Contact - (IFHT), RWTH Aachen

2010

Annual Report

A NNUAL R EPORT 2010

1

Institute for High Voltage Technology

Institute for High Voltage Technology

Institute for High Voltage Technology Schinkelstraße 2 DE-52056 Aachen Phone: Fax: e-mail: Internet:

+49 241 80-94931 +49 241 80-92135 [email protected] www.ifht.rwth-aachen.de

Editor:

Univ.-Prof. Dr.-Ing. Armin Schnettler Institute for High Voltage Technology RWTH Aachen University

Editorial: Editorial deadline:

Dipl.-Ing. A. Strauchs, Dipl.-Gwl. M. Keller 31.12.2010

2

A NNUAL R EPORT 2010

Preface Research at IFHT Fields of Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Scientific Reports of Research Group Circuit Breakers . . . . . . . . . . . . . . . . . . . . . . . . 12 Scientific Reports of Research Group Insulation Systems and Diagnosis . . . . . . . . 19 Scientific Reports of Research Group Sustainable Energy Systems . . . . . . . . . . . . . 32 Scientific Reports of Research Group Systems Engin. and Asset Management . . 45 Scientific Reports of Research Group Fault Arcs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 PhD Theses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 HumTec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Institute for High Voltage Technology

Contents

Teaching at IFHT Lectures and Laboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Diploma- , Master-, Bachelor- and Project Theses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Seminars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

The Institute Chronicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Technical excursions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Service Portfolio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Test Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EEIM Award 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

112 116 121 129 130 134 138

A NNUAL R EPORT 2010

3

Institute for High Voltage Technology

4

Dear friends and partners of the Institute, dear Ladies and Gentlemen,

recently electrical power engineering received an amount of attention from the media we dreamed about 20 years ago. However, this public attention leads to rushed political actions and a complicated discourse about reasons, aims and solutions towards alternative energy sources. Of course we have to transform the existing system according to the future needs. However, due to the high importance of reliable and secured energy for the economy and security for a country, this task must be prepared thoroughly and rash experiments and interference by political interests have to be avoided. In addition to the well-known aspects of energy supply, such as security, reliability, sustainability and price, the aspect of public acceptance gains more and more importance. This is especially true for nonrenewable energies and for the necessary measures for the reinforcement of the grid which require a lot of persuasion of authorities and citizens before they can be commissioned - if at all. These developments prove that our holistic approach toward supplying electricity from concepts and scenario analyses over technologies and systems to an economical, ecological and ethical assessment of options - is a good one. We are convinced that through our research topics and expertise we are in a good position to contribute to the future design and development of power supply. This annual report 2010, for the first time in a separate German and English edition is supposed to give you a short but concise overview about our topics, research and results. To be able to maintain the format and also accommodate the increased scope of our topics and thus the increased number of papers it became necessary to split the an-

A NNUAL R EPORT 2010

nual report into a German and an English edition. My 10th year as head of the institute was dominated by a growth of both personnel and research topics: • The money from third party funds (i.e. non basic funds) spent in 2010 amounted to more than 3 million Euros with rising tendency. • There were more than 250 oral exams in our core lectures and another 200 written exams in the Bachelor study program. • More than 60 theses (Bachelor, Master and Diploma) were submitted at the IFHT. • 17 Research associates were integrated to our staff (some temporarily). • 8 members of our staff received renowned awards. • 5 dissertations were successfully completed in 2010. • A large number of new and interesting projects started or are in development. We have to admit that these developments in 2010, however positive they are, have shown the limits of what is possible within the environment of a university. The fact that members of our staff received awards despite their high workload makes us proud and proves that growth does not necessarily compromises the quality of work - a very important aspect for universities which are usually not acquainted with strong growth. The PhD degrees awarded to Dr. Kittipong Anantanavich, Dr. Ming Chark Tang, Dr. Stefan Federlein, Dr. Michael Keßler and Dr. Torsten Wirz made 2010

Preface the implementation of the Testing center for storage systems and Smart Grid technologiesänd social life at the Institute can be found in this annual report. I want to thank all friends and partners of the institute for their support and suggestions and our staff for their commitment and creativity. We consider the success of the years past as an acknowledgement of our work and as an incentive for further improvement. Together with you we want to find the solutions for the future of our energy systems. Best regards from Aachen Yours

Institute for High Voltage Technology

also a successful year concerning our research. For the years to come we also expect an annual average of five to six PhDs. The works of Mrs. Anja Strauchs and Mr. Andrey Mashkin in the field of syntactic foam were acknowledged by the IEEE DEIS Graduate Award and the EEIM Award, respectively. For their theses Mr. Daniel Eichhoff received the ABB-Diploma Award, Philipp Masmeier the Otto-JunkerAward and Stanislav Votruba was granted the SAG Award, 2nd grade. Especially pleasant is the victory of our student project „Plug into Future“ within the competition „Energy for Ideas“ of the Federal ministry for education and research (BMBF) which was done by Bartholomäus Wasowicz and Fabian Potratz in close cooperation with Thomas Dederichs. Information regarding teaching, research,

Aachen, January 2010

5

A NNUAL R EPORT 2010

6

Institute for High Voltage Technology

A NNUAL R EPORT 2010

Research at IFHT

A NNUAL R EPORT 2010

7

Institute for High Voltage Technology

Institute for High Voltage Technology

8

Fields of Research Research Group “Circuit Breakers” Manager: Dipl.-Ing. Andreas Kurz Short description of the research topics: Circuit breakers are one of the most important protective devices in electrical power systems. Therefore, it is of huge importance to understand the physical process of a switching operation. Thus, the research group follows two different approaches. On the one hand so called CFD-(Computational-Fluid-Dynamics)-Simulations are used to model the physical phenomena. Due to this information parameters can be figured out, which are in general difficult to observe e. g. the temperature. On the other hand the required equipment and know-how to investigate the phenomena experimentally is available in the research group. While developing new switching technologies measurements of the temperature, the pressure, the gas flow as well as a determination of the interruption limit are part of the basic investigation methods. Due to an intensive interaction of both approches some results could be transferred into new inventions. Members of the group: Dipl.-Ing. Mathias Behle (until Mai) ∗ Dipl.-Ing. Daniel Eichhoff ∗ Dipl.-Ing. Matthias Hoffacker ∗ Dipl.-Ing. Andreas Kurz ∗ Dipl.-Ing. Gregor Nikolic ∗ Dipl.-Ing. Ming-Chark Tang (until April) Student assistants: Stefan Buß ∗ Polson Keeratibumrungpong ∗ Tim Kubiczek ∗ Andreas Meyer ∗ Thien-duc Nguyen ∗ Gregor Nikolic ∗ Karl Oberdieck ∗ Michael Weuffel ∗ Tobias Wild

Research Group “Insulation Systems and Diagnosis” Manager: Dr.-Ing. Michael Keßler Short description of the research topics: The research group covers a wide variety of technical challenges: ranging from the development of new insulating materials, up to a condition assessment of insulating systems by means of different diagnostic methods. In the field of hybrid insulation systems, syntactic foam as a new insulating material with low specific density is in the main focus. Especially a possible correlation of the dielectric strength and the adhesion between microspheres as filler and the matrix is investigated. In the field of diagnosis, the assessment of polymeric insulation systems by using ultra sound is another focus. Besides the detection of small impurities in different materials like elastomers (e.g. in cable joints), boundary layers between two different materials and inhomogeneous cross-linking are detectable. A new application of ultra sound diagnosis is the temperature measurement at inner interfaces. An efficiency advancement of ignition systems in combustion engines via „Multisparc Ignition“ is a new research field of this group. Instead of one single sparc, many short discharges are ignited in series to assure a steady ignition of the fuel-air mixture.

A NNUAL R EPORT 2010

Fields of Research

Research Group “Sustainable Energy Systems” Manager: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Short description of the research topics: The research group is focused on modelling and assessing integrated energy systems. These systems are characterized by merging the markets for different forms of energy due to technological developments (e.g. Combined Heat & Power, heat pumps or electric vehicles). The expertise of the group lies especially in generation, transmission and distribution technologies. The methods implemented in the process of modelling include classic scenario analysis, load flow and stability assessments as well as economical analyses (for example potential or risk assessment). This is complemented by extensive knowledge in the field of Life Cycle Assessment according to ISO 14040/44. Sustainability is perceived as a multi-crtiteria optimization problem, which must include technical, economical, and ecological assessment criteria. Much effort is directed into supporting the further development of a methodology for technical eco-efficieny analysis. Due to the close cooperation with the interdicipinary project house “Humtec“ in the field of „Ethics for Energy Technologies“ the research is rounded off by taking ethical problems into consideration. Members of the group: Dipl.-Ing. Dipl.-Wirt.-Ing. Thomas Dederichs ∗ Dipl.-Ing. Dipl.-Wirt.-Ing. Markus Gödde ∗ Dipl.-Ing. Thomas Helmschrott ∗ Dipl.-Ing. Stefan Krengel ∗ Dipl.-Ing. Claas Matrose ∗ Dipl.-Ing. Hendrik Natemeyer ∗ Dipl.-Wirt.-Ing. Baris Özalay ∗ M. Sc. Thomas Pollok ∗ Dipl.-Ing. Martin Scheufen ∗ Dipl.-Wirt.-Ing. Eva Szczechowicz ∗ Dipl.-Wirt.-Ing. Bartholomäus Wasowicz ∗ Dipl.-Ing. Sebastian Winter Student assistants: Mitra Ariatabar ∗ Reinhold Bertram ∗ Christian Bredtmann ∗ Philipp Erlinghagen ∗ Pascal Golor ∗ Philipp Grüneberg ∗ Andreas Haidl ∗ Mirco Hinnenkamp ∗ Julie Kramer ∗ Sören Kreimeier ∗ Shanta Kumar ∗ Julian Langstädtler ∗ Moritz Mittelstaedt ∗ Mark Müller Giebeler ∗ Thomas Nemeth ∗ Florian Nießen ∗ Tim Noffz ∗ Jens Noschinski ∗ Cora Sophie Petino ∗ Fabian Potratz ∗ Stephan Raths ∗ Erik Reimann ∗ Annika Reitz ∗ Andreas Roehder ∗ Yvonne Surmann ∗ Julia Tecklenburg ∗ Philipp Tünnerhoff ∗ Johannes Turinsky ∗ Markus vor dem Berge

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Members of the group: Dipl.-Ing. Gregor Brammer ∗ Dr.-Ing. Michael Keßler ∗ Dipl.-Ing. Jens Knauel ∗ M. Sc. Andrey Mashkin ∗ Dipl.-Ing. Benjamin Preidecker (Daimler AG) ∗ Dipl.-Ing. Christoph Roggendorf ∗ Dipl.-Ing. Anja Strauchs ∗ Dipl.-Ing. Daniel Winkel Student assistants: Denys Demenko ∗ Paul Duisberg ∗ Henning Frechen ∗ Pierre Honoré ∗ Sebastian Kamps ∗ Patrick Poss ∗ Alexa Rombach ∗ Christoph Schneider ∗ Stefan Seibel ∗ Daniel Vobis ∗ André Wagner ∗ Matthias Wiesmann ∗ Serdar Yildiz

9

Institute for High Voltage Technology

10

Research Group “Systems Engineering and Asset Management” Manager: Dipl.-Ing. Christian Hille Short description of the research topics: A safe, reliable and efficient operation of electrical grids is the main focus of the research group Systems Engineering and Asset Management. This task requires a fundamental and detailed knowledge of the behaviour of each relevant grid component, in order to be able to model the complex system’s behaviour. Additionally, the system point of view enables identification and description of emergent properties. Current and future performance, reliability and lifetime of installed equipment are assessed in close cooperation with grid operators and manufacturers. The main focus is the development of asset-simulation software and expert-systems to support grid operators in their decisions towards an efficient network operation. Protection systems of actual and future distribution networks are presently another focus. A massive penetration with decentralized energy resources and accumulators falsifies classical assumptions concerning the distribution systems’ operating conditions. Protection systems are analyzed and designed from an integral point of view, looking at the single components’ and the system’s properties and behaviour at once. Thereby a secure operation of today’s distribution networks and those of the future shall be guaranteed, also taking into consideration the prospective long term coexistence of future’s active and classic passive distribution systems. A technological and economic comparison of different options shall enable macroeconomic optimal recommendations. Members of the group: M. Sc. Tirinya Cheumchit ∗ Dipl.-Ing. Stylianos Filippidis ∗ Dipl.-Ing. Christian Hille ∗ Dipl.-Ing. Tilman Wippenbeck Student assistants: Tim Gaßmann ∗ Karina Klusmann ∗ Pascal Koehn ∗ Simon Sandor

Field of Instruction and Research “Gas Discharge Engineering” Univ.-Prof. Dr. rer. nat. Gerhard Pietsch Short description of the research topics: Fault arcs in electrical installations cause temperature and pressure rise in switchgears and switchgear buildings. In principle, fault arc safety can be proved experimentally by fault arc tests and theoretically by pressure rise calculations. The advantage of pressure rise calculation is that it can be performed with applications, for which no experimental possibilities are available, e.g. when the influence of pressure rise on buildings is investigated during the design of a substation. The existing pressure calculation methods will be extended particularly for SF6 -insulated installations. In this case, changing SF6 -air mixtures in the relief room have to be taken into account. The extended calculation method can be evaluated by comparing calculated with experimental results. A further problem is related to the replacement of SF6 by air for tests of SF6 -insulated switchgear due to environmental concern. This

A NNUAL R EPORT 2010

Fields of Research

Members of the group: M. Sc. Kittipong Anantavanich

Institute for High Voltage Technology

task is to find out, whether there exist possibilities to replace tests with SF6 by those with air.

11

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Development of a measuring system investigating the spatial resistance distribution of switching arcs For many years the development of gas circuit breakers is surrounded by computational fluid dynamics simulations. They can help to unterstand and thus to improve the switching process of the breaker. The used physical models of these simulations are often crosschecked by experimental investigations. The spatial arc resistance distribution of the arc is one of the fundamental aspects which are related to the interruption capability of the circuit breaker. However it could not yet be determined by experiments as all known methods have either a influence on the arc and the gas flow or can only applied at special conditions. The aim of the ongoing research project is the development of a new method to determine the spatial arc resistance distribution of a quenched switching arc. This can lead to a better physical understanding of the cooling process and thus to improved simuilation models.

Introduction

12

Circuit breakers are important for the electric energy supply. They are responsible for disconnecting the current at nominal conditions and in case of a fault. The current is interrupted during its natural current zero crossing by cooling the electric arc. This cooling is established by blowing the arc with the filling gas of the breaker. For a SF6 circuit breaker the arc resistance at a time instant 200 ns before current zero (R200) can be used as an indication to predict the success of the interruption capability [1]. Different cooling mechanisms along the axis of the arc cause a non-linear resistance distribution in the arc. The measurement of this distribution would render possible a conclusion on these mechanisms and their influence on the interruption capability [2]. Thus a non-invasive method to determine this spatial resistance distribution of a blown switching arc is the objective of this project.

Approach For the measurement of the spatial arc resistance distribution the potential drop along the axis of the arc is determined. The arc

A NNUAL R EPORT 2010

resistance distribution is then calculated by knowing the current. The potential distribution will be determined by measuring the electric field surrounding the arc. Therefore a reference potential is required which is created by mounting a conductive cylinder around the insulation nozzle. The cylinder is then connected to ground. Measuring electrodes are placed between this cylinder and the nozzle (Fig. 1) to measure the arc potential. measuring electrode PP PP PP q P

C1

Ce

cladding electrode Ci

Cn

Ri

Fig. 1: Measuring principle (schematic)

The material of the nozzle acts as a dielectric and separates the electrode from the arc. This construction is a capacitive field probe. Between electrode and grounded cladding tube a capacity Ce is generated. Furthermore this yields to different stray capacitances Ci between arc and measuring electrode. The sum of all these capacitances C1 ...Cn

Field of Research: Circuit Breakers

ϕ1

C1

I1 ..

ϕi

Ci

Influence of radius and position of the arc Using a capacitive coupling, correct dimensions of the field probe and a well chosen load of the sensor the electric field strength at the wall of the conductive cladding tube can be determined [3]. Assuming a short part of the arc with the length ∆x is an ideal cylindric conducter with the radius rarc the measured field strength can be calulated analogue to a cylindrical capacitance:

Ii ..

ϕn

Cn

In Ce

ln Rm

Cm

Um

Fig. 2: Model of capacitive coupling

In Fig. 2 the mathematical model of the signal coupling of a sensor loaded with Rm and Cm is given. With given arc potentials ϕi the measured voltage at the sensor electrode is calculated by:

Um =

Cm +

Pn

Pi=1 n

Ci ϕi

i=1 Ci

−

ϕ

E0 ∝

Im

j Rm ω

By knowing the signal of multiple sensors the arc resistance distribution can be calculated. Therefore the potential of the arc in front of the sensor should have the highest influence on the measured voltage. Thus the coupling capacitance Ci which is lineal from sensor to arc should have a high value while the capacitances beside should have low values. This can be established by design features. To achieve a linear frequency response of the transfer function of the system the load resistance Rm has to be as high as possible with respect to the band width of the signal to be measured.



r0 rarc



Institute for High Voltage Technology

is the coupling capacitance between sensor and arc. The setup corresponds to a capacitive divider with Ci as the upper capacitance of the divider and Ce as lower capacitance of the divider. For measuring the potential distribution several sensors are located along the axis of the arc. Each sensor is measuring a partial voltage of the arc.

∗ r0

Here ϕ is the potential of the part of the arc and r0 is the radius of the cladding tube. Obviously neither ϕ nor rarc can be determined by knowing r0 and E0 . The assumption of a constant value for rarc is not valid as the arc radius will vary due to its dependence on the quenching and on the current. By means of a second field probe - which is assembled at the same part of the arc with a radius r1 unlike r0 the electric field strength E1 can be calculated analogue. Thus both unknown variables (ϕ und rarc ) can be determined. However the assumed symmetry along the rotational axis will not occur in the experiment. The arc will move inside the insulation nozzle. Hence r0 and r1 as well are unknown. To be able to determine position, thickness and potential of the arc anyway the amount of field probes is increased. By placing four sensors surrounding the same short part of the arc the system of equations can be solved.

Results In a first approach four capacitive coupled sensors are placed around an arc burning inside a insulation nozzle. The design is shown in fig. 3. Three sensors are identical and separated by an angle of 120◦ from each

A NNUAL R EPORT 2010

13

Institute for High Voltage Technology

other (S1-S3). The forth sensor (S4) is shifted outwards.

S1 S2

S3

Acknowledgement

S4

This Project is funded by Deutsche Forschungsgemeinschaft (DFG). Fig. 3: Sensor arrangement

The feasibility of the above explained measurement method can be shown by experiments. The measured voltages at sensors S3 and S4 for a non-quenched arc - which is powered by a 950 Hz resonance circuit - are shown in fig. 4. 0.06

S3 S4 +8,6dB

0.04 0.02 sensorvoltage [V]

0

14

ple groups - each consisting of four sensors - along the axis of the arc. For analysing the experimentally measured data a software is developed to calculate the spatial resistance distribution. Therefore the mathematic theory of inverse problems is used.

-0.02 -0.04 -0.06 -0.08 -0.1 -0.12 -0.14 19

19.5

20

20.5

21 21.5 Zeit [ms]

22

22.5

23

Fig. 4: Measured sensorsignal

Here the signal amplitude of sensor S4 is amplified by 8.6 dB. This allows a qualitative comparison between both voltages. Differences between both voltage characteristics can be explained by a varying arcradius respectively by a changing arc position.

Outlook As the principle feasibility of the method can be shown the next step is placing multi-

A NNUAL R EPORT 2010

References [1] Knobloch, H., Habedank, U.: Behaviour of SF6 high-voltage circuit breakers with different arcextinguishing systems at shortline fault switching; In: Science, Measurement and Technology; IEE Proceedings; Bd. 148 273−279; 2001 [2] Tang, M.C.: Widerstandsverteilung in Schaltlichtbögen von Selbstblasleistungsschaltern während der Stromnulldurchgangsphase; Dissertation, Institut für Hochspannungstechnik; RWTH Aachen; 2010 [3] Andreas Küchler: Erfassung transienter elektromagnetischer Feldverteilungen mit konzentrierten und räumlich ausgedehnten Sensoren; Fortschritt-Berichte VDI; Reihe 21: 7; Düsseldorf; 1986

Contact Dipl.-Ing. Matthias Hoffacker [email protected] +49 241 80-94944

Field of Research: Circuit Breakers

In future electrical power systems based on DC-technology (direct current) appropriate switchgears are essential for switching operations during normal conditions as well as in case of fault conditions. AC-circuit breakers, widely used in present grids, are unable to perform these operations due to their basic physical process of ’current zero switching’. Hence, the development of special DC-circuit breakers is necessary. For this purpose, firstly, the technical requirements must be determined by means of simulations. It is observed, that fault currents in a DC-grid rises with a higher steepness compared to AC-systems with similar short circuit power. Concerning the interrupting capability, this means that DCcircuit breakers generally have to fulfill enhanced requirements.

Introduction In the future, the DC-technology will be increasingly used in electric power systems. At present, the HVDC-technology is already used for the energy transmission over long distances and for the connection between offshore windfarms and the onshoregrid. In many cases of application, HVDC is the only reasonable alternative [1]. In electric distribution grids, the share of converter-connected, decentralized power generation units, especially photovoltaic units and wind energy converters, will further increase as well as the ratio of converter-fed loads in the grid will rise. Hence, the DC-technology can be advantageous for medium-voltage distribution grids regarding technical and economical issues [2]. The realization of the internal medium-voltage grid in an offshore-windfarm is assessed as a possible application for a medium-voltage DCgrid (MVDC), which interconnects the dcbusses of the wind energy converter (WEC) units with the central HVDC-platform [3]. In this way, the convertion to the 50 Hz AC-voltage at the WEC units, the transformation of the 50 Hz current and the rectification for the HVDC-connection can be avoided. These MVDC-grids require appropriate

switchgears in form of DC-circuit breakers for a reliable and save operation. These circuit breakers must be able to perform the switching of nominal currents under normal operation and they have to assure the save interruption of short circuit currents in case of a fault. At present, appropriate circuit breakers are not available for MVDC-applications. Furthermore, ACcircuit breakers are not suited for the interruption of direct currents, as they essentially require a zero-crossing of the current, which does not occur in the DC-case. For a targeted development of the needed DC-circuit breakers, firstly, the technical requirements for these switchgear, which arise in future DC-grids, have to be identified. Technical standards, providing applicable figures and values, have not yet been defined.

Model of a MVDC-Grid The technical requirements for DC-circuit breakers can be derived from theoretical investigations in a model DC-grid. The bipolar MVDC-grid in an offshorewindfarm, which is used for the simulations, is given in Fig. 1. The nominal voltage of the exemplary system is set to +/- 7,5 kV. For reasons of clarity, only one single pole of the system and one branch

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Technical requirements for circuit breakers in future DC grids

15

Institute for High Voltage Technology

16

of the grid are depicted. The grid links the wind energy concerter units (WEC), having a nominal power of 5 MVA each, with the central DC/DC-converter platform, from where the electric power is transmitted to the onshore grid by a HVDC-link.

In normal operation the mechanical switch carries the current due to its comparatively low losses. To interrupt the current, the mechanical switching contacts are opened and simultaneously, the anti-parallel thyristors are fired. Hence the current commutates into the parallel path and charges the capacitor. If the charging voltage of the capacitor exceeds the grid voltage, a negative di/dt is applied over the grid inductances (maily the inductances of the cables in the grid). This leads to a decrease and finally to an interruption of the current. The varistor limits the maximum charging voltage to its residual voltage.

Fig. 1: Model of the DC-Grid for a OffshoreWindpark

One possible fault, considered in the model of the DC-grid, is given by an earth fault of the cable L3. In this case, the circuit breakers CB1 and CB2 as the responsible protective devices have to interrupt the fault current. Technically, different concepts exist, how a DC-circuit breaker can be realized. Whereas special requirements in addition to general ones can arise for every breaker topology. In the scope of the presented simulations a Snubbered Mechanical Circuit Breaker (SMCB) is investigated. The SMCB is a so-called hybird circuit breaker, that combines a mechanical switch with powerelectronic devices and passive components, as shown in Fig. 2. The main features of the SMCB are its simple control and the utilization of cost-effective thyristors instead of active-turn-off devices like IGBTs [4].

A NNUAL R EPORT 2010

Fig. 2: Setup of a Snubbered Mechanical Circuit Breakers (SMCB)

Exemplary results The presented exemplary results are based on the application of a SMCB-model, used to interrupt an ideal earth fault in the model DC-grid. Before the fault occurs at t = 0 s, the system operates with a nominal current of 1 kA. The simulated development of the current through the breaker, the voltage over the breaker, and the grid voltage during the switching process are given in Fig. 3˙Immediately after the occurance of the fault, the current increases with a rate of rise up to 75.4 A/µs and crosses the detection level of 2 kA after 18.5 µs. The internal delay of the breaker after the detection of the fault is assumed to be an additional 100 µs.

Field of Research: Circuit Breakers

Current mech. switch

8 Current I [kA]

Current parallel path

6

Current Varistor

4 2 0 0

100

200 300 Time t [µs]

400

500

12

Voltage U [kV]

10

Voltage over circuit breaker

8

Wmag =

6 4 Grid voltage 2 0 0

100

200 300 Time t [µs]

400

500

Fig. 3: Current and voltage during the clearing of the earth fault

After the separation of the mechanical contacts, the current commutates into the parallel path and reaches a maximum value of 8.8 kA after t = 140 µs, when the charging voltage of the capacitor in the breaker UC exceeds the voltage of the grid. The voltage UC further increases and is limited by the varistor after 170 µs to a maximum value of 11 kV. In this phase the current commutates into the varistor path. With the preset parameters for the grid and the SMCB the entire interruption process is completed after 270 µs. The system voltage, which dips down to 2.0 kV during the fault, recovers to its nominal value of 7.5 kV after the successful disconnection of the fault affected section of the grid. The increasing voltage over the capacitor due to the charging by the commutated current leads to a dielectric stress of the contact gap in the mechanical switch immediately after the contact separation. In order

Institute for High Voltage Technology

10

to avoid a dielectric breakdown and hence a failure of the entire breaker, the stress of the contact gap must be controlled by a sufficient dielectric strength of the gap. Because the mechanical contact are still in the opening process and have not reached the maximum contact travel, a voltage rise of less than 1 kV/µs can already cause a dielectric stress, which may lead to a breakdown [5]. Hence. the rate of rise of the voltage over the breaker has to be considered in the dimensioning process of the capacity. The energy stored in the grid in form of magnetic energy 1 2 · Lgrid · Imax 2

is compensated by the breaker during the switching process, This is done, firstly, by charging the capacitor and secondly, by the dissipation of ohmic losses, especially in the varistor. The maximum amout of energy, which has to be compensated, must be taken under consideration during the development of a DC circuit breaker as an additional parameter.

Conclusion and Outlook The behaviour of a DC-grid under fault conditions and the subsequent electric stresses for circuit breakers during the interruption of fault currents deviates inherently from the situation in present AC-grids. The DCcircuit breakers have to fulfill increased technical requirements concerning their interruption capability, which arise from the lack of natural current zero crossings and the high steepness of the DC-fault current. The dielectric stress of the breaker by a transient recovery voltage (TRV) after the current interruption, as in high voltage ACsystems, does not occur in the DC-case. In fact, the DC-circuit breaker is dielectrically stressed by the voltage that is internally

A NNUAL R EPORT 2010

17

Institute for High Voltage Technology

18

build up during the switching process to achieve a negative di/dt over the grid inductance. Additionally, the DC circuit breaker must be able to compensate the magnetically stored energy from the grid. The investigations show that DC-grids require high performing protection technologies and secondary installations, especially concerning the required speed of the fault detection, the signal processing and the signal transmission. The value of 100 µs, which is assumed for the entire time delay in the scope of the current investigations, is less than the operational time present technology can provide. The technical requirements derived from the current investigations will be used for the development of a DC-circuit breaker for MVDC applications. For this purpose, the SMCB will be compared with several concepts for DC-circuit breakers, that use vacuum switching technology in order to identify the most qualified topology, which will be used as a basis for the design of a MVDC cricuit breaker prototype. With regard to the vacuum circuit breaker, the effect of current chopping on the dielectric stress of the DC-grid and the DC-circuit breaker will be analyzed. Additionally, the required equipment for development and performance tests of DC circuit breakers will be setup according to the simulative results from these investigations.

References [1] Kirby, N. et al: HVDC transmission for large offshore wind farms, IEE

A NNUAL R EPORT 2010

Power Engineering Journal, Vol 16, pp.153-141,2002 [2] Larruskain, D. et al: Transmission and Distribution Networks: AC versus DC, 9th Spanish-Portugese Congress on Electrical engineering, Marbella, Spain, 2005 [3] Martander, O.: Wind Farm Configuration and Energy Efficiency Studies - Series DC versus AC Layouts, Chalmers University Göteborg, Ph.D. Thesis, 2006 [4] Meyer, C.: Key Components for Future Offshore DC Grids, RWTH Aachen University, Aachener Beiträge des ISEA Band 46, Ph.D. Thesis, 2007 [5] Holaus, W.: Ultra fast switchesBasic elements for future medium voltage switchgear, Swiss Federal Institute of Technology Zurich (ETH), Ph.D. Thesis, 2001

Contact Dipl.-Ing. Daniel Eichhoff [email protected] +49 241 80-97348 Dipl.-Ing. Andreas Kurz [email protected] +49 241 80-90270

Field of Research: Insulation Systems & Diagnosis

Many parameters of the cable production process are based on long experiences and are optimized for every special cable type. During the combined extrusion process of the cable sheath, there is a risk that gas inclusions can occur between the polyethylene sheath and aluminium screen caused by missing glue. Due to those inclusions water can permeate into this delaminated areas and can cause corrosion of the aluminium screen, which leads to reduced life time of the cable. With ultrasound diagnosis it is possible to identify and localize these gas inclusions at interfaces.

Delaminations in power cables

Theory of detection

In the last production step of power cables for high voltage (HV) and extra high voltage (EHV) an aluminium screen is bent around the cable core and both ends are welded together.

A sound impulse propagating from the transducer (TD) through the sample is reflected at innner interfaces and at the backwall and propagates back to the transducer which receives the impulse (Fig. 1).

The screen shall carry the short circuit current in a fault situation and prevent water intrusion into the cable, which could lead to chemical degradation of the insulation material. After welding, the screen is led into the extruder head. Inside the head the hot screen is covered with a thin glue layer and a polyethylene (PE) sheath is extruded on top. There is a risk for delaminations during the sheathing process of the cable. Water can permeate into these delaminations and cause corrosion of the screen. Due to the by corrosion reduced cross sectional area the ampacity might be influenced negatively. It is possible to perform a non-destructive online test of the interface between the aluminium screen and the polyethylene sheath directly after the extrusion process for detection and localization of potential defects. The positive aspect about this monitoring is the knowledge of the defect’s position. Moreover it gives the chance to optimize the production parameters during the running extrusion process of the sheathing line. Hence rejections can be minimized.

Institute for High Voltage Technology

Prototype for detection of delamination in power cables using ultrasonic diagnosis

19

Fig. 1: Top: Sound propagation and reflection at inner interfaces Bottom: Time signal of the reflected impulses

A reflection of waves occurs generally at interfaces between two materials with a different specific sound impedance Z which is defined as product of density ρ and sound velocity c of the material: Z =ρ·c The sound waves behave at material boundaries corresponding to the mathematical model of electrical waves. The reflection

A NNUAL R EPORT 2010

Institute for High Voltage Technology

and transmission coefficients are given by [1]: Z2 − Z1 R= Z2 + Z1

2 Z2 T = Z2 + Z1

where Z1,2 are the specific sound impedances of the materials. In order to detect delaminated areas of the cable sheath the time signal of the ultrasound reflection is analysed. In figure 2 the intact reflection signal (black) is compared to the signal of an air inclusion (gray). The difference between these signals can be found by correlating the measured time signal with a reference signal. 1000

amplitude / a.u.

500

0

−500

20

−1000 100

200

300 400 time / a.u.

500

600

Experimental Setup In a first step an applicative test bench with a single transducer for testing of the interface between aluminium screen and polyethylene sheath is build up. This setup is used for verification of the theory of detection of gas inclusions at the interface for this special kind of cable sheath. Figure 3 shows the test bench with a sheath sample mounted. The setup uses a single transducer for scanning. The ultrasonic transducer is mounted on a special construction which provides a water reservoir for coupling the ultrasound into the sample. For the movement of the sample and the transducer a micro positioning system is used. While moving the transducer continuously along the sample the ultrasonic signal is recorded. At the end of every line the sample is rotated. A combination of the ultrasonic measurement data and the measurement position gives the possibility for location of defects at the interface between aluminium screen and the polyethylene sheath within the scanned sample.

Fig. 2: Time signal of synthetic testsample, black: intact, gray: delamination

Using signal correlation the differences in ultrasonic signals can be evaluated automatically after or even during measurement. The similarity of two signals s(t) and g(t) is described by the normalized cross correlation function [2]: psg =

R +∞ −∞

s∗ (t) g(t − τ )dt p Es Eg

To make the result independent from the particular energy of the signals s(t) and g(t) it is normalized, where Es and Eg are the energies of the signals. As a result psg returns values in the range of -1 to +1 where +1 is the largest similarity.

A NNUAL R EPORT 2010

Fig. 3: Experimental setup for automated scanning

The measurement data collected during the scanning process are stored for investigations on the acoustic properties of the sheath material, post processing and analysis as well as the development of a detection algorithm. Figure 4 shows a K-scan [3], a 2d-plot of the correlation coefficients depending on the position. Black areas rep-

Field of Research: Insulation Systems & Diagnosis

180 0.8 160

0.6

φ direction / a.u.

140

0.4

120

0.2

100

0

80

−0.2

60

−0.4

40

−0.6

20

−0.8 20

40 60 80 x direction / a.u.

100

120

It is possible to correlate the measured data to a certain position of the cable which enables the possibility for finding the delaminated areas. The online evaluation and data analysis allows a direct user interaction e. g. display of a warning if delaminations are detected. Using this information given by the ultrasound measurement system it is possible to mark the delaminated areas and to correct production parameters, in order to guarantee a constant quality of the power cable sheath and a minimization of rejections.

−1

Fig. 4: Scan of synthetic testsample, black: intact, white: delamination

Industrial Prototype The developed ultrasound measurement prototype is an adaptive system, which can be used in the production line without changing the existing production process (Fig. 5).

Acknowledgement The authors thank NEXANS Charleroi/Calais for the good cooperation and the preparation of special sheath samples.

References [1] Werkstoffprüfung mit Ultraschall, J. Krautkrämer, H. Krautkrämer, Springer 1986 [2] Signalübertragung, J. R. Ohm, H. D. Lüke, Springer 2004 [3] Detektion und Evaluation von Inhomogenitäten im Volumen und an inneren Grenzflächen in polymeren Isoliersystemen mittels Ultraschall, P. Walter, Dissertation IFHT, Aachen 2006

Fig. 5: Multisensoral transducer arrangement

Moreover the measurement technique is contactless so that there is no need for a physical contact between the system itself and the extruded cable sheath. This ensures a non-destructive examination of the extruded PE sheath. In order to cover the surface of the cable several ultrasound sensors are placed around the cable sheath.

Institute for High Voltage Technology

resent an intact interface between the aluminum screen and the PE sheath. Delaminations show less similarity to the reference signal and are therefore coloured in white.

Contact Dipl.-Ing. Gregor Brammer [email protected] +49 241 80-94910 Dipl.-Ing. Jens Knauel [email protected] +49 241 80-97346

A NNUAL R EPORT 2010

21

Institute for High Voltage Technology

Design and Construction of a Test Bench to Characterize the Behavior of Batteries in High Voltage Storage Systems Battery energy storage systems could be able to compensate volatile energy sources, such as wind and solar energy. One system topology for medium voltage grids uses high voltage batteries. This concept should provide an improvement of the overall system efficiency. High voltage batteries may be advantageous for future DC-grids as well. However, as there is little expertise about series-connected batteries up to several kilovolt, the charging and discharging behavior of such systems will be investigated. For this purpose, a test bench for high voltage storage systems is built to analyze these processes for different battery technologies.

Introduction

22

Strongly increasing power generation based on volatile energy sources, such as wind and solar energy, sets new requirements on design, planning and operation of power systems. In addition, the demand for regulating reserve power and ancillary services is drastically increasing, forcing power system operators to intensify the use of system flexibility measures and non conventional operating concepts. Here, battery energy storage systems or electrochemical energy storage systems in general may offer unique and scalable solutions for high power and long term energy demands in the range of up to 100 MW and several 100 MWh. To reach these power levels one option is to use high voltage storage systems. Large battery energy storage systems show high reliability and robustness. Thus, the usage in high voltage applications is feasible in general [1]. However, the series connection of batteries up to several kilovolt leads to nontrivial charging and discharging characteristics of such a storage system. To investigate the cell balancing behavior for different battery technologies, a high voltage battery test bench is under construction. The concept of the test bench is shown in Fig. 1.

A NNUAL R EPORT 2010

Fig. 1: Principle drawing of the test bench

The test bench consists of an insulation transformer with connection to the low voltage grid, an inverter, a DC/DC converter as charge/discharge unit and the high voltage storage system. The power capacity of the power electronics and the grid connection is in the range of 100 kVA. In the basic configuration the storage system consists of 360 lead-acid batteries (12 V / block) connected in series.

Power electronics The classical two-level hard-switched converter is state-of-the-art for low voltage applications, being the most commonly used topology in industry. For grid connection of such converter it is necessary to have a higher dc-link voltage than it is usually used for drives, in order to be able to inject power into the grid. The converter is connected to the 50 Hz low voltage grid. The converter’s switching frequency is 3 kHz. In order to comply with the grid codes, an LCL output

Field of Research: Insulation Systems & Diagnosis

High voltage storage system The battery storage system consists of a large number of cells in series, thus the optimum topology and reliability are major issues. Lead-acid batteries are robust and competitively priced in comparison to other battery technologies. The test bench’s battery is constructed of 360 12 V lead-acid blocks, with a capacity of 27 Ah each. The test bench’s battery system can store approximately 120 kWh of electrical energy. The topology of the test bench mainly consists of six battery building blocks with 720 V each. Thus, the nominal string voltage is ± 2160 V. In the case of fully charged batteries including the voltage drops over the cables, fuses and disconnectors the storage system voltage is about ± 3 kV. Figure 2 shows the battery storage system. Each battery building block consists of ten substrings of 72 V. The batteries in the test bench are stored in two separate battery racks, each contain-

ing three of the 720 V building blocks. This design was chosen to reduce connection length in order to decrease the internal resistance of the storage system.

Institute for High Voltage Technology

filter is used together with a current control which is able to compensate low order harmonics. The DC/DC converter, a dual-active bridge topology, acts as a charging/discharging unit for the high voltage battery. In the test bench, there is a wide range of operation conditions. Depending on the charging status of the batteries, the voltage varies between 3.3 kV and 6 kV. The charging/discharging power varies between 4 kW and 100 kW. Low power operation is needed to investigate the voltage distribution of the series connected battery cells. The converter is optimized for the whole operation range. The output voltage is controlled by the grid coupling inverter to be constantly 750 V. Hence, a high voltage transformation ratio from output to input is necessary. Therefore, a galvanically isolated topology was chosen to benefit from the winding ratio of the included transformer.

Fig. 2: Battery storage system

The battery monitoring system for the test bench consists of 360 LEM Sentinel sensors. Herewith the voltage, temperature and impedance of each battery block are measured independently.

Safety infrastructure Due to the high voltage and the high energy, elaborate safety requirements for a high voltage battery test bench are necessary. Therefore a special safety infrastructure is developed and integrated into the system. In the laboratory operation regular maintenance and adaptation of the system is indispensable. In the special case of the high voltage storage system it is not possible to switch off or to ground the high voltage. The maximum DC touch voltage is 120 V [2]. Therefore, within the safety infrastructure the high voltage storage system can be separated into battery stacks of 72 V. To avoid long wiring distances a special disconnector system, which is di-

A NNUAL R EPORT 2010

23

Institute for High Voltage Technology

rectly constructed into the battery shelves, was developed. Figure 3 shows the principle of these disconnectors for the first two levels of the battery rack. All in all, 60 disconnectors are realized in this way.

and the wide operation range of the power electronics in the test bench allow a variation of different parameters to find the optimal topology. For example it will be possible to adapt the voltage level of the storage system and to investigate the charging and discharging behavior for other battery technologies by replacing some of the lead-acid battery strings. The usage of battery storage systems in medium voltage grids requires a high reliability of the system. Therefore, with the monitoring systems for each 12 V block a detailed examination of possible failures within high voltage storage systems can be performed. In addition, a connection of some battery strings in parallel allows a reliability analysis of redundant topologies.

Acknowledgment Fig. 3: Disconnector system in the battery shelve

24

The control of power electronics, the battery monitoring, as well as the door of the safety cage, emergency switches and smoke detectors are integrated into the safety circuit. Hence, monitoring all critical conditions is possible. In case of emergency the test bench can be switched into a secured status.

Intended investigations In the presented test bench it is possible to investigate the charging/discharging behavior of different battery technologies in a high voltage series connection. Efficiency and reliability of such a system are of special interest and can be investigated with the described set-up. The main focus is on the necessary measures for cell balancing in series connected batteries. The modular construction of the high voltage battery

A NNUAL R EPORT 2010

The project this report is based on was funded by E.ON AG as part of the E.ON Research Initiative.

References [1] P. A. Taylor, Update on the Puerto Rico electric power authority’s spinning reserve battery system, Battery Conference on Applications and Advances, 11th Annual , pp. 249-252, 1996 [2] IEC 60364-4-41, Protection for safety - Protection against electric shock, International standard, 2007

Contact Dipl.-Ing. Christoph Roggendorf [email protected] +49 241 80-93032

Field of Research: Insulation Systems & Diagnosis

The current activities deal with the experimental and theoretical investigation of syntactic foam. The investigation focuses on the impact of temperature and electrical dc field stress on the conductivity and breakdown strength of this kind of material. As a main result an exponential dependency of the material’s conductivity on temperature and electrical field stress can be determined. Furthermore, in a limited temperature range, an increase of the electrical breakdown field strength with increasing temperature is observed. Above a certain threshold temperature, the breakdown field strength decreases.

Introduction Syntactic foam is a new composite material consisting of hollow microspheres (HMS) and a polymeric matrix. Due to its specific foam structure the material features a very low density. In the present investigation syntactic foam of hollow glass micropsheres and an epoxy resin matrix is used. Possible filling degrees of the hollow microspheres go up to 55 Vol.-% and reduce the material’s weight by 40 % compared to pure epoxy resin. Syntactic foam is already successfully used as insulation material in high voltage components, where a compact and lightweight design is required. Fig. 1 shows a Scanning Electron Microscopy picture of the material. The name Syntactic foam origins from the specific material structure, which differs from the structure of conventional foamed polymers. If an insulation system is stressed by an electrical dc field, the field inside distributes according to the conductivities of all materials used. Therefore, a detailed knowledge of the conductivity and breakdown field strength of each material or component is required. For solid dielectrics, e.g. epoxy resin, a correlation between the electrical conductivity and the temperature as well as electrical field stress exists [1,2,3]. Furthermore, a dependency of the material’s breakdown field strength on the temperature is known [4]. Ongoing investigations focus

on the dependencies between conductivity, breakdown field strength, temperature and material’s field stress.

Institute for High Voltage Technology

The Behavior of Syntactic Foam under Electrical DC Field Stress

25 Fig. 1: Syntactic foam of epoxy resin and HMS of glass

Theoretical Basics Whereas an ideal insulation material features theoretically zero conductivty, inside the real insulation material charge carrier transport occurs if the material is stressed with an electrical field [1]. Hence, the real insulation material features a very low conductivity, which can be described by [2]: X κ= (qi ni bi ) n = Charge carrier density q = Charge b = Charge carrier mobility

A NNUAL R EPORT 2010

Institute for High Voltage Technology

26

Assuming that the charge q is constant, the conductivity depends on the charge carrier density n as well as the charge carrier mobility b. The following dependency is known for certain temperature and field stress ranges, according to the Van’t Hoff law [2]:

ature as well as the field stress is known [5, 1]: κ(T, E) = κ0 · eαT · eβE Typical coefficients for polymers are [6]: α in the range of 0,074 - 0,114 K−1 and β in the range of 0,018 - 0,128 mm / kV.

n · b ∼ exp(−F/kT ) k = Boltzmann constant F = Thermic activation energy T = Temperature Thus, the correlation between the term n · b and the temperature can directly be measured. Beside the temperature, the field stress inside the material also influences the mobility as well as the density of the free charge carrier. This can be ascribed to the so-called Schottky effect, that describes the reduction of the thermic activation energy F to Fef f by the increase of the electrical field E [2]: s e3 E Fef f = F − 4πǫ0 ǫr e = Elementary charge ǫ0 = Dielectric permittivity of vacuum ǫr = Dielectric permittivity of the insulation material At a constant ambient temperature T the number of free, for the charge carrier transport available electrons increases. Furthermore, the electrical field stress of the material results in a higher charge carrier mobility. Hence, the density n as well as the mobility b of the free electrons increases. According to the equation for the conductivity κ, a higher field stress E results in a higher conductivity. Thus, for a certain temperature and field stress range, the following dependency between the electrical conductivity of a polymer and the temper-

A NNUAL R EPORT 2010

The electrical breakdown field strength of polymers is a function of temperature as well [4]. Here, is has to be distinguished between temperatures below and above the glass transition temperature TG of the polymers. Whereas below TG the electrical breakdown field strength decreases slowly with the temperature, the breakdown field strength decreases rapidly above TG .

Material Production and Object Geometry In the present investigations a hot curing epoxy resin system with a glass transition temperature of 130-140 °C is used as matrix material. The epoxy resin is filled with hollow microspheres of glass by 55 Vol.%. The mean diameter of the microspheres ranges below 100 µm. For the conductivity measurements, plate specimens with a thickness of 3 mm are used. For the breakdown tests specimens with embedded spherical electrodes of 12 mm diamter and 3 mm distance are produced. To determine the breakdown field strength of the material at glass transition temperature, a different epoxy resin system with a lower glass transition temperature of 100-110 °C is used.

Test Setup and Procedure The test setup of the conductivity measurements is built up according to DIN VDE 0303-30 (Fig. 2).

Field of Research: Insulation Systems & Diagnosis

A protective ring electrode configuration is used to seperate the volume and the surface current. A high voltage dc generator with a maximum output voltage of 100 kV provides the required dc voltage. To investigate the conductivity as a function of temperature, the whole test setup is placed in a climate chamber. To determine the dependency on the electrical field stress, the specimens are stressed with high voltages up to 50 kV. The temperature during the measurements is 20 °C. To investigate the dependency on the temperature stress, the temperature inside the climate chamber is varied between 20 °C and 100 °C. For this test the voltage level is 1 kV. The test setup for the breakdown investigations of syntactic foam under dc field stress is built up according to DIN VDE 0303-21 und 0303-22. The specimens are placed in transformer oil to avoid surface discharges. The transformer oil along with a heater pump are also used to heat the specimens to the required test temperatue. The specimens are stressed in a voltage step test. The test starts at 30 kV and the voltage is increased stepwise for 5 kV every 2 min until breakdown occurs.

Experimental Results Fig. 3 shows the dependency of the electrical conductivity on the electrical field stress in the material.

Fig. 3: Dependency between conductivity and field stress

Institute for High Voltage Technology

Fig. 2: Test setup according to DIN VDE 030330

The meausured curve features an exponential increase of the conductivity with higher field stresses. This effect correlates with the theoretical basics of polymers. Hence, it can be ascertained that also for syntactic foam a correlation between the electrical field stress and the conductivity exists.

27

Fig. 4: Dependency of conductivity on temperature

Fig. 4 shows the dependency of the electrical conductivity on the temperature. In analogy to the conductivity measurements for different electrical field stresses, an exponential increase of the electrical conductivity can be observed.

A NNUAL R EPORT 2010

Institute for High Voltage Technology

28

References [1] Bärsch, Kindersberger: Nichtlineare dielektrische Funktionseigenschaften von Dielektrika, ETGFachbericht 110. VDE Verlag. [2] Beyer, Boeck, Möller, Zaengl: Hochspannungstechnik, Springer Verlag, 1986. Fig. 5: Dependency of breakdown field strength on temperature

The results of the breakdown tests are given in Fig. 5. The values are given as mean values with 95 % confidence intervals. Up to a temperature of 85 °C the breakdown voltage increases with higher temperatures. The effect can be ascribed to the internal field distribution in syntactic foam. For higher temperatures the conductivity of the epoxy resin matrix between two adjacent microspheres increases and leads to a more homogeneous field distribution [6]. For temperatures above 85 °C a decrease of the breakdown voltage can be observed. This complies with the theorie of decreasing breakdown field strength of polymers at temperatures above TG .

Conclusion and Outlook In the present investigations a correlation between electrical conductivity and temperature as well as electrical field stress can be observed for syntactic foam. The conductivity of syntactic foam increases exponentially with ambient temperature and field stress. Furthermore, the results show that the breakdown voltage under dc stress depends on the temperature. For the use of syntactic foam as insulation material in high voltage equipment the correlations observed in the present investigations have to be considered.

A NNUAL R EPORT 2010

[3] Huber, Weddingen: Epoxy casting resins with improved electrical conductivity, Colloid & Polymer Sci. 259, pp.852-858, 1981. [4] Artbauer: Elektrische Festigkeit von Polymeren, Kolloid-Zeitschrift und Zeitschrift für Polymere, Band 202, Heft 1, pp.15-25, 1964. [5] Hampton, Chang, Hobdell: What happens to materials under HVDC?, CIGRÉ Session 2000, P2-001. [6] Strauchs, Mashkin: Impact of Temperature and Electrical Field Strength on the Conductivity and Breakdown Field Strength of Syntactic, ETG-Fachbericht 125. VDE Verlag.

Contact Dipl.-Ing. Anja Strauchs [email protected] +49 241 80-94920 M.Sc. Andrey Mashkin [email protected] +49 241 80-94939

Field of Research: Insulation Systems & Diagnosis

After the discovery of superconductivity in 1911 the transmission of electrical power seamed to be possible. Since 1986 ceramics are investigated, which become superconductive at comparatively higher temperatures. Using liquid nitrogen (LN2 ) whith a boiling point at T = 77 K for cooling instead of liquid helium the operating costs of superonducting utilities can be reduced. Thus an economical operation is possible. For insulating the nominal voltage against ground potential, a syntactic foam consiting of a polymer, which is filled with microspheres, is to be applied. Exemplarily a superconducting power cable will be illustrated.

Introduction Prototypes of superconducting power cables mainly have two different designs, firstly the warm dielectric (WD) and secondly the other is the cold dielectric (CD) design [1].

cable is interchanged (Fig. 2). Inthere the inner conductor is electrically insulated at first befor a vacuum insulates thermally. In this case the electrical insulation is cold and consists of polypropylene laminated paper (PPLP), which is impregnated with LN2 . For this reason it it even possible to modify the CD cable design by making the inner conductor massively and cooling the superconductor through the electrical insulation.

Institute for High Voltage Technology

Syntactic Foam in Superconducting Cables

29

Fig. 1: Schematical layout of a WD cable

Both designs are based on a hollow copper conductor taped with superconductor material. The hollow conductor is passed through by LN2 . Thereby the conductor is cooled down to temperatures of about 70 K [2]. The difference between the WD cable and the CD cable lies in their construction of thermal and electrical insulation. The WD cabel insulates the cooled down inner conductor thermally at first by vacuum. Thus the following electrical insulation has ambient temperature and conventional insulation materials as crosslinked polyethylene can be used (Fig. 1). The order of thermal and electrical insulation in a CD

Fig. 2: Schematical layout of a CD cable

Due to the fact that the delivery of the cable is only performed in warm status, the LN2 of before routine tested cables has to be removed. Hence the routine test results loose their significance. After installing the cable it will be filled with LN2 again, but thereby a new untested insulation will be created. To prevent this effect a solid insulation system should be applied. That would mean, that in spite of LN2 dumping for the purpose of delivery, rotine test results keep their significance. The task of a

A NNUAL R EPORT 2010

Institute for High Voltage Technology

solid insulation system schould be fulfiled by a syntactic foam, which is presented in the following.

Syntactic Foam Syntactic foam is a composite of polymer with enbedded hollow microsphers. The microspheres have diameters of several 10 µm. The expression syntactic foam is derived from the foamy microscopic structure of the material. Fig. 3 shows a micrograph of syntactic foam took by a scanning electron microscope. This foam consists of glass hollow microsphers (1) enbedded in epoxy resin (2).

spheres, matrix material or filling degree. 3. The lower permittivity of syntactic foam leads to an decrease of the cable capacity and thus results in a less reactive power demand. 4. Based on the cheaper microsphers compared to the matrix material the material costs of the insulation system decrease. 5. The density of the insulation system is reduced by the considerable lower sphere density. Hence the cable is much lighter and e.g. transport costs can be saved.

Requirements for the syntactic foam

30 Fig. 3: Micrograph of syntactic foam

By variation of different production and material parameters like type of spheres, wall sickness of the spheres, matrix material or filling degree various material properties can be specifically changed. For example the material density depends on the filling degree or the mechanical strength on the size of the microspheres. There are some advantages in using syntactic foam in superconducting power cables as solid insulation over pure polymers: 1. The shrinkage of the foam by cooling down is lower compared to pure material. 2. A specific change of material properties can be done by variation of production and material parameters like type of spheres, wall sickness of the

A NNUAL R EPORT 2010

The main issure of the syntactic foam is to insulate the high voltage potential of the inner conductor of the cable agianst ground potential. Therefore it must have a sufficent breakdown strength. Furthermore the relative permittivity εr as well as the loss factor tan δ should be as low as possible. This would entail less charge energy in case of DC voltage and less losses in case od AC voltage. The losses within the cable occur in form of heat and leads to a raise up of cooling complexity. Because of low efficient refrigerators the total losses increase. Hence a low electrical conductivity of the insulation system is recommended. Is the syntactic foam applied in a power cable it is exposed, in addition to electrical stresses during operation, to mechanical stresses as well. The cable is fureld on a reel so the dielectricum has to be flexible enough to stay proper after fureling. Nevertheless, it has to be rugged enough to hold the inner conductor in the cable center. For commissioning it is nesserary to cool the cable down to liquid nitrogen temperature. Thus, with a shrinking of the syntactic foam is to reckoned.

Field of Research: Insulation Systems & Diagnosis

First results Measurements of the breakdown field strength of syntactic foams in quasi homogenious field (sphere-spherearrangement, sparking distance 2 mm) are performed at room temperature and 77 K. For these tests epoxy is used as matrix material and hollow microsphers made of glass with a mean diameters of 40 µm are used as fillers with filling degrees of 30 vol.-% and 50 vol.-%. The results, presented in fig. 4, show no significant influence of temperature on the breakdown field strength.

More measurements of the breakdown field strength of further syntactic foam compositions will be performed to get the impact of temperature on the breakdown behaviour of syntactic foam.

References [1] A. Bulinski, J. Densley, High Voltage Insulation for Power Cables Utilizing High Temperature Superconducticitiy, IEEE International Conference on Conduction and Breakdown in Solid Dielectrics, 1998.

Institute for High Voltage Technology

Finaly, due to the cables low operation temperature, the syntactic foam needs certain thermal properties. In the modified CD cable design the superconductor is cooled through the electrical insulation system. Hence, to cool down the superconductor by commissioning and to hold the superconductor temperatur under his transition temperature during operation the syntactic foam must have a sufficent thermal conductivitiy. As well as in AC application the heat losses during DC caused by harmonics are in the range of 1 W/m [3], which has to be dissipate through the syntactic foam.

[2] C. Matheus, Technische und wirtschaftliche Einsatzmöglichkeiten supraleitender Energiekabel, Dissertation RWTH Aachen, 2005. [3] R. Bach, W. Prusseit, Supraleitung in EVU-Netzen - Konzept für eine effiziente Energieversorgung von Ballungsräumen, EW Medien und Kongresse GmbH, 2009.

31

Contact Dipl.-Ing. Daniel Winkel [email protected] +49 241 80-93033

Fig. 4: Breakdowm field strength and confidence intervalls of two syntactic foams at different temperatures

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Generation of Synthetic Low and Medium Voltage Grid Models As a consequence of a rising number of decentralized generation as well as electric vehicles, there are new challenges to planning and operation of electric distribution grids. In order to make generally accepted predictions about the necessity of grid reinforcement and the amount of money that needs to be invested, representative grid data is required as a basis for calculations. In the following, different approaches will be presented that enable for a generation of grid models that are equal to the entirety of German distribution grids in terms of their electrical and topographical characteristics.

Introduction and Objective

32

Several studies at the Institute for High Voltage Technology analyse the integration of electric mobility and renewables into the distribution grids at medium and low voltage levels. Different data sets for both voltage levels are publicly available - statistical data about grid assets and topologies for LV level and coordinates of substations for MV level - and can be used in two different approaches to create grid models, which will be described in the following sections. However, a combination of the grid models created with either one of both approaches is possible and allows for an integrated study of operation principles looking at both voltage levels at the same time. Additionally, a combination of synthetic LV grid models with real MV grid data is possible and allows for an assessment where electrical grid data is only available for MV level. Synthetic grid models in this context are defined as data sets, that can be used for load flow study like data from real grids, but that are artificially generated.

Composition of Low Voltage Grid Models An assessment of the integration of new technologies into distribution grids cannot

A NNUAL R EPORT 2010

be done only by taking rated power and a coincidence factor into account. Moreover, predictions about average and maximum load on transformers and cables as well as voltages along lines can only be achieved by load flow calculations, because of the rising complexity and the increasing number of possible operating conditions. As a sufficient amount of grid data that is required for this kind of studies is not available, synthetic grid models have to be developed. The required statistical information is available from [1], [2] and [3]. Data sets that are required for the synthesis of grid models and that are used in the approach developed at IFHT are shown in the following list. A classification of different grids and a connection with medium voltage grids can be done according to different types of settlements. The data considered within this approach includes amongst others length of lines, branching of lines, types and their cross sections, the number of households and their respective transormer rating. Besides, the distance between houses, number of housholds per house and the building type have been taken into consideration. The information about buildings is also used in order to identify the possible size of decentralized generation plants, taking the size of roofs (for PV installation) and the heat demand of differ-

Field of Research: Sustainable Energy Systems In order to provide region specific conclusions their inherent and mainly historically driven topology and electrical grid characteristics must be taken into account. For this purpose a distinction between rural, urban and suburban area has been made, which is incorporated into a two stage model for generation of synthetic medium voltage grids. This model considers a scatter plot of distribution grid stations and a transformer station respectively in order to approximate a realistic medium voltage grid based on the data given. The scatter plot is based on freely available geo referenced local transformer stations. The first step only approximates possible groups of transformer stations, which identify possible topological structures such as ring or mesh solutions. Based on the statistical results related to topological and electrical properties of real medium voltage grids (see above) a travelling salesman approach is utilized to connect different local transformer stations according to the possibly shortest path. An exemplary result is shown in figure 1 comparing the generated grid (highlighted in black) with the original (highlighted in grey).

Topological generation of synthetic medium voltage distribution grid structures The approach illustrated within this chapter generates synthetic medium voltage grids, which enables very general conclusions compared to the approach mentioned above - related to integration potential of decentralized resources. Based on an analysis of electrical and topological properties of available real medium voltage grids first general conclusions related to topological and the respective electrical properties have been drawn. The main data, which was analysed include population density, number of local networks stations, length of ring topology, cable length and the category of regions mentioned below.

A NNUAL R EPORT 2010

Institute for High Voltage Technology

ent types of houses into account. A synthesis of grid models is carried out based on the probability distribution functions of this data and offers the advantage that an infinite number of different grids can be created. Therefore, also grid models with rather extreme values of parameters can be used for evaluation. Acquiring an equal amount of data about real grids and converting it for load flow calculations would require much more effort. Moreover, data sets of the low voltage level that can be used for load flow calculations are quite often not even available from grid operators. The algorithm to create synthetic grid models works based on statistical data of today’s distribution grids. This means that it is not be mistaken for an algorithm for grid planning based on information about the load to be supplied. Therefore, grid structures that have evolved over long periods of time are implicitly represented. Hence, challenges to distribution grids and required measures for reinforcement can accurately be calculated.

33

Institute for High Voltage Technology

References [1] Jörg Scheffler, Bestimmung der maximal zulässigen Netzanschlussleistung photovoltaischer Energiewandlungsanlagen in Wohnsiedlungsgebieten Dissertation an der TU Chemnitz, Chemnitz, 2002 [2] Georg Kerber, Statistische Analyse von NS-Verteilungsnetzen und Modellierung von Referenznetzen http://www.hsa.ei.tum.de/ Publikationen/2008/_Kerber Witzmann_Energiewirtschaft_32008.pdf Seitenabruf: 15.05.2009

Abb. 1: comparison of real grid (black) and generated grid (grey)

34

The second step converts the heuristically generated solution into a cost pattern indicating the minimal expenses necessary related to installation and material cost. Besides, regional aspects such as course of the roads or forests etc. are taken into account, which impact possible routes for cabling significantly. Depending on the regional details taken into account this solution might already present a pretty good approximation, which is then benchmarked with a real grid topology. The two stage approach enables generation of a large number of medium voltage grids, which follow real grid data. If generating a large number of medium voltage grids general conclusions related to integration possibilities of decentralized generation becomes possible.

A NNUAL R EPORT 2010

[3] Georg Kerber, Belastungsfähigkeit von Niederspannungsnetzen bei PV-Einspeisung http://www.energieagentur.nrw.de/ files/Belastungsfähigkeit von Niederspannungsnetzen.pdf Seitenabruf: 11.09.2009

Contact Dipl.-Ing. Claas Matrose [email protected] +49 241 80-94912 Philipp Erlinghagen, B. Sc. [email protected] Thomas Pollok, M. Sc. [email protected] +49 241 80-94958

Field of Research: Sustainable Energy Systems

In many countries a strong transformation regarding the generation pattern of electrical energy is taking place leading to distributed and volatile energy conversion systems. In Germany for example, it is planned to build 25GW offshore wind turbines until 2030. This transformation of generation structures and the continued expansion of cross-border electricity trade are placing new demands on the transmission networks, making measures to increase the transmission capacity necessary. At the Institute for High Voltage Technology, long-term strategies such as expansion measures for the electrical infrastructure and the necessary boundary conditions are analyzed and discussed.

Due to the current changes in the electrical energy supply, transmission networks are facing new challenges. Today’s transmission systems were built and designed for a centralised generation close to load hotspots. In many countries however, a transformation of generation structures, particularly towards renewable energy sources, is taking place. The renewable energy sources are usually decentralized, volatile and located far from load hotspots. Also, there are additional factors such as the expansion of international trade in electricity. These developments will set up new demands on the electrical transmission networks. Network congestion will thus constitute a major barrier for a further integration of electrical energy supply on a renewable basis. In particular, Germany with its strong promotion of renewable energy sources is facing these problems.

EU network model To enable the integration of electrical energy from renewable sources into the transmission system, an expansion of the network will be inevitable. An analysis of the necessary expansions requires an exact knowledge about the structure of the network, load and production patterns as well as their change.

For such considerations a model of the European electrical transmission network was developed at the Institute for High Voltage Technology which also contains load and generation patterns in a high geographical resolution. The model contains several thousand nodes, lines and transformers and more than 2500 generation units (see Figure 1). All network nodes and generation units are geographically referenced so that the effects of local changes in generation can be mapped to the transmission network. The power generation structure is modeled by means of various pan-European databases, again including detailed information about the locations of the power plants. For the mapping of the temporal change in generation patterns different scenarios for the country-based evolution of these patterns are introduced. The scenarios specify the development of the installed capacity for each country and each type of power plant. With the knowledge about planned power plant projects and the age structure of the existing power plants, it is possible to obtain a spatial information about the development of the European generation structure. Thus, an estimation about the location of new power plants and their properties can be made, including an estimation where old generation infrastructure will be shut down.

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Model of the european transmission grid for the assessment of expansion measures

35

Institute for High Voltage Technology

Fig. 1: Model of the transmission grid, utilisations according to OPF in case of high load and strong wind

36

Fig. 2: Installed generation capacity of wind power plants in 2030

The elements of the adapted power plant databases are then assigned as generators to the nodes of the network model, which is made possible due to the geographical informations contained in the models. The load and the installed capacity of wind power plants and photovoltaics can be resolved in node-based data using various methods. As an example, the projected installed generation capacity of wind power

A NNUAL R EPORT 2010

plants allocated to the network nodes in 2030 is shown in Figure 2. The striking peaks at the German North Sea coast result from offshore projects. By combining the models a simplified model of the entire European electricity supply is created. To simulate this model, various methods are used, including Power Flow Calculation (PF), Optimal Power Flow (OPF) or Security Constrained OPF (SCOPF). Depending on the application, the objective function and the constraints of the formalisms can be formulated in a problem-specific way. For example, a cost or emission minimum can be found, in compliance with network restrictions. Despite the large number of parameters and variables, a full OPF including the consideration of reactive power and losses (AC-OPF) for the respective model can be computed for a single point in time sufficiently fast to enable the use in scenario analysis methodologies, Monte Carlo simulations etc.

Field of Research: Sustainable Energy Systems

Network extension measures Using geographical references makes it possible to simulate proposed network expansion measures and their infrastructural impact. For an economic assessment of these measures the generation cost are considered which are determined via an OPF. If the expansion measure removes weak network points it is directly reflected in a noticeable cost reduction. A technical assessment can be conducted for instance by simulating the reduction effect of a specific expansion measure on the network stress in case of an asset outage, the so called (n−1)situation. For an ecological assessment the CO2 -emissions resulting from a cost driven OPF are considered. This approach can be sophisticated by using a simplified Ecological Optimal Power Flow which is able to reveal expansion measures that are economically, but not ecologically effective and vice versa. At the Institute for High Voltage Technology the European network model and the methodology for the assessment of future expansion measures are constantly refined. For the purpose of highly detailed analyses, the assessment of the European transmission network is expanded by also considering the underlying network levels. Furthermore, the networks of the Scandinavian countries, the United Kingdom and Ireland have been added to the model of continental Europe. This enables the simulation of future large scale projects such as the “North Sea Grid“ which has been proposed by the

European Commission and is intended to connect the offshore wind farms with the respective European neighboring countries. Furthermore, also the modeling and the associated assessment of the competing technologies for these kinds of projects, such as hybrid networks, flexible AC transmission systems (FACTS) and meshed high voltage DC networks (meshed HVDC), are subject to the current and future research at the Institute for High Voltage Technology.

Summary In order to facilitate the integration of power generation based on renewable energy sources into the European network the Institute for High Voltage Technology uses its existing models which are constantly refined to assess different network capacity enhancement strategies. As long-term solutions, network expansion measures, in particular of the European transmission network, are investigated and assessed using high detail models developed at the Institute for High Voltage Technology.

Acknowledgement The network model was developed for the EU-FP7 project IRENE-40 (www.irene40.eu).

Contact Dipl.-Ing. Hendrik Natemeyer [email protected] +49 241 80-94913 Dipl.-Ing. Sebastian Winter [email protected] +49 241 80-94914 Dipl.-Ing. Martin Scheufen [email protected] +49 241 80-94941

A NNUAL R EPORT 2010

Institute for High Voltage Technology

In order to model different states of the power transmission network, different scenarios are defined which are suited to illustrate the varying conditions of the network. They are designed to simulate different load patterns as well as the fluctuating availability of renewable energy sources by considering also local power peaks for instance due to strong wind conditions in the North Sea.

37

Institute for High Voltage Technology

38

Optimisation of the operation of integrated supply systems for electricity and heat Based on a simulation model for the analysis of heat pumps and dispersed units for combined heat and power generation a tool has been designed which calculates the optimal operation strategies for these units by applying generation and load management. At the same time the influence of the addition of electrical and thermal storage systems on distribution networks is analysed using different objective functions.

Introduction A strong decentralization of energy supply is expected to occur in Germany in the course of the coming years. As a result of a growing share of renewable energy sources with their inherent volatility and uncontrollability of generation, action is needed in order to balance the influence of fluctuating renewable energy sources and to ensure an uninterrupted power supply. The integration of energy storage systems into the existing electricity networks is an important component in order to realize the projected expansion of renewable energy sources and at the same time ensuring security of supply. The use of thermal energy storages offers the possibility to adjust the operation of heat pumps and cogeneration units to the electricity demand in the network without endangering the securing of the heat supply.

Modeling of Plants The decentralized conversion technologies are mainly connected close to the loads in the respective distribution networks at a low voltage level and are usually operated independently (from each other). Systems of combined heat and power (CHP) work mostly in heat driven mode, which means that their operation depends on the local heat demand. To investigate the influence of increasing decentralized energy production in the existing networks and in order to perform an economic and ecological analysis, a bottom-up modeling approach of the

A NNUAL R EPORT 2010

various decentralized conversion technologies to account for the stochastic operating behavior has been implemented in MATLAB [1]. Furthermore, the use of thermal and electrical storage technologies in intelligent distribution networks is examined under different operating strategies. The technologies considered are combined heat and power plants, heat pumps and, additionally for each facility, a thermal tank to avoid the undesirable clock-operation of the systems. In addition, stochastic generation profiles for photovoltaic and small wind turbine generators based on historic time series data are used. Exemplarily the approach for CHP units is described. The rated power of the cogeneration facility is set in the model to a reasonable level with respect to the heat demand in the building assuming dual-mode operation. In addition, a peak load boiler is included for each plant to support CHP on the coldest days of the year. As input parameters the model uses a heat load profile. Depending on building type, year of construction and rehabilitation a practical supplement is added [2]. Each CHP plant is either turned on or turned off at each time interval of the day according to the local heat demand and the condition of the existing thermal storage. Since the state of the thermal storage and heating demands of the individual objects differ from each other, the output of the model comprises a stochastic production profile for each day of the year. In Fig. 1, for example, the production pro-

Field of Research: Sustainable Energy Systems

20

15

15

10

10

5

5

0

0

2

4

6

8

10

12

14

16

18

20

22

0 24

Fig. 1: Synthetic production profile for CHP (example)

Storage systems

Optimisation of the operation The modelled generation profiles represent the heat driven operation of CHP units. Consequently the operation hours depend on the demand of thermal energy by the object to be supplied. In this operation mode, the generated electricity is fed into the distribution network without taking the current demand for electricity into account. Combined with additional local electricity generation (e.g. from photovoltaic) a situation might occur in which the surplus of energy endangers the stable operation of the electric power grid. 30

Balance of electrical power

generation of heat

thermal power [kW]

electrical power [kW]

generation of electricity 20

of their placement from exogenous influences.

Institute for High Voltage Technology

file (electricity and heat) of a cogeneration plant for a house on a winter day is presented. It shall be noted that the profiles for each household are generated individually so that the entirety of units is modelled accurately.

20

A range of technologies is able to store thermal energy. However using today’s knowledge only systems based on sensible heat storage1 are economically feasible. At temperatures which are typical for domestic heating the typical technical solution would be an insulated water reservoir. Compared to other materials such as concrete or sand the high thermal capacity of water (1.17 Wh/(kg*K)) leads to a sufficiently compact design. Latent heat storage systems using phase change materials or systems based on reversible chemical processes are not included in the model because of their high costs and limited availability on the market. However, they can be easily added to the model by using new parameters for costs and losses. For storing electrical energy the assessment was limited to different types of battery storages differing in investment costs, efficiency and standby losses. With the focus on distribution networks they can be seen as the preferred technology because of their flexible size and the independence

10

0

-10

0

2

4

6

8

10

12

14

16

18

20

22

-20

-30

time

Heat driven CHP operation

Generation management

Fig. 2: Intelligent operation control of CHP units

By the introduction of thermal energy storages it is possible to shift the operating hours of the CHP units to time intervals with insufficient generation (see Fig. 2) because the heat storage guarantees that the demand of thermal energy is met at any time. This optimisation can be carried out by using the objective of a distribution system with a high degree of autarky. This means in this case that the distribution network is almost independent of the transmission network and is mainly supplied by local generation from renewable energy sources and CHP. A second ap-

1

Sensible heat storage systems directly store thermal energy by heating a storage medium like water, stone oder salt.

A NNUAL R EPORT 2010

39

Controllable CHP unit in operation

Balance ofel. power

Target

Controllable CHP unit out of operation

and demand of roughly 20%. In order to also compare production costs of the specific technologies, life spans and costs have been assumed according to [3]. The analysis shows that a combined operation of CHP units and thermal storage systems within an intelligent distribution network can provide similar levels of autarky at much lower costs compared to the installation of battery storage. Firstly, the advantage of heat storages rises faster with lower additional costs. However, there is point of saturation when additional heat storage yields no further benefits. 90%

Grade of energetic autarky

Institute for High Voltage Technology

proach for the examination is the harmonisation of power flows. The objective function aims to reduce the maximum load of the assets in order to reduce transmission losses. Apart from the different objective functions both approaches are based on the implementation of a generation management for CHP units and a load management for heat pumps, respectively. In the following the approach for CHP units is described. Based on the temperature of the heat storage and a number of technical parameters it is checked whether the current condition of the CHP unit necessitates changes to the current state of operation. Using a projection for the following 24 hours the state of operation is analysed with a time resolution of 15 minutes. If deemed beneficial a change in the state of operation will be done. The constraints are shown in the simplified Fig. 3.

85%

80%

75%

70% 0

5.000 Li-Ion

40

CHP management

Set out of operation

Minimal duration of operation reached

Water

15.000 Annuity [€/a] Li-Ion+Water

20.000

25.000

Lead+Water

Fig. 4: Autarky of the system using different storage technologies

Set into operation

Thermal load covered Pel < required load

Pel < required load

10.000 Lead

New temperature of storage

Boiler

Generation management

Fig. 3: Generation management for CHP units

The approach for heat pumps follows the same principles while different technical parameters have to be taken into account and the objective of the controlling is inverted as a heat pump is an electrical load.

Exemplary results The developed model has been used for an exemplary supply area with 102 private households. In the network area examined the local electricity generation from CHP (56%), photovoltaic (36%) and wind energy (8%) matches approximately the annual energy demand of 300 MWh. However, there are seasonal differences between generation

A NNUAL R EPORT 2010

An example for the relationship between costs and autarky is shown in Fig. 4. It displays the results for the supply area described above using today’s costs and life spans for battery storage systems. A scenario analysis has been carried out which showed that for any generation mix there is always a region in which heat storages are superior to batteries when considering the ratio of cost- to autarky increase. The combined operation of electrical and thermal storage (dashed lines) allows higher grades of autarky at lower costs compared to the sole operation of battery storages. Whereas the objective of autarky aims to bring the load balance down to zero, the target value becomes variable with the objective of power flow harmonisation. For the supply area described above a foresight on generation and load has been done with a resolution of 15 minutes. The mean value

Field of Research: Sustainable Energy Systems energy sources and CHP a supply area can be self-sustaining for some periods of time. For this purpose the objective function for load and generation management could be changed in a way that the energy exchange with the transmission grid becomes dependent on the electricity prices on the wholesale markets in order to generate additional income towards the distribution grid.

References

80

60

[1] T. Smolka: Ökologisch-technische Auswirkungen dezentraler Energieversorgungsszenarien mit Blockheizkraftwerken in elektrischen Verteilungsnetzen; Disseratition, RWTH Aachen, Aachen 2008

Balance of electical power [kW]

40

20

0

-20

-40

Institute for High Voltage Technology

of the load balance then is the target value for load and generation management. This relationship is shown in Fig. 5 for a selected week. The target value is time dependent and rose slowly during the week from a starting point of -18 kW. As a result the analysis of the assets shows that the loads on lines and secondary substation transformers are significantly lower than in the unaffected operation.

-60

-80 0

12

24

36

48 Status quo

60

72

84 96 108 120 132 144 Time [h] Autarky (energ.) Homogenization

156

Fig. 5: Load balance of the system in a selected week depending on the objective function

Conclusion and outlook In a first step the introduced method allows simulating the operating behaviour of dispersed energy conversion units and the influence of a significant number of these units on the distribution networks. Second, a generation and load management has been implemented in order to control the load balance of the supply area according to the chosen objective function. Therefore the operation of thermal and electrical energy storage has been included. In the future the model shall be linked to an optimisation tool for designing the setup of the heat and electricity supply in a specific area. Using local generation based on renewable

[2] W. Lutsch, F. Witterhold: Perspektiven der Fernwärme und der Kraft-Wärme-Kopplung; Arbeitsgemeinschaft für Wärme und Heizkraftwirtschaft - AGFW e. V. , Frankfurt 2005 [3] U. Bünger et al.: Energiespeicher in Stromversorgungssystemen mit hohem Anteil erneuerbarer Energieträger; ETG Task Force Energiespeicher, Energietechnische Gesellschaft im VDE (ETG), Frankfurt 2009

Contact Dipl.-Ing. Sepideh Doroudian [email protected] Dipl.-Ing. Stefan Krengel [email protected] +49 241 80-93034

A NNUAL R EPORT 2010

41

Institute for High Voltage Technology

42

Agent-based traffic models for local emissions analysis By using an agent-based traffic model and a separate examination of every single vehicle, the streets capacity utilization can be determined in a very high time resolution. The previously made analyses of time-differentiated charging processes can be extended by another spatial component. With an examination of movement patterns it is possible to get an estimation for the locally generated emissions and to analyze the spatial linkage of loading processes of the vehicles.

Introduction The expected strong increase of electric mobility outlines a new challenge for grid operators regarding reliable and safe operation of energy supply networks. An integrated model of electricity networks and transportation networks gives the opportunity to analyze the impacts on power supply caused by a comprehensive integration of electric vehicles considering both energy and transportation flows. Therefore, synthetical driving patterns are generated in the beginning and are substituted by real driving patterns later on.

This was extracted in seperate layers. (Fig. 1) shows the levels „street network“ and „residential areas“.

Fig. 1: Street network and housing areas

Traffic model The traffic and the road network have a special significance in the traffic modeling. With the knowledge of the exact position of each vehicle, it is possible to state the availability and the potential of each vehicle and its storage capacity in a certain region. In a further step, the knowledge about the driving behavior is used to control charging processes. In order to generate schedules for the vehicles, it is necessary to know where the drivers live, what job-related drives they do or where they drive to for private reasons. In addition to the street information, the geological data that is used also contains information about the land use.

A NNUAL R EPORT 2010

Integrated modeling Through a combination of both models, the traffic network and the electricity network, an integrated traffic and energy flow model is realized as it is shown in (Fig. 2). Another component is a central control which effects and controls the charging process. The more information it receives from the vehicles, the more effective it is. The data about the standing times and future driving destinations generate the biggest benefits. The vehicle must be informed by the driver with these data, e.g. through a modified gps-system, so that the charging processes during the parking time can be controlled optimally in time and space.

Field of Research: Sustainable Energy Systems

Fig. 2: Integrated simulation model

Results of the integrated modeling The impacts of an uncontrolled charging behavior of the vehicle population on the electricity network was analyzed. Therefore three scenarios for the availability of charging infrastructure, „everywhere“, „at home“ or „at home and at work“ were considered. The increase of the overall load caused by electric vehicles in a certain region is illustrated in (Fig. 3). The considered rural region has a size of approximately 2000 km2 with 290.000 inhabitants and 150.000 automobiles of which 30% are assumed to drive electrically. The peak load is 360 MW. It has to be noticed, that in consideration of a single day only 65% of the electric vehicles actually drive. The rest is parked and connected to the grid in a best case scenario.

Fig. 3: Load increase by electric vehicles

In a further step, the ecological impacts of electronical vehicles on their environment can be captured and a comparison to conventional automobiles can be made. In order to be able to make an appropriate comparison between electric vehicles and conventional vehicles, it is necessary to create definite, ecological critical areas, the socalled impact categories, in which the comparison should occur. Therefore categories are considered in which a significant difference between the two types of cars is noticeable in respect of the ecological impacts. The investigations of the fine dust is one of the categories, because an electric vehicle does not emit fine dust. Furthermore, an electric motor does not produce noise or any local CO2 emissions. The addressed impact categories are therefore fine dust, noise and CO2 emission. In order to outline the influence of an integration of electric vehicles in these categories, there has to be a description of the current load state in the considered region. Based on this description, the influence on the regions is shown by increasing the number of electric vehicles. An electric vehicle saves, compared to a conventional vehicle, up to 70% of the emission in the fine dust category. The remaining 30% are emitted by tire and street abrasion and cannot be avoided by electric vehicles. With the help of data about the emission per automobile, statements about the streets fine dust pollution can be made in g/km. A conversion methode was established with real measured data. The resulting factor serves as an expected value for statistic analyses for the probability of a limit violation. As a result of the simulation in the fine dust category it has to be recorded, that a low penetration rate, which is realistic in the future, would only have a small impact on fine dust in the annual average. A percentage of 10 % of electric vehicles would lead to a decrease of fine dust

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Ecological impacts

43

Institute for High Voltage Technology

pollution by 1.7 % in a strongly used street. However, this decrease would have a substantial influence on the probability of the limit violation. It would fall from over 60% to under 40% in a strongly used street as it is shown in (Fig. 4). A noticable improvement of the quality of living is only realistic with 30% of electric vehicles on a medium term.

Fig. 4: Fine dust concentration for different scenarios and the probability of limit violations

44

The calculation of the CO2 emission follows the same pattern. In order to calculate the load, the data about the CO2 emissions of all vehicles on the road is used. This is based on the noise averaging level LDEN. The calculation of each factor depends on the traffic volume and the average percentage of trucks on the road. In the

A NNUAL R EPORT 2010

range of the local CO2 emission a tremendous relief through electric vehicles can be expected. The road traffic noise reduces enormously through an integration of electric vehicles mainly in housing areas. With a percentage of 10% of electric vehicles the load decreases about 6% in streets where the speed limit is 30 km/h and in streets where the speed limit is 50 km/h it only decreases about 2%. Overall, a strong integration of electric vehicles leads to a relief regarding the considered impact categories especially in the city centers. However, with a low penetration rate the effect is small. Therefore, electric mobility is going to improve our quality of living only on a medium term.

Contact Dipl.-Ing. Thomas Helmschrott [email protected] +49 241 80-97340 Dipl.-Wirt.-Ing. Eva Szczechowicz [email protected] +49 241 80-94916

Field of Research: Systems Engineering and Asset Management

The reliable estimation of quantity structures of the equipment within electrical transmission systems represents a challenge all over the world due to insufficient and inconsistent public data. The results of such assessments, particularly of expensive equipment such as power transformers, are part of the decision-making in the context of business activities in the energy sector. The presented method of synthetic mesh generation enables calculatign a network for a country on the from scratch, which fulfills the supply task, whereas the procedure itself is based on publicly available information and technical, economic and geographical conditions. The resulting synthetic network enables the approximation of the quantity structures of expensive equipment in the actual electrical transmission system of a country.

Introduction There is no global data base available to determine the installed quantity structures of equipment in electrical transmission systems of individual countries or regions. In European countries, there is only information on the number and cumulated capacity of power transformers available as well as of route lengths available. These are subject to the inconsistent and sometimes incomplete inquiry. The exact knowledge of the actual equipment quantity is an important basis for economic and strategic decisions such as the buying of power systems, infrastructure investment, opening new markets for products and services as well as the assessment of the own market position. Especially in countries with poor public data like emerging markets, an option to estimate the quantity structures is quite important. A preliminary approach for the derivation of a transferable regression model for the number and capacity of power transformers on the one hand confirmed the presumed correlation with selected input data, on the other hand showed high divergences when transferred to countries - especially non-European ones - outside the dataset for parametrization. A significant influence of

parameters not regarded due to a lack of availability and of operator-specific philosophies is likely to be responsible. For the synthetic mesh generation, the historical developments are considered to be not reconstructable. Therefore, geographical boundary conditions and the spatial distribution of loads and generation are of relevance. Based on current input data, a minimal reference network is developed which can fulfill the supply task of the transmission network. Precondition for the application is a centralized production structure by large-scale power plants in the investigated countries. Results of the process are the network topology and the quantity structure of costly equipment in a synthetic network.

Analysis of available input data Cumulative data with a national focus is collected worldwide but the range of variables, criteria for the collection and spatial resolution differ greatly between the gathering competent authorities. Thus, collection and standardization of data on quantity structures of equipment out of individual, scattered sources is impossible for a global data base with data such as • installed generation capacity

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Synthetic generation of network topologies using publicly available data

45

Institute for High Voltage Technology

46

• anually generated, imported and exported electric energy • annual energy consumption and losses • total population • degree of electrification In contrast, geo-referenced data assign geographic points or areas to properties. Properties of surfaces are then present in the form of raster data. Geo-referenced data about generation sites is also available worldwide or can be obtained by a proven systematic approach with calculable effort. The initial degree of the coverage of centralized generation compared to the cumulative variables is usually over 70%. Geographical information such as the discrimination between land and water areas, elevation profiles etc. are available for all countries worldwide. Globally available geo-referenced data related to load balancing, which are used as the basis for the estimation: • population density

Fig. 1: Flowchart of the process

To determine the load, a correlation between the distribution of electrical load and the population on the national territory is adopted. Basis of this assumption is the mutually dependent position of focal points of population and economic sectors. In figure 2, the determined load map for the Netherlands is shown.

• land use • electric night lights • degree of electrification (restricted) The further, combined processing of these different types of input data is shown in the next section.

Procedures and results of a typical application The process of synthetic mesh generation creates a technically acceptable, reliable and economic electric transmission network. The process follows the steps shown in figure 1. There are several variants of a technically acceptable system generated, whereas the most economical result is finally selected. The steps are outlined below.

A NNUAL R EPORT 2010

Fig. 2: Identified load distribution

The algorithm to generate the supply areas of the distribution network level (including the high voltage level) uses the determined

Field of Research: Systems Engineering and Asset Management

Discussion and outlook The method shown by the example of the Netherlands identifies a circuit length for the transmission network of approximately 90% and a number of transformer of approximately 92% of the real values. Crossborder interconnections and different high voltage levels are currently not taken into account. In contrast to purely statistical methods, the synthetic mesh generation is able to consider both the geography of a country as well as structural information such as the distribution of the population. The generated electrical power is also based on technical and economic conditions to ensure the best possible representation of reality, with the aim to approximate the quantity structure of expensive equipment for as many countries as possible. Due to the „green field“ approach of the economic optimization of an electrical network by planning from scratch, without taking historical data into account, the synthetic mesh generation additionally shows potential concerning the generation of optimal reference- or target networks for the current supply task.

Kontakt Dipl.-Ing. Christian Hille [email protected] +49 241 80-94915 Dipl.-Ing. Tilman Wippenbeck [email protected] +49 241 80-90015

Fig. 3: Synthetically generated network with supply areas including substations and power plants

A NNUAL R EPORT 2010

Institute for High Voltage Technology

load map. Each of these supply areas includes, as a black box without detailed structural mapping, the load of a calculated portion of the geographic land area up to a certain limit. In all cases, a substation is used as a central anchor point between the transmission and distribution network level. All of the substations, along with the production sites, form the nodes that need to be connected by the transmission system. Both the rating of power transformers and the interpretation of the electric circuits of the transmission system are designed for a reliable supply. An overloading in the maximum load case and the failure of equipment or a circuit is avoided by an appropriate design and redundancies. From all generated power variants, the costoptimal synthetic network is identified as a result of the procedure by comparing the profitability. The economic evaluation includes annual pay ranges out of operating costs and the annual investment costs of the equipment. Fig. 3 shows such a synthetic system for the Netherlands. Displayed are the supply areas, the transmission grid and the production sites. Their level of performance is expressed by the diameter of the corresponding circular areas.

47

Institute for High Voltage Technology

48

Challenges for protection systems in distribution grids with significant dispersed generation High future penetration levels of decentralized generators and storage devices will potentially endanger the functionality and selectivity of traditional protection systems in electrical distribution systems. The broad range of deciding factors prevents up to now an efficient assessment of the threat potential and an objective decision on necessary remedies. This article gives an overview on the influencing factors, shows some example protection threats and sharpens the questions to be answered.

Protection in distribution systems Selective protection of low voltage (LV) and medium voltage (MV) distribution systems is mainly achieved by non-directional overcurrent protection devices. Examples of such are: • fuses (LV+MV) • LV line switches for homeinstallation • overcurrent-relais + circuit breakers Successful usage and rating of those protective devices depend on several assumptions. Firstly, one assumes a mono-directional power-flow to the periphery of the distribution system in normal and in faulted operation. Secondly, a significantly differing current level between those states are assumed. In the absence of decentralized generators and storage devices the mentioned protective devices are a robust, economic and fieldtested solution for ensuring an adequately protected operation of distribution systems.

Scenarios of distribution system developments The growing integration of decentralized generators (e.g. photovoltaics, combined heat and power plants, small wind turbines and run-of-river power stations) and of bidirectional storage devices (e.g. electric vehicles) - in the following summed up as “decentralized infeed” - will constantly or from

A NNUAL R EPORT 2010

time to time violate the validity of the named assumptions. Besides this fundamental change of the current flows in distribution systems experts discuss and predict a change of the grid itself. Migration from radial feeder operation to closed rings or even meshed structures may improve voltage maintenance and minimize influences of decentralized infeed on power quality. Experts consider an increase of controllable components and sensors necessary. Sectioning points, loads, generators and storage devices shall be (remotely) controllable. For this purpose one needs an extensive amount of automation and communication technologies. Smart meters may become a way to extend the existing minimal amount of sensors drastically. The sum of these developments aims towards an active control in distribution systems’ operation. The latter shall enable an optimal utilisation ratio of the equipment and an efficient use of electrical energy. To ensure a high reliability of supply temporary islanded operation is suggested to avoid supply interruptions in case of blackouts of the upstream grid. The changes depicted by these scenarios need huge investments by distribution system operators in order to build up the necessary infrastructure. Huge circuit lengths and huge component numbers lead to the

Field of Research: Systems Engineering and Asset Management

assumption, that a gradual change will rather take place in existing distribution systems than a sudden change over all systems. Thus, classical and new active distribution systems will prospectively coexist for a long time.

Scenarios of decentralized infeed behaviour Due to the change of the distribution systems’ structure and components a development of the requirements towards decentralized infeed is expected and needed. Sources connected to the low voltage grid shall nowadays disconnect from the grid rapidly in case of a power system fault or in case of unintentional islanding. This passivation of the grid is among other things intended to reestablish the conditions that were assumed when designing and dimensioning the traditional protection system. Today, distributed generators connected to the medium voltage network are already required to be able to supply ancillary services. Due to the increasing penetration of lowvoltage distribution systems with decentralized generators and storage devices (e.g. electric vehicles) we expect an extension of this requirement to low-voltage infeed. In this case, the now favored passivation is no longer applicable to preserve traditional protection system functionality. In dis-

tribution systems with local high penetrations of decentralized infeed might apply the Microgrid-concept in the future, part of which is a connection to the overlying network with lower rating than necessary to supply the installed load at once. Without load-shedding a simultaneous disconnection of all decentralized infeeds would result in critical overload conditions. As long as the described infrastructure for controlling the distribution systems is not available comprehensively, an uncontrolled behavior of the distributed infeed needs to be assumed.

Threats to traditional protection systems In case of uncontrolled decentralized infeed we expect an influence on the functionality of traditional protection systems. An example is an unacceptable delay of operation of protective devices. This for instance may be caused by blinding of feeder protection in a situation as depicted in fig. 1. This effect may lead to a loss of service live or destruction of the protected components and may furthermore endanger persons or animals. Another example is an unselective sympathetic tripping of healthy feeders in case of a fault in neighbour-feeders as depicted in fig. 2. Under certain conditions the protec-

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Abb. 1: Blinding of feeder protection FA by Generator G. Relevant parameters for modelling according to IEC 60909

49

Institute for High Voltage Technology

Abb. 2: Sympathetic tripping of feeder protection FB by fault in feeder A. Relevant parameters for modelling according to IEC 60909

tion of feeder B will trip although the fault itself is located in feeder A.

50

Other cases similarly lead to unnecessary costs and service interruptions. One main challenge of the integration of decentralized infeed is, that protection investigations need to simultaneously consider so far independent parts of the distribution system. One needs to consider the grid in total with all connected components. Even in case of a controlled operation of the distribution system proper protection needs to be ensured, for instance when automation or communication systems fail. Furthermore, new system states previously unknown will exist, which will violate completely the assumptions that originally lead to today’s protection concepts. These system states may not be observable and will lead to a lack of operation of the protection concept. An example is an inner overload situation in which traditional protection concepts will no longer trigger as the overload currents do not pass the sensing points of the installed protection system.

A NNUAL R EPORT 2010

Resulting Questions The underlying basic problem is that the limit of the proper operation of the installed protection systems is not yet as well quantifiable as necessary. Single case investigations as done by protection divisions and by researchers offer appropriate solutions for specific distribution system cases. Yet, it does not seem possible to transfer those specific results efficiently to the broad range of diverse distribution systems today and in the future. The mentioned single case investigations will not become superfluous as knowledge of the functional limits of traditional protections system rises. Development of methods to gain this knowledge in a more general way will be a key to efficiently answer the following questions for a huge number of distribution systems: • In which distribution systems exist a need for an adaption of the protection system? • Is it possible to develop rules, which allow for a quick and efficient esti-

Field of Research: Systems Engineering and Asset Management

• Of what amount is the actual and foreseeable necessity of adaptions in the distribution systems? • Will functional limits of the protection concepts be reached without prior violation of other restrictions (voltage maintenance, steady state current carrying capability)? • Which measures will be applicable to preserve the traditional protection system functionality? • Which requirements apply to newly designed protection concepts? • Which measures are economically and technically appropriate?

The still unreached identification of functional limits of the traditional protection systems is due to the new complexity of protection investigations. At the moment a project to answer the above questions is actively developed by the Institute for High Voltage Technology in close cooperation with industrial partners. As a first step a simulative analysis will be conduct therein. Secondly, the necessary modeling and the results will be validated on real distributions systems in the test and verification center of IFHT.

Institute for High Voltage Technology

mation without detailed modelling of the single case?

Kontakt Dipl.-Ing. Tilman Wippenbeck [email protected] +49 241 80-90015

51

A NNUAL R EPORT 2010

Institute for High Voltage Technology

52

Life Management of Power Transformers based on Online Monitoring Data Life management of power transformers aims to operate them close to the end of their lifetime. This report emphasizes on the prediction of their remnant lifetime regarding aging of solid insulation by different operation scenarios. The work covers enhancing the aging model by considering more vital aging causes, investigating the correlation of operation to aging, and developing the remnant lifetime forecasting approaches. Time-series data from an online monitoring system are employed.

Introduction The condition of the solid insulation is irreversible. Thus, the transformer’s lifetime is generally constrained by its solid insulation lifetime. A well-known end-point criterion is the degree of polymerization (DP) of 200. However, the transformer has to be out of service in order to cut some parts of the paper for this test and therefore the condition cannot be assessed continuously. Hence, an approach to derive aging according to its causes which are related to the operation is the means to acquire aging characteristics and is the main goal of this work. This information can assist decision making in order to choose the optimal action on the transformers, e.g. limiting the overloading, controlling the cooling units, doing the drying process or planning for the relocation or replacement of the aged units. The research overview is shown in Fig.1.

Fig. 1: Overview of this research

Aging model by fuzzy logic approach The solid insulation is made mainly of cellulose and therefore deteriorates over time due to three main causes; heat, water and oxygen. In order to ease and enhance the

A NNUAL R EPORT 2010

verification of the aging model with actual data from a large population of transformers, the model is firstly developed by a fuzzy logic approach (FL) based on experiment results from [1]. Hotspot temperature (HST) and water in paper (WIP) are used as inputs. Oxygen, which is another main cause for aging, is not taken into account in these models because it is mostly kept in low level by the oil preservation system [2]. The rate of aging is the outcome of the fuzzy model. The age can be finally derived by the summation of all results from the multiplication of rate of aging and time interval. The developed procedures can be seen in Fig.2.

Fig. 2: Age derivation by FL

The zero-order Sugeno model, a precise fuzzy logic model, is employed here due to its computational efficiency.

Influence of operation on aging The top oil temperature (TOT) is for calculating the HST. The neural network (NN) may become a useful tool for determining the TOT. Time-series data from historical databases belonging to a monitoring system of a 5-years-in-service transformer for about 2 years are used to train the network. This sample transformer is

Field of Research: Systems Engineering and Asset Management

Fig. 3: Layouts of developed neural networks

Here, the measured TOT is employed as test input for the case RNN1, and the TOT resulting from the previous time of the same model, is used for the case RNN2. The mean absolute error (MAE) between measured and computed TOT defines the breaking condition and is the main factor to evaluate the performance of the model. Moreover, the absolute error distribution of all data of each method is also investigated. In some cases, the maximum absolute error (Max AE) is really high due to abnormal operation. The performance of all developed methods and conventional calculation method are shown in the table below. Test data1 refers to timeseries data acquired for 150 days while test data2 is the data measured for 3 years. Methods

Train data

SNN1 SNN2 TPNN RNN1 RNN2

3.22 2.58 2.51 0.17 0.17

MAE( ◦ C) Test Test data1 data2 3.80 3.14 3.24 0.22 3.20

3.47 3.12 2.94 0.34 5.87

Max AE( ◦ C) 90% 95% of test of test data2 data2 7.78 10.32 5.73 7.16 5.73 7.02 0.82 1.28 16.43 25.95

The obtained results from both SNN models together with the measured TOT are illustrated in Fig.4. By taking the status of cooling units into account as SNN2, the

computed results are closer to the measured ones, especially during strong variation periods of current or ambient temperature. 800 TOT by measurement TOT by SNN1 TOT by SNN2 Load current

75

700 Current (A)

Top oil temperature (°C)

85

65

600

55

500

45

400 75

76

77

78

79 80 Time (days)

81

82

83

84

Fig. 4: Measured and computed top oil temperature by SNN1 and SNN2

SNN2 and TPNN provide almost the same performance because here the applied input data are the measured data. The amount of test data gives nearly no effect to their performance. Therefore, these methods can be used for long term forecasting purposes.

Institute for High Voltage Technology

characterized by 350 MVA, 400kV, and an OFAF cooling type. The feed-forward architecture consisting of one hidden layer is used here. The networks are trained by the Levenberg-Marquardt algorithm. The activation function of the hidden layer is the sigmoid and a linear function is assigned to the output layer. Static NN (SNN), temporal processing NN (TPNN) and recurrent NN (RNN) [3], are applied in this work and their layouts are shown in Fig.3.

Approach to forecast aging characteristics In order to forecast aging characteristics, extrapolation of existing data is not possible due to its dependencies on several factors. The profiles of aging relevant factors are derived by a pattern learning approach and used for forecasting purposes. Statistical and clustering methods are used for achieving the daily and yearly patterns of ambient temperature and load current from timeseries data. Both are deployed together by the modulation principle.

Implementation and results The developed aging models are fed with data from the sample transformers being in service for 5 years. Its prognostic aging behavior is forecasted with three operation scenarios: • Scenario 1: values of ambient temperature, load current and water in oil are annually constant as from their profiles

A NNUAL R EPORT 2010

53

• Scenario 3: load current is limited at only 50%. The simulated aging characteristics are presented in Fig.5. 15

9

References

6 Age of 5−year−in−service Prognostic age with scenario1 Prognostic age with scenario2 Prognostic age with scenario3

3 0 0

5

10

15

20

25 30 35 Time (years)

40

45

50

55

60

Fig. 5: Predicted aging characteristics

Based on the nominal lifetime of 15 years and the prediction with scenario 1, the remnant lifetime of this transformer is around 16 years. By considering load growth, the remnant lifetime is reduced, while it is increased by limiting load.

54

acteristics of power transformers including aging from the commissioning until the present and also the prognostic one can be derived by the developed approaches. The remnant lifetimes can be achieved by different operation scenarios. This information is helpful to support decision making for network operators, maintenance personal and network planners.

12 Age (years)

Institute for High Voltage Technology

• Scenario 2: load growth is assumed as 3% per year and

Conclusion The aging models based on the experiment results from [1] are developed. The rate of aging for each period of time can be derived according to HST and WIP. The fuzzy logic approach is applied in order to make the models suitable for verification and adjustment by a numerical method with the actual data. The neural network approach is applied for computing TOT. The aging char-

A NNUAL R EPORT 2010

[1] A. M. Emsley and G. C. Stevens, Review of chemical indicators of degradation of cellulosic electrical paper insulation in oil-filled transformers, IEEE Proceedings Science, Measurement and Technology, 1994. [2] IEEE guide for loading mineral-oilimmersed transformers. IEEE Std. C57-91-1995. [3] Q. He, J. Si and D. J. Tylavsky, Prediction of top-oil temperature for transformers using neural networks, IEEE Transactions on Power Delivery, 2000.

Contact Tirinya Cheumchit, M.Sc. [email protected] +49 241 80-94937

Field of Research: Systems Engineering and Asset Management

Since January 2009, the financial resources which are available to a german network operator are reduced by the efficiency requirements of the legally valid incentive regulation. At the same time, share of power system components with increased age - due to the electricity grid expansion in the 60s and 70s - necessitates replacement investments, whereas the cost pressure is additionaly increased. To mitigate these constraints, short term operational costs can be saved depending on the chosen maintenance strategy. The capital costs, however, can only be adjusted in the long term, so the cost development determined by the chosen maintenance strategy of a system operator is of fundamental importance for the development of an optimal asset management strategy. This article discusses techniques for optimizing maintenance and to determine the influence of these measures on the revenue cap development.

Introduction In order to reduce costs, maintenance offers a considerable potential for optimizations. Annually about 1-1.5 percent of the replacement value of the network is invested in maintenance [1]. The saving of costs in the field of operational asset management is often associated with a decreased reliability of the equipment. Therefore to maintain a high level of network performance quality, the influence of maintenance on the frequency of interruptions in the asset strategy has to be observed. In the incentive regulation, the reliability is taken into account by a so called „element of grid-quality“ (Qelement), so that there is an additional link between the maintenance and the revenue cap. For the network operator, this means that by changing the maintenance strategy, the limit of allowed revenues and therefore profit margins can be affected by the costs and resulting quality. Analysis to depict and explain the relationship between altered maintenance strategy and the resulting cost savings are currently conducted at the Institute of High Voltage Technology of RWTH Aachen (IFHT). Inter alia variations in the cycles of maintenance of the strategy and approaches for the textual adapta-

tion of measures are regarded. The influences of the incentive regulation, for example dividing total costs into different components and the planned Q-element (2012), are as well considered. The relevance of the latter one increases by the dynamically rising, partly contradictory demands of customers and regulators for a sustainable high supply quality at low costs. Since onlinemonitoring systems are economically not feasible for many types of equipment, usually time-based maintenance is performed. These recently used and applied maintenance measures with fixed time intervals do not represent the most cost effective option because the uncertainty about the current condition of equipment associated with those can lead to unnecessary or delayed actions. Therefore in this article a new method for status prognoses is presented, which shall enable the transition towards a condition-based maintenance. The cost shares, the resulting revenue cap as well as the Q-element are included as examples. Due to the expected relevance of cost accounting it is presented by the example of a distribution network.

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Challenges in operative asset management due to incentive regulation

55

Institute for High Voltage Technology

Condition assessment and prognosis In the following, the Condition is the degree of ability of each equipment to run the function or functions for which it is created without any major failures. Here, the condition can by defined by the degree of influence on the undisturbed use of the investigated equipment, which shall be quantified using a valuation model. Starting point for structuring the valuation method is the idea that each equipment operates as a multi-stage unit, defined by its functions according to Cigré. In case of an examplarily regarded circuit breaker in this study, these are • opening and closing on command • switching of currents • loss-free transmission • the upholding of the insulation condition and

56

• protection functions [2] Therefore the acquisition and assessment (weighting) of all relevant parameters for these functions of equipment are seen as a measure of the condition. This detection can be implemented by online monitoring systems, which are often already used in Germany for expensive assets such as power transformers. However, to integrate equipments like circuit breakers and disconnectors into the new conditionbased maintenance strategy, a procedure operating on the basis of already available data is required. Starting point for this is in the considered case the information from maintenance. With knowledge about the development of detected minor failures (frequency of occurrence) during maintenances it my be possible to predict the function-relevant parameters. The procedure for condition prediction is shown in a simplified way in the following figure.

A NNUAL R EPORT 2010

fig. 1: Method for a data-based contition prognosis

Based on experience of past operations and the minor failure behavior of known components the time of first maintenance is set. In this first measure, all relevant parameters are checked and the condition (before the measure) is detected. Starting from the results of the parameter verification their development is adapted and with this knowledge, the following maintenance time is set. This is done recursively, as long as the equipment condition does not require a replacement. Using this method, for about 74 percent of the regarded equipment a sufficiently accurate prediction can be reached. A further analysis of this procedure is in progress at IFHT.

Effect of maintenance on the revenue cap To evaluate whether a maintenance strategy not only optimizes resources, but as well represents a profitable procedure in terms of the incentive regulation in Germany and to assess the question of replacement or re-operation, the relation between maintenance and revenue cap has been determined. In order to consider the Q-element in terms of its influence, these studies were

Field of Research: Systems Engineering and Asset Management

fig. 3: Considered grid-components in the asset simulation [2]

fig. 2: Typical failure rates of assets

Based on data taken from a study of the „Forschungsgemeinschaft für elektrische Anlagen und Stromwirtschaft (FGH e.V.)“ [3], a modeling of the failure rates using the model functions displayed in fig.2 was done by regression calculating. Assuming that only the wear-related failures are influenced by the maintenance, only for this exponential function there is a dependency of the exponent c on the maintenance. The adaptation of this parameter to the respective maintenance was done by examining different network operators with their equipment failures and maintenance strategies. The considered equipments can be found in fig. 3. As a result of a reliability assessment, the change of supply interruptions starting from a base scenario and thus the Q-element in its currently covered shape can be determined. The basic scenario consists of a maintenance interval of 10 years and 2 years for inspections respectively together with an average replacement age of the considered components of 40 years.

Institute for High Voltage Technology

conducted using a representative mediumvoltage network. To simulate the impact of maintenance on the failure behavior of equipment, initially the relation outlined in fig. 2 was assumed.

As a part of a complete asset simulation for an urban distribution network operator, then the profit as the difference between revenue cap and costs can be determined. The result for various maintenance and investment scenarios shows fig. 4.

57

fig. 4: Result of the asset simulation for an average grid operator

There are three possible scenarios, which have a nearly equal gain. The actual scenario, a scenario with more maintenance and a scenario with increased age and increased replacement maintenance intervals. However the latter one will have a negative impact on the image of the network operator due to increased equipment failures, which has not been quantified here. The scenario with increased replacement age but increased maintenance has to be regarded as inefficient.

A NNUAL R EPORT 2010

Institute for High Voltage Technology

58

Summary Network operators are currently faced with major tasks that can be mapped in their entirety only by complex procedures. The incentive regulation has to be mentioned as a relevant factor. Asset simplified simulations can help to develop estimations for the development of the company profit. The presented results of such simulations considering the Q-element show, that neither the cost- nor the quality-optimizing strategy leads to the maximum profit. An exception does exist, if network operators would operate their grids at a very low level of quality. As the Q-element is probably limited the additional profits due to high cost savings are in this case only partially compensated. However, additional reductions in profit due to penalties and reputational damage allow the conclusion that minimum quality strategies are not sustainable anyway in reality. In general the conditionbased maintenance strategy is regarded as cost-optimal and therefore in terms of the incentive regulation as efficient. For this purpose a new data-based method was presented that can adequately predict the condition of about 75 percent of the observed circuit breakers over a period of 8 years under the given conditions. The asset simu-

A NNUAL R EPORT 2010

lation as well as the condition forecast are still under investigation at IFHT.

Sources [1] J. Wilckens, Den Herausforderungen der Anreizregulierung mit Lean Methoden begegnen, proLean Consulting AG, Online-artilce for „Bundesverband Deutscher Unternehmensberater BDU e.V.“,2010. [2] CIGRE WG C1.1, Asset management of transmission systems and associated CIGRE activities, 2006. [3] U. Zickler, A. Schnettler, Betriebsmittelstatistiken für das AssetManagement: Möglichkeiten und Nutzen, FGH-Conference „AssetManagement in Verteilungsnetzen; Methoden, Daten, Praxiserfahrungen“, Heidelberg 2006.

Contact Dipl.-Ing. Christian Hille [email protected] +49 241 80-94915

Field of Research: Fault Arcs

Although fault arcs occur seldomly in electrical installations, they do not only put a risk on the operation personnel but also have a serious impact on the installations and the switchgear building by the exhaust of hot gas and the consequent pressure rise. Hence, these consequences have to be considered during the planning phase and the dimensioning of the installation in order to limit the pressure rise to acceptable values, e. g. by construction measures. For this purpose calculation methods are employed to determine the expectable pressure rise in case of a fault. The presented ISC method (Improved Standard Calculation), a development based on the established PRESSURE-software, allows a fast and reliable pressure calculation considering energy absorbers, gas mixtures of air and SF6 , and any number of rooms or volumes.

Institute for High Voltage Technology

Improved Calculation Method for the Pressure Rise due to Fault Arcs

Introduction In case of a fault arc (internal fault) in a medium-voltage switchgear an electric power in the order of several tens of MW is dissipated leading to a temperature rise in the insulation gas and a consequent pressure rise in the arc-affected switchgear compartment. To prevent an uncontrolled burst of the containment, the pressure is released through relief openings or ducts into other volumes of the switchgear or the building, the switchgear is installed in. The energy transported by the gas flow subconsequently leads to a pressure rise in these volumes. The mechanical strength of these rooms against overpressure is limited and only amounts up to 10 mbar for brick walls, for example. Hence, the pressure stress especially of the switchgear building has to be considered carefully in the pressure calculation process [1].

59

Fundamentals of Pressure Calculation

Fig. 1: Basic principle of this calculation method PRESSURE

The software PRESSURE, which is used for pressure calculation so far, is based on the solution of the fluid dynamic equations and the first law of thermodynamics [2]. The basic principle of this calculation method is depicted in Fig. 1.

The pressure rise ∆p in the fault affected compartment (arc room) VAR within a calculation time step ∆t depends on the electric power of the arc Parc , the thermal transfer coefficient kp and especially on the

A NNUAL R EPORT 2010

Institute for High Voltage Technology

60

properties of the insulation gas expressed by its adiabatic index κ: ∆p =

(κ − 1) · kp · Parc · ∆t . VAR

The mass ∆m, exhausted within a calculation time step ∆t through an opening (area A, discharge coefficient α) is determined by the density ρ and the flow velocity v of the gas: ∆m = α · A · ρ · v · ∆t. These equations together with the determination equation for the temperature rise ∆T form a system of differential equations, which is solved iteratetively for all volumes of the considered arrangement using the Runge-Kutta method. This solution process finally yields in a spatially averaged pressure development for the entire switchgear and the building. The quality of the pressure calculation depends on the accuracy of the input data. Especially, the arc voltage Uarc , determining the dissipated arc power Parc , and the temporal variation of the energy transfer coefficient kp as a function of the gas density have to be taken under consideration. Furthermore, a reliable pressure calculation requires accurate gas data [3]. For instance, small deviations of the adibatic index of SF6, which is only marginally greater than one, can already lead to significant errors in the calculated pressure rise.

Functionality of the ISC Method The established PRESSURE-software, which is based on the so-called standard calculation method [2], has been extended and developed at IFHT in the scope of the publicly funded research project Reduction of Pressure Stress in Electrical Installations due to Internal Arcing supported by the German Federation of Industrial Research Associations (AiF). This development mainly includes the integration of ad-

A NNUAL R EPORT 2010

ditional functionalities and leads to the Improved Standard Calculation (ISC) method, which constitutes a considerable progress in comparison with previous calculation methods. Whereas the PRESSURE-software only allows the pressure calculation for standard applications with three compartments and one opening each, neither the number of volumes and openings nor their arrangement is now limited in the ISC method. This allows the calculation even for complex installation layouts, in which several relief openings into different relief rooms are used simultaneously, for example. The integration of arc energy absorbers, used in relief openings and ducts to reduce the energy content of the exhaust gas, is an important issue of the ISC method. Employing a physical model, both the absorption of energy and the flow resistance are implemented [4]. Another feature of the ISC method is the possibility to calculate the pressure rise not only for air-insulated, but also for SF6-insulated switchgear. The calculation method considers the mixing of the insulation gas exhausted from the arc-effected compartment with the ambient air in the other volumes of the switchgear and the building, respectively. Hence, the variation of the gas properties is taken into account. The input data is based on real gas data and the physical gas properties are determined by spline-interpolation from pre-calculated tables. Thus, a high accuracy in the determination of the gas data is achieved, leading to a significant increase of the reliability of the entire pressure calculation process [4]. The ISC-software provides a graphical user interface, which allows a fast and efficient operation of the calculation tool. Due to the realization in the JAVA®-environment, the implementation is independent from the operating system. The input values for the arc voltage and the kp -factor are derived from a broad data

Field of Research: Fault Arcs

Conclusion Having developed the Improved Standard Calculation method , a pressure calculation tool is available, which features by a minimum of restrictive pre-conditions, a high degree of variability, and the use of real gas data. This allows a reliable pressure calculation in a wide range of medium-voltage applications for both air and SF6-insulated switchgear. The ISC-software is ideally suited to eval-

uate other spatially-averaging pressure calculation methods.

References [1] Schumacher M.: Untersuchungen zur Modellierung der Druckbelastung von Schaltanlagengebäuden durch Störlichtbögen, Ph.D.Thesis, RWTH Aachen, 1994 [2] Friberg, G.: Berechnungsverfahren zur Bestimmung der Druckbelastung in elektrischen Anlagen im Störlichtbogenfall, Ph.D.Thesis, RWTH Aachen, 1998 [3] Anantavanich, K., Pietsch G., Eichhoff, D.: Importance of SF6-air gas data for pressure calculation due to fault arcs in electrical installations, XVIII International Conference on Gas Discharges and their Applications, Greifswald, 2010 [4] Anantavanich, K.: Calculation of Pressure Rise in Electrical Installations due to Internal Arcs Considering SF6-Air Mixtures and Arc Energie Absorbers, Ph.D.Thesis, RWTH Aachen, 2010

Contact Dipl.-Ing. Daniel Eichhoff [email protected] +49 241 80-97348

A NNUAL R EPORT 2010

Institute for High Voltage Technology

base, containing reliable values for these figures. By means of a integrated module for the calculation of the electric arc power Parc the required input values are immediately available for the pressure calculation. The possibility to consider the response pressure of relief openings and rupture disks further increases the degree of variability in switchgear layouts. Hence, the ISC method covers a wide range of application and is able to calculate the pressure rise for nearly all commonly installed configurations of medium-voltage switchgear without any restrictions concering special types of switchgear, single manufacturers, or building layouts. As the ISC-software outputs not only the calculated pressure but also further physical values like the gas density or the temperature, the user is able to check the plausibility of the results immediately. In contrast to spatially-resolved calculation methods, which usually require a simulation time of several hours, the pressure calculation of the ISC-software is completed within a short period of time.

61

Institute for High Voltage Technology

Calculation of Pressure Rise in Electrical Installations due to Internal Arcs Considering SF6-Air Mixtures and Arc Energy Absorbers PhD Thesis: M.Sc. Kittipong Anantavanich Date of the oral examination: March 5th, 2010 Reporter:

Univ.-Prof. Dr.rer.nat. Gerhard Pietsch Univ.-Prof. Dr.-Ing. Hans-Jürgen Haubrich

Nowadays, there exist already reliable pressure calculation methods, which are able to determine pressure rise in electrical installations and corresponding buildings due to internal arcs. However, up to now none of them is able to consider SF6 -air mixtures, which are of importance particularly in SF6 -insulated medium voltage switchgear installations. The same is true for arc energy absorbers, which are recently installed with an increasing number. However, their current modelling must be improved and the implementation in pressure calculation methods validated. In this thesis, the aforementioned drawbacks of existent calculation methods are overcome in implementing SF6 -air mixtures and the effects of arc energy absorbers in pressure calculation methods.

Introduction

62

Internal arcs cause sudden temperature and thus pressure increase in electrical installations, which may endanger personnel, electrical equipment, installation rooms or buildings as well as the security of power supply. The proof of internal arc withstand is usually performed by tests in high power laboratories or by pressure calculation especially in cases, where tests are not possible or impractical. Nowadays, there exist reliable pressure calculation methods, which can determine the pressure development during internal arcing. Two kinds of calculation methods have been proved to be necessary for practical applications. The first one is the so-called “CFD calculation method”, which delivers spatially resolved results including all relevant physical effects e.g. pressure waves by solving the fundamental hydrodynamic equations with a Computational Fluid Dynamics (CFD) solver. The second one is the

A NNUAL R EPORT 2010

“standard calculation method”, which provides spatially averaged results rather fast by solving a simplified equation system. The application area of these calculation methods is so far limited. This is especially true for SF6 -insulated switchgear installations with (air-insulated) pressure relief rooms, in which SF6 -air-flows with varying SF6 concentration occur, and for switchgear installations with arc energy absorbers. The objective of this thesis is to include SF6 -air mixtures and arc energy absorbers into both pressure calculation methods for general applications and to further develop the computer program based on the standard calculation method.

Modelling of SF6 -Air Mixtures and Arc Energy Absorbers The focal point of the description of SF6 air mixtures is the generation and treatment of gas data with changing gas composition. For the generation of these gas data, there exist two possibilities, which are on the one

Dissertations

Fig. 1: Effect of absorbers

Arc energy absorbers have influence on the pressure development by two ways, through heat absorption and flow resistance (Fig. 1). In order to desribe these effects, different existing as well as newly introduced model approaches are analysed.

determined by interpolation. For this purpose an optimisation of the interpolation process is carried out to achieve fast convergence with numerous iterations with the CFD calculation method (in short “CFD method”) resulting in reasonable computing times. In order to determine the changing gas compositions in the flow domain, an additional conservation equation for SF6 mass fraction is introduced. For the implementation of absorbers the absorbed heat energy and the pressure loss have to be included into the fundamental hydrodynamic equations. This is done by introducing heat sinks into the conservation equation of energy and friction forces into the conservation equation of momentum. Standard calculation method With the standard calculation method the SF6 mass fraction is directly included into the equation system. Real gas data depending on pressure, temperature and gas composition are also used here. In order to include the effects of absorbers, the grid efficiency is chosen to describe the heat absorption and the flow resistance is considered by a concentrated effective opening. The absorbed heat energy is included into the equation system through heat sink and the flow resistance through the discharge coefficient. Based on the standard calculation method with the newly introduced extensions, a new software package for the calculation of pressure rise due to internal arcs is developed (Improved Standard Calculation (ISC) Method). The program is written in Java and thus platform independent.

Implementation of Gas Mixtures and Arc Energy Absorbers in the CFD and Standard Calculation Method

Validation

CFD calculation method During pressure calculation, gas data of SF6 air mixtures are obtained from the threedimensional table. Intermediate values are

The CFD and ISC methods are validated with the inclusion of SF6 -air mixtures and arc energy absorbers. This is achieved by comparing experimental with calculated pressure and temperature develop-

A NNUAL R EPORT 2010

Institute for High Voltage Technology

hand the application of ideal gas mixing laws with temperature and pressure dependent gas data of the components and on the other hand the minimisation of Gibbs free energy of all components together with the solution of the Boltzmann equation. With the first possibility, interactions including chemical reactions between the gas components are not taken into account. With the help of two examples, it is demonstrated that the consideration of interactions and chemical reactions is necessary. In order to obtain reasonable computing times, the gas data cannot be generated in every iteration step during the solution of the differential equation system. Therefore, they are provided in a three-dimensional table with reference values depending on pressure, temperature and gas composition (SF6 mass fraction).

63

room are depicted. 3.5

switch compartment (ISC method) relief room (ISC method) switch compartment (CFD method)

Pressure (bar)

3.0

relief room (CFD method) switch compartment (measurement) relief room (measurement)

2.5

2.0

1.5

1.0

0

100

200

300

400

Time (ms)

Fig. 3: Measured and calculated pressure developments in the switch compartment and the relief room (air test) 1.04

ISC method CFD method measurement

1.03

Pressure (bar)

Institute for High Voltage Technology

ments, respectively. In total six arrangements are investigated. The dimensions of the six arrangements under investigation cover a wide range of application, which is typically found in medium voltage (MV) switchgear with SF6 -insulated faulty compartments. In the following, a comparison with one of the investigated arrangements is shown as an example. It consists of a compact MV switchgear installation (Fig. 2). A five-layer arc energy absorber is installed between the relief and the transformer room. The switch compartment is filled with air and SF6 , respectively (air test: 12.7 kA, 0.76 s; SF6 test: 16 kA, 1s). The results from the ISC and CFD method are compared with those from measurements. With the CFD method the model approach “in-line tube bundles” is applied for the heat transfer and that of “crossed in-line tube bundles” for the pressure loss.

1.02

1.01

1.00

0

100

200

300

400

Time (ms)

64

Fig. 4: Measured and calculated pressure developments in the transformer room (air test) 1.25

1.20

ISC method CFD method

Pressure (bar)

measurement 1.15

1.10

1.05

Fig. 2: Schematic arrangement of the MV switchgear installation with effective volumes (SC: switch compartment, RR: relief room, TR: transformer room)

In Fig. 3 and 4 the pressure developments in the switch compartment, the relief room, and the transformer room in case of the air test are shown. In Fig. 5 the measured and calculated pressure developments in the relief room in case of the SF6 test without the transformer

A NNUAL R EPORT 2010

1.00

0

200

400

600

800

1000

Time (ms)

Fig. 5: Measured and calculated pressure developments in the relief room without the transformer room (SF6 test)

The results of the validation for the CFD and ISC method can be summarized as follows: CFD method

Dissertations

ISC method In general the ISC method yields reliable results as long as pressure waves and gas flows do not dominate the pressure development.

Conclusion The significant results of this thesis are related to the modelling and inclusion of SF6 air mixtures as well as the inclusion of the effect of arc energy absorbers into the pressure calculation methods (CFD and standard calculation method) for general applications. The main concept of the modelling of SF6 -

air mixtures is the consideration of gas mixture properties with varying mixing ratio during the gas flow of SF6 into surrounding air. For this purpose the gas composition must be determined during the calculation. In order to take into account all important effects, which have influence on the pressure development, real gas data are applied. The inclusion of SF6 -air mixtures into both considered pressure calculation methods leads to reliable calculation results. Arc energy absorbers influence the pressure development through heat absorption and flow resistance. In order to describe both effects, existing approaches are analysed and improved. The inclusion of both effects into the pressure calculation methods is achieved by introducing heat sinks and friction forces into the corresponding conservation equations. Accordingly, reliable results are obtained. Based on the standard calculation method together with the inclusion of mixtures and absorbers, a platform independent computer program is developed, which gives fast results, is simple to handle, and is suitable for a wide range of application.

Contact Dr.-Ing. Kittipong Anantavanich [email protected] +66 2 436 1312

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Regarding SF6 -insulated switchgear, the CFD method delivers reliable results. For a wide range of application, real gas data of mixtures must be applied, which consider interactions between the components of the insulating gas including chemical reactions. In order to describe the heat absorption and the flow resistance of absorbers simultaneously, two model combinations are proposed concerning the applicability. The first combination (grid efficiency approach for the heat absorption and an effective opening for the flow resistance) is suitable, if measuring data are available. If not, a second combination (in-line tube bundles for the heat absorption and crossed in-line tube bundles for the flow resistance) is recommended.

65

Institute for High Voltage Technology

Modeling the type specific failure occurrences of high voltage switching equipment PhD Thesis: Dipl.-Ing. Stefan Federlein Date of oral examination: April, 29th, 2010 Reporter:

Univ.-Prof. Dr.-Ing. Armin Schnettler Univ.-Prof. Dr.-Ing. Gerd Balzer

On the basis of an efficient asset management network operators will be able to optimize their future operation. Therefore, knowledge about age dependant failure occurrences of equipment has to be integrated into the asset management software tools of network operators. Precise predictions for future equipment behavior will help network operators in the decision making process. This thesis deals with models describing the type specific and age dependant failure occurrences in consideration of applied maintenance actions. Using the example of different types of circuit breakers, first the ageing behavior in case of the current maintenance and following in case of modified maintenance scenarios will be modeled and investigated.

Introduction and motivation

66

Increasing cost pressure due to regulatory guidelines, changing topology of electrical power generation as well as ageing of equipment and very volatile load flows are the main challenges for transmission and distribution system operators. Minimizing the total cost and providing a high security of supply at the same time are the major aims of system operators. An efficient asset management will help to achieve these aims. The optimal usage of equipment as well as determining their maximal lifetime are the essential aspects in this optimisation process. One of the basic requirements is a comprehensive knowledge about equipment, their type specific behaviour as well as the impact of maintenance strategies on minor and major failure occurrences. Modeling the type specific failure occurrences of high voltage circuit breakers in consideration of applied maintenance actions is the main objective of this thesis.

A NNUAL R EPORT 2010

Especially, it is focused on the correlation between failures and the specification of maintenance actions. Following the point of view is not a comprehensive network system but rather on an certain high voltage equipment (123 kV and 245 kV outdoor circuit breakers).

Database In cooperation with German network operators a database is established. It includes essential information of outdoor substations e.g. technical data, failures, operating times, maintenance data. Afterwards five circuit breaker types were identified whose available data is adequate to the modeling and the application. In addition numerous maintenance protocols will be captured and evaluated which was not done before. The findings during maintenance actions are now available as a so called bit mask that presents the condition at maintenance time. Following all minor failures are collected in that way.

Dissertations

The basic for the following model is the evaluation of the ageing behavior considering the current applied maintenance actions. Built on empiric knowledge the failure model in case of current maintenance actions will be developed and applied on five representative circuit breaker types. The resultant ageing models show a bath tub curve approach (see figure 3); thereby the specific differences in the behavior of the considered equipment could be pointed out clearly.

ing truncated Weibull distributions the expert knowledge is formulated mathematically. Theses functions are integrated in the model as a time dependant failure probability. The probabilistic progression from a minor to a major failure is attended by a time dependant and not regenerative random process. Hence, the model estimates the failure occurrence after the normal service interval is exceeded is thus apparent as a superposition of the current failure occurrence λ(t) with the additional failures which occur as a result of a damage ∆λ(t). Therefore, for the changed behaviour, the following resultant failure occurrence λ′ (t) is produced by ′

λ (t) = λ(t) + ∆λ(t) = λ(t) +

n X

Institute for High Voltage Technology

Failure occurrences by current maintenance

∆λi (t)

i=1

Fig. 1: Failure occurrences of different circuit breakers by current maintenance

The teething items of the circuit breakers cover the first years up to two till eight years. With the exception of one breaker, all ageing and burn-out phenomena start in average between 10 and 15 operating years. Afterwards failures could increase extremely. The average failure rate is in a range of two to eight failures per 100 circuit breaker years.

Progression of a minor to a major failure occurrence - m2M-Model Based on heuristic knowledge, inspected damages (minor failures) during maintenance actions and current failure occurrences a new model called m2M1 is developed. This m2M model evaluates the impact of deferred maintenance actions on the failure occurrences. In this process, us1

n describes the maximum of different damage types i which can be inspected during maintenance actions. By means of the truncated Weibull distributions deduced from expert knowledge, the probability pi (t) for a failure occurrence at time t can be delivered. In consequence ∆λi (t) is calculated by ∆λi (t) = pi (t) ·

t−1 Y

( 1 − pi (j) ) · hi, IHM

j=1

In this case, the second factor illustrates the probability, that the minor failure has not entailed a major failure in the time period before time t (not regenerative process). The relative minor failure frequency is represented by hi, IHM and weights the failure occurrence accordingly.

Failure occurrences by modified maintenance In this thesis 14 scenarios are considered. The deferred maintenance intervals (in years) always refer to the basic scenario

m2M: minor to Major failure

A NNUAL R EPORT 2010

67

Institute for High Voltage Technology

0/0 with an inspection interval of two years and a service interval of eight years. Consequently, scenario 1/2 illustrates inspection intervals of three years and service intervals of ten years. Figure 2 shows the ageing behavior for three different scenarios. It is obvious, that already a slight extension of maintenance intervals induce a significant increase of failure occurrences (scenario 1/2). Furthermore, a scenario with very long maintenance intervals (e.g. scenario 4/8) procures increasing failure occurrences that are disproportionate higher than the extension of the maintenance intervals. This effect belongs to minor failures with long progression times for m2M, because these minor failures will develop to major failures only in combination with long time intervals. Considering short intervals, these minor failures will not induce major failures and consequently they have no negative effect on the failure occurrences.

tenance effort is reduced by 45 percent.

Fig. 3: Failure occurrences in different maintenance scenarios

68 Fig. 4: Maintenance effort in different scenarios

Fig. 2: Maintenance scenarios for a 123 kV circuit breaker

Figure 3 and figure 4 compare all considered scenarios. The expected changes in failure occurrences as well as in maintenance efforts (= fix costs) are compared to the basic scenario 0/0. Considering the very low failure rates of circuit breakers in any case, the scenario 2/4 seems to be very efficient. On the one hand the failure occurrence increases three times, but on the other hand the main-

A NNUAL R EPORT 2010

The application of the m2M model on three types of circuit breakers presents plausible results and the following consequences can be drawn out. According to the resultant failure rates the extension of maintenance intervals is tend to be defensible. Tough the type specific differences of the circuit breakers have to be considered. In dependence on the technical characteristics already slight extensions of the intervals will induce a very high failure potential. Independent on the circuit breaker types, the efficiency according to deferred maintenance intervals increases easily. An

Dissertations

Furthermore, the analysis of damages shows, that especially the first service action is able to detect many damages, which often refers to teething items. In that case, it is recommended to hold the first service action at the current time (eighth operating year) and to defer the following service cycles only. Hence, the potential of failure occurrences will not exhaust in the first operating years.

Conclusion In this thesis a new model is developed, which describes the progression from a minor to a major failure (m2M). This model predicts the resultant failure occurrences in case of change maintenance actions. The application of the m2M model is reviewed using the example of circuit breakers. Now, the model is available for network operators to assess other equipment in dependence on the available data.

References [1] S. Federlein, C. Hille; C. Neumann, B. Rusek, A. Schnettler: Nutzung historischer Instandhaltungsinformationen zur Modellierung des Betriebsmittelverhaltens. International ETG Congress, FT4 - 5.1, Duesseldorf 2009 [2] S. Federlein, C. Hille, A. Gaul, A. Schnettler: New methods to assess the impact of maintenance and the condition of network. CIRED 2009, Session 1, No. 0870, Prague (Czech Republic) 2009

Institute for High Voltage Technology

inspection can detect only less critical damages and is in consequence less effective.

Contact Dr.-Ing. Stefan Federlein [email protected] +49 (241) 80-94959

A NNUAL R EPORT 2010

69

Institute for High Voltage Technology

70

Elastic Syntactic Foam in High Voltage Applications PhD Thesis: Dr.-Ing. Michael Kessler Date of oral examination: July, 2nd, 2010 Reporter:

Univ.-Prof. Dr.-Ing. Armin Schnettler Univ.-Prof. Dr.-Ing. Volker Hinrichsen

Future insulation systems have to be lighter and more compact than those used today. Nevertheless they must be reliable and feature a long lifetime. Additionally, due to a general increased environmental awareness of the customers, environment-friendly materials are demanded. A non-combustible, solid material is insensitive to leakages in housings, which may occur when gaseous or liquid insulation materials are used and may lead to faults and environmental hazards. Using an elastic and compressible solid material also avoids crack formation, which may occur in conventional solids under thermal load cycles. Elastic syntactic foam is a promising material for this task. This kind of composite material consists of a silicone matrix and gas-filled polymeric microspheres in a range of several 10 µm used as filler. In this work, mechanical and physical properties of elastic syntactic foams depending on their constellation are investigated. The focus is on the electrical properties in the short term range. Moreover, the behavior of this material under electrical field stress is investigated with the aim to develop a model describing the processes during an electrical breakdown for AC and DC voltages. The investigations show that discharge structures generally are located in the silicone elastomer. Hence the breakdown process for AC and DC voltage in elastic syntactic foam is an electrical breakdown inside the silicone matrix.

Introduction and motivation There is a variety of demands on modern insulation systems for high voltage applications. Not only electrical ageing but also thermal stress is a challenge for materials. As a result of changing weather conditions or fluctuating load, thermal load cycles lead to mechanical stress inside the insulation system. Crack building inside the housing or ablation of the insulating material leading to a breakdow and outage may occur. Gaseous insulation materials like Sulfurhexafluorite (SF6) or liquid materials (e.g. transformer oil) are commonly used in high or extra high voltage applications. The requirement to avoid leakages is a serious disadvantage, since leaking of the in-

A NNUAL R EPORT 2010

sulation material is harmful to the environment and leads to a reduction of the dielectric strength with a breakdown as last consequence. A solid material avoids the disadvantages mentioned above (a so called “dry” Insulation material [1]). However, high temperature gradients are a danger to solid materials. A possible solution for this task is an insulation material filled into the component in the liquid state curing there to solid material. A high elasticity, compressibility and a strong adhesivity to other solids make a material able to withstand thermal load cycles. Such a solid material is called elastic syntactic foam. This material consists of a silicone gel matrix with a low viscosity and polymeric microspheres in a dimension of some 10 µm used

Dissertations

Fig. 2: Dielectric strength of different elastic syntactic foams depending on the filling degree

Fig. 1: ESEM picture of an elastic syntactic foam [2]

New application fields for elastic syntactic foams are ignition systems. Due to the strong temperature gradients in the engine compartment during motor start or as a result of splash water contact during drive in winter times, there is a serious temperature stress for components. As a consequence of the thermal stress crack building may occur in traditional epoxy-based insualtion systems.

Since the intervals are overlapping, no binding statement can be made. There is also broad scattering for 20 Vol.-% filling degree under AC stress. The results for all elastic syntactic foams are around 25 % lower in comparison with pure silicone gel (0 Vol.-% in Fig. 2). The average dielectric strength of pure silicone gel under AC stress is 71.4 kV/mm (peak value). A similar trend is found for DC investigation. There is a slight increase of the average dielectric strength with increasing filling degree. The influence of the inorganic coating on the dielectric strength is shown in Fig. 3.

Short time dielectric strength One focus of this work is on the investigation of the dielectric strength of elastic syntactic foam under AC and DC stress depending on several parameters. A sphere to sphere arrangement of electrodes is used for the investigations (diameter of spheres 12 mm, sparking distance 3 mm). The evaluation is done using the 95 % confidence interval. Peak values are given for AC results. If calcium carbonate coated microspheres are used as filler, a slight increase of the dielectric strength with increasing filling degree can be found for AC (Fig. 2).

Fig. 3: Dielectric strength of elastic syntactic foams depending on coating

With calcium carbonate coated microspheres (average diameter 90 µm) and uncoated microspheres (ca. 80 µm) are considered with 40 Vol.% filling degree respectively. The average dielectric strength of syntactic foams filled with coated spheres

A NNUAL R EPORT 2010

Institute for High Voltage Technology

as filler. Fig. 1 shows an ESEM (Environmental Scanning Electron Microscopy) picture of elastic syntactic foam. This material combines the excellent adhesion of the silicone gel to other polymeric and metallic surfaces with the compressibility of the gas filled microspheres.

71

Institute for High Voltage Technology

72

is 54.3 kV/mm for AC. Hence, it is 76 % higher than the dielectric strength for syntactic foams filled with uncoated microspheres. For DC applications, the usage of coated microspheres leads to 93 % higher average dielectric strength in comparison to uncoated microspheres.

Investigations on well defined microsphere arrangements Optical investigations form another important backing for the development of a breakdown model for DC and AC voltage. These investigations make the detection of origin and development of discharge channel possible. Samples of coated microspheres in pure transparent silicone gel are considered. To make the production process easier big microspheres with an average diameter of 200 µm are used to build up the structures. Fig. 4 illustrates an example of a microsphere row inside a tip to tip electrode arrangement. Fig. 5 shows the breakdown channel. To create the channel a step test is used (start voltage 15 kV, step 2 kV/min). The results show that all breakdown channels proceed around the microspheres. Although the microspheres are centered between the tip electrodes and hence on the axis of highest field strength, the discharge channel does not proceed through the microspheres but moves them aside. Further optical investigation in the short term range show that the breakdown channel develops in the silicone gel matrix independent from the degree of homogeneity of the electrode arrangement.

Fig. 4: Microcsphere arrangement in silicone gel

A NNUAL R EPORT 2010

Fig. 5: Microsphere arrangement after breakdown

For the determination of the field distribution inside well defined microsphere arrangements, simulations are performed. A three-dimensional FEM (finite elements method) tool is applied. This tool works with an automatic, adaptive mesh method using tetrahedral elements. The simulations are done with background field strength of 20 kV/mm, since this value is lower than the dielectric strength of pure silicone gel according the data specification. The field distribution in the plain of central points of an microsphere matrix is shown as cut view in Fig. 6. The background ~ a, ∞ is vectored from left to right. field E

Fig. 6: Simulation of the field distribution in a microsphere matrix

The highest field strength occurs at the outermost microspheres of a row (25 kV/mm). However, the field strength decreases locally to 15 kV/mm between two microspheres. Looking at a microsphere row vertically to background field shows that there is a zone of reduced field strength (10 kV/mm) left and right of the microsphere row. But the field strength is strongly enhanced between the micro-

Dissertations

Breakdown model Due to the presence of microspheres field displacements occurs inside the silicone gel matrix leading to local field enhancements. These enhancements can activate local discharges in the silicone gel. The discharge process itself is known from literature: Electrons accelerated in the gas channel cause chain scission of the silicone [3]. If the discharge impacts on a mineral coated microsphere, the discharge can be hampered. Using a high filling degree of microspheres helps to avoid the expansion of the discharge channel as a result of the pressure rise inside the gas channel and the elasticity of the silicone gel matrix. Is the discharge process able to continue around the microsphere and bridge the complete insulation distance, the high short circuit current destroys the thermoplastic sphere shells thermally. Since the investigation deliver comparable results, the breakdown model is valuable for DC and AC voltage.

References [1] Eitle, R.; Kaumanns, J.: Trockene Freiluftendverschlüsse mit Stützeigenschaft, Fachbeitrag aus Elektrizitätswirtschaft; Heft 11.2000 [2] Keßler, M.: Einsatz elastischer syntaktischer Schäume in der Hochspannungstechnik, Aachener Beiträge zur HOCHSPANNUNGSTECHNIK Band 16, Verlagshaus Mainz GmbH, 2010 [3] Oesterheld, J.: Dielektrisches Verhalten von SilikonelastomerIsolierungen bei hohen elektrischen Feldstärken,Fortschr.Ber.VDI Reihe 21 Nr.196, Düsseldorf VDI Verlag, 1996

Institute for High Voltage Technology

spheres. Thus the dielectric strength of the silicone gel can be exceeded.

Contact Dr.-Ing. Michael Kessler [email protected] +49 241 80-94936

A NNUAL R EPORT 2010

73

Institute for High Voltage Technology

74

Resistance Distribution in Switching Arcs of Self-Blast Circuit Breakers at the Current Zero Phase Dissertation: Dr.-Ing. Ming Chark Tang, MBA Date of oral examination: 5th March, 2010 Advisor:

Univ.-Prof. Dr.-Ing. Armin Schnettler Univ.-Prof. Dr.-Ing. Michael Kurrat

Self-blast circuit breakers are widely used as switching elements in electrical power systems. In case of a short-circuit fault the breaker contacts are separated leading to an ignition of an electric arc. Hence, the current flow is carried by the electric arc itself. Due to a cooling of the arc at the current zero phase, the arc can be successfully extinguished. The switching gap is converted from a conductive state to an insulating one. In this thesis a measuring arrangement is developed to determine the partial resistance distribution of the switching arc during the current zero phase. This arrangement is adapted to a selfblast circuit breaker model. The physical cooling mechanisms can be evaluated by using an Computational Fluid Dynamics simulation of this process. Detailed knowledge of the resistance distribution leads to a better understanding of the processes during the current zero phase with filling gases such as SF6 and CO2. This enables further optimisation of gas circuit breakers.

Resistance Development in Circuit Breakers At the current zero phase the current is decreasing towards the current zero crossing. Thus, the ohmic heating power is decreasing at this phase. In order to achieve a successful interruption the electric arc inside a circuit breaker has to be quenched, which starts some microseconds before the actual current zero crossing (see Figure 1). During this time period the cooling power of the arc inside a circuit breaker exceeds the ohmic input power by the current flow. The energy of the electric arc is decreased and the resistance of the arc is increasing. At the postarc current phase (t > 0) the transient recovery voltage is generating a current flow inside the switching arc. Due to the net loss of energy the resistance of the electric arc is

A NNUAL R EPORT 2010

still increasing shortly after the current zero crossing. If the ohmic heating power of the electric arc exceeds the cooling power the resistance of the switching arc will reach a maximum. Due to further re-heating of the electric arc the resistance is decreasing to a conductive state. Hence, the arc quenching process is not successful. In the opposite case the cooling power remain higher than the ohmic heating power during all the time after the current zero crossing. The resistance of the switching arc will increase to a non-conductive state. Hence, the switching arc is quenched and the switching process of the circuit breaker is successful.

Dissertations

Fig. 1: Development of the current, the voltage and the resistance at the current zero phase and at the post-arc current phase [1]

The resistance development during the current zero phase is understood as a spatial integrated value, which is further used as an evaluation criteria for circuit breakers. But no information is known about the spatial distribution of the resistance within the switching arc. Further, the relevant physical phenomena of this process are not quite clear. A detailed knowledge of the resistance distribution and the cooling mechanisms lead to a better physical understanding of the switching process. Hence, a more goal-orientated analysis of gas-blast circuit breakers with various filling gases can be performed and further design and performance optimizations can be carried out.

Concept of Investigation The concept of investigation is illustrated in Figure 2. The analysis of the resistance development and distribution and the correlating cooling mechanisms are the main goal of this investigation. Therefore, an appropriate test arrangement is developed to evaluate the main effects. Influences by other parameters are minimized by the design. The test arrangement is implemented in an experimental circuit breaker model and a simulation model. The experimental results are used to verify the simulation

Fig. 2: Schematic Diagram of the investigation concept

For the experiments a self-blast circuit breaker model is used with common cylindrical nozzle shapes. Within these nozzles a measuring arrangement is integrated to measure the partial resistances at discrete areas of the switching arc. The specific areas represent relevant areas of specific physical processes. The measuring arrangement is investigated to have a negligible influence on the basic processes at the current zero phase of a switching operation. A special measuring equipment is utilized to have a high time resolution of the measuring signals during the current zero phase. The simulations are performed with a CFD (Computational Fluid Dynamics) software. The main equations of a CFD solver are based on the laws of mass conservation (Eq. (1)), of momentum conservation (Eq. (2) (4)) and of energy conservation (Eq. (5)):

A NNUAL R EPORT 2010

Institute for High Voltage Technology

model. The simulated results are used to determine physical parameters and processes which are not accessible to measurements. The combined results of both approaches are evaluated to one consistent understanding. The investigations are performed with the filling gas SF6 on the one hand and CO2 on the other hand. The differences of the switching behaviour of these two gases are evaluated as well.

75

Institute for High Voltage Technology

Results of the Investigations and Analysis of the Switching Behaviour

Additionally, calculation models for turbulent effects and for radiation phenomena are taken into account as well as the equation for the electrical field. The test arrangement is implemented into the simulation software. The simulation model proves to be able to simulate the physical effects occurring during a current zero phase.

The simulated results of the SF6 circuit breaker model and the CO2 one are presented in Figure 4 for the time 200 ns before the current zero crossing for the same simulation conditions. On the left side of the dashed line the results refer to the SF6 circuit breaker model whereas on the right hand side the results refer to the CO2 circuit breaker model. Around the stagnation area the temperature distribution of both simulations are very similar. Especially for the CO2 circuit breaker model a radial widening of the temperature distribution can be observed within the nozzle. Thus, the resistance in the nozzle is smaller than that at the area around the stagnation point. Contrary, a higher resistance distribution can be determined in the nozzle of the SF6 circuit breaker model.

76 Fig. 3: Illustration of the arcing area and sectionalizing into discrete resistance areas

The switching area of the self-blast circuit breaker model is illustrated in Figure 3. It is a static and rotational symmetric arrangement. On the left and on the right hand side the electrodes are placed. Between these electrodes is the arcing area. In the middle of the arrangement the flow channel of the quenching gas is indicated. The quenching gas consists mainly of the filling gas of the circuit breaker. The total resistance of the switching arc is given by Rges. The total resistance consists of the partial resistances RDiff (resistance of the diffuser area), Ra and Rb (resistance of two separate parts of the nozzles) and RSP (resistance at the stagnation point area).

A NNUAL R EPORT 2010

Fig. 4: Comparison of the temperature distribution and the resistance per length of a SF6 circuit breaker and of a CO2 one

The experimental results are interpreted in conjunction with the physical analysis of the simulations. For the partial resistances and the main cooling mechanisms (by convective cooling Pconv on the one hand and by turbulent cooling Pturb on the other hand) the following conclusions can be made for the SF6 circuit breaker model:

Dissertations

Using SF6 as filling and quenching gas in self-blast circuit breakers the resistance at the stagnation point is approximately as high as the results in the nozzles. Thus, the total resistance is determined by these two areas. The main cooling mechanism at the stagnation point is dominated by convective cooling effects. In the nozzles the turbulent cooling is the main cooling mechanism. Hence, both cooling mechanism are important for the generation of the total resistance. If CO2 is used as the filling gas in the circuit breaker model the resistance is mainly generated in the area of the stagnation point. Thus, the convective cooling is the dominant cooling mechanism for CO2 circuit breakers. Nevertheless, the total resistance of the CO2 circuit breaker model is lower than the value of the SF6 one. Hence, the switching performance of SF6 gas is better than of CO2. A more detailed presentation and discussion of the investigations are described in [2].

Summary The thesis gives a more detailed insight of the physical processes during the current zero phases of gas filled circuit breakers. Further, the analysis of the spatial resolved resistance distribution gives a better understanding of the resistance development and accordingly the dominant cooling

Institute for High Voltage Technology

Using the CO2 circuit breaker model the following relationships can be determined:

effects for the arc quenching. Depending on the switching arc area two different cooling mechanisms can be effective. The convective cooling occurs mainly at the area around the stagnation point whereas the turbulent cooling is the main effect within the nozzles. The intensity of cooling and thus the amount of the resistance development is depending on the quenching gas of the circuit breaker. In SF6 circuit breakers both cooling mechanism are relevant to generate an arc resistance. For a CO2 circuit breaker the convective cooling is the dominant effect. Hence, design and optimization criteria need to be developed for the two gases individually.

References [1] C. Kahlen, R. Dommerque, M. Schwinne, M.-C. Tang, A. Schnettler: Application of a New Measuring System for Investigations of the Interruption Behaviour of a SF6-Selfblast Circuit Breaker Model, XVIth Symposium on Physics of Switching Arc, Brno, 2005 [2] M.-C. Tang: Widerstandsverteilung in Schaltlichtbögen von Selbstblasleistungsschaltern während der Stromnulldurchgangsphase, Dissertation RWTH Aachen, Schriftenreihe: Aachener Beiträge zur Hochspannungstechnik, Band 13, ISBN 3-86130-676-X, 2010

Contact Dr. Ming Chark Tang [email protected] +49 221 676-2003

A NNUAL R EPORT 2010

77

Institute for High Voltage Technology

78

HumTec Modelling of the future energy supply structure HumTec is an interdisciplinary project house of the RWTH Aachen University, which is built up in the frame of the German excellence initiative. The project house has the aim to advance the interdisciplinary work between the humanities, natural- and engineering science. The Institute for high voltage technology (IFHT) provides together with the Thermal Process Engineering the engineering competences in our work group Ethics for Energy Technology (EET). The models are developed with computer scientists and mechanical engineers.

Description of the project The main focus of the project is the development of an integrated assessment model, which contains different modules. The model enfolds a global point of view and has its final date in the year 2100. This model is able to built up connections between the development of the population, consumption of resources, energy supply, use of land areas, climate and the economy, whereby every research group has its own part and focus. One aim is to analyse the political influence on the different model parts. With the help of the model various scenarios for the future energy mix are calculated. The paths are based and optimized on the present data and prognoses for costs. In addition there are limits for technologies and other constraints. As consequence of the increasing world population, higher living standards and the shortage of fossil fuels questions of the future energy supply and demand become more important in social and political decisions. For this reason the energy model considers especially the development of the markets for fossil fuels. The global resource model should evaluate the production paths and price evolution for crude oil, natural gas and coal. The allocation of the resources is optimized with economical criteria, that means that the total cost for the extraction of each resource are minimized. The distribution of the resources is consid-

A NNUAL R EPORT 2010

ered on a global level, whereas the world is divided into six world regions, so called agents. The agents represent and interact not only as the supplier but also as the consumer for each region on this macro level. Before starting modelling the most relevant influence factors are filtered and analysed. This includes the possibility of substitution between the fossil and the reneweable fuels and the transformation and integration of technological progress into the model. Because of the shortage of fossil fuels the reneweable energies, such as wind, solar and biomass energy, will become more and more important. The final time of modelling in the year 2100 underlines this aspect. The consideration of social and political factors influences more and more the development in the frame of the energy supply. For this reason these aspects have to be examined and taken into account. Exemplary are shown two research activities. On the one side „DESERTEC“, in which the electricity of north african solar energy plants should support the energy supply of Europe and other side the new technology „Carbon Capture and Storage“ (CCS). This technology is able to capture the carbon dioxide from the power generation process. After that the captured carbon dioxid is transported to a suitable storage. Technical solutions need connection and interaction of social questions to the known economical and technical ones, what is especially

HumTec of the exported electricity and develop their countries and so the whole african continent. Despite to all aspects the danger of Europe’s reliance on North African solar electricity, compared with one, as currently from the oil-producing countries, has to be excluded. For this reason, in a holistic assessment, the different open questions regarding energy security, socio-economic development and international cooperation have to be clarified to ensure the long-term success of the project „DESERTEC“. Another development, which is a typical example for the interface of the work of EET, is CCS. Besides the necessary economical and technical considerations of the technology there are also expected problems relating its acceptance and penetration. The mass media and the people need answers for the open questions especially for the storage of the carbon dioxide, so that the power supply companies first have to convince the people and need the legal conditions for the realization of CCS. These examples impressively show that the future energy mix is not only influenced by economical, ecological and technical criteria, but also by social and political ones.

References [1] DLR, Strom aus der Wüste, Deutsches Zentrum für Luft- und Raumfahrt, 2010

Contact Dipl.-Wirt.-Ing. Baris Özalay [email protected] +49 241 80-93041

A NNUAL R EPORT 2010

Institute for High Voltage Technology

shown by the open and discussed questions in the project „DESERTEC“. This project has the aim to realize the high potential of solar energy in the region of north africa, to meet the increasing electricity demand of Europe. The electricity transport to europe should be realized by an HVDC system and produce 15 percent of the european electricity in 2050 [1]. There are basically four different possibilities to install electrical lines from North Africa to Europe. One way leads from morocco through the street of Gibraltar to Spain, a second one goes from Tunisia through Sicilia to Italy. The third possibility is the long sea way from Libya to Greece and the last one is described by the terrain connection about the Middle East and Turkey . A single two-point HVDC link from Egyptian up to the load center in Central Europe would cover a distance of about 4000 km and that is why it will not be the first route to be built. The connection to Greece is also unlikely as the first line path, because the sea connection is with 500 km very long. Realistically, HVDC links with multiple entry points on the way from North Africa to the load centers will be in central Europe, especially Germany and France. Thus, the electrical energy on their way to the heart of Europe can be distributed from various points. A country would not approve the building of a high-power transmission line without it even to be connected and participate with the use of the electrical energy. As connection points are load centers suitable, primarily nodes near large cities. Next to such technical and economical considerations also social factors influence the development of the project. The affected African countries could benefit from the revenues

79

80

Institute for High Voltage Technology

A NNUAL R EPORT 2010

Teaching at IFHT

A NNUAL R EPORT 2010

81

Institute for High Voltage Technology

Institute for High Voltage Technology

82

Introduction of Bachelor and Courses in Electrical Engineering

Master-

With the winter term 2007/2008 the Bachelor of Science (B.Sc.) in electrical engineering was introduced at RWTH Aachen university and replaced the former Diploma-degree. Winter term 2010/2011 brought the introduction of the consecutive Master of Science (M.Sc.). Together with this changes, content and structures of the study programmes were reevaluated and improved. There also have been changes for the IFHT.

New Lectures and changed contents Due to the change to the new Bachelor and Master degrees an adaptation of content and structure of the teaching in all institutes of the faculty was necessary. Now all courses are modularized and listed in a module handbook. For the specialization of power engineering in the Master degree the IFHT coordinated the compilation of this handbook. Moreover, the range of courses tought at the IFHT was adapted as shown in the following. Components and Installations for the Supply of Electricity The lecture Components and Installations for the Supply of Electricity is compulsory for 5th semester bachelor students specializing in power engineering. It was introduced in the winter term 2009/2010. Prof. Schnettler and Prof. Hameyer from the Institute for Electrical Machines (IEM) teach the lecture together. With the topics wind, hydro and steam power-plants as well as generators and transformers Prof. Hameyer is focussing on the generation of electricty. Prof. Schnettler is focussing on the components of the grid, such as overhead lines, switchgear, surge arresters, measurement devices and test circuits. For the compilation of the lecture (including script and powerpoint slides) both institutes were granted funds from study fees. With over 100 students the lecture was very well attended which shows that among the students the power engineering specializa-

A NNUAL R EPORT 2010

tion is very popular. High Voltage Engineering In the diploma study programme the lecture High Voltage Engineering was divided into two consecutive parts. This had to be changed for the Master study programme as the students have, due to practical semester and thesis semester, only two semesters left for attending lectures. Therefore, greater flexibility had to be made possible. To achieve this, content and title of the two lectures have been adapted in a way that enables the students to visit the lectures High Voltage Engineering - Insulation Systems and High Voltage Engineering - Test Systems and Diagnosis and take the respective exams independently. Until now, the content of the latter has been reevalued and changed and is presented with a completely new set of powerpoint slides. For this adaption the institute was granted fund from study fees as well.

New Seminars and Projects Also the range of projects and seminars offered at IFHT has been adapted to the new courses’ requirements. Seminar Energy Supply in the Future The former seminar Technology of Gas Discharges has been replaced by the new seminar Energy Supply in the Future. Participants have to present topics on relevant future developments in Power Engineering. The seminar is open to Bachelor and Mas-

Teaching their supervising tutors for help and advice.

Fig. 1: The fastest car

Highlight of the project was a time trial in which the cars of the student groups raced each other and the car of the supervisors. The cars had to fulfill all safety requirements and drive the race course at least two consecutive times. Picture 1 shows the winning team’s car. The participants enjoyed the project and it is planned to repeat it in the coming years.

Institute for High Voltage Technology

ter students. Bachelor students are awarded 3 CP for participation and a ten minutes presentation. Master students have the opportunity to gain 4 CP by handing in an additional three to five page paper. 4th Semester Project In summer term 2010 the faculty’s institutes for power engineering (IFHT, ISEA, IAEW, IEM, ACS) together offered a project for 4th semester students. In groups of three the participants built a roadworthy electric model car. To achieve this they had to learn about batteries, DC-motors for the drive, DC-DC-converters, servomotors for the steering and the required measurement and control systems. As the car had to drive a racetrack independently a special attention had to be given to the control and protection to avoid damage and provide security. The basis was a construction kit that was the same for all student groups. The students had to design and built the circuit boards, assemble the cars and program the control software. In five units they were given the necessary background information on track and speed sensors, DC-DC-converter, motors and programming of the micro controller. Moreover, they always could ask

83

Contact Dipl.-Gwl. Maximilian Keller [email protected] +49 241 80-94924

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Teaching at IFHT High Voltage Engineering Insulation Systems (Summer term)

Components and Installations for the Supply of Electricity (Winter Term)

Lecturer: Prof. Dr.-Ing. A. Schnettler L2/E1, from 6th semester, International Master Program: 2nd semester German and English

Lecturer:

Content • Overvoltages in power systems: Travelling waves, overvoltage protection • Breakdown-phenomena: Statistics, breakdown in gases, solids and liquids, vacuum breakdown, arcing, pollution flashover

84

Prof. Dr.-Ing. A. Schnettler Prof. Dr.-Ing. habil. Dr. h.c. K. Hameyer L2/E1, from 5th Semester Bachelor German only

Content This lecture is offered in collaboration with the Institute for Electrical Machines (IEM). • Introduction to electrical power engineering

• Insulation systems and dielectrics

• Electric field

• Technical excursion

• Gas discharge technology

High Voltage Engineering Test Systems and Diagnosis (Winter Term) Lecturer: Prof. Dr.-Ing. A. Schnettler L2/E1, from 7th semester, International Master Program: 3rd semester German and English

Content • Generation and measurement of high AC, DC and impulse voltages • Dividers and Shunts • High voltage equipment: Construction, capacitors, bushing, transformers, breakers • Electromagnetic compatibility (EMC) for power systems

A NNUAL R EPORT 2010

• Introduction to the most important components for the supply of electricity • Testing of equipment

Protective Measures and Equipment in Power Supply Systems and Electrical Installations Lecturer: Dr.-Ing. B. Thies L2/E1, Master program in Electrical Power Engineering German and English

Content • Regulations and standards, national, regional and world-wide safety regulations and standards, legal authority of standardisation, test certificates

Teaching

Power Cable (Winter Term)

• Protective measures of low voltage installations against direct and indirect contact, configurations of the network, safety, protectional and functional extra low voltage, evaluation

Content

Engineering

Lecturer: Dr.-Ing. D. Meurer L2/E1, Master program in Electrical Power Engineering English only

• Introduction: cable industry, cable market, power cables in networks, history, technical standards, transmission properties • Cable Components: materials, conductor, insulation, screen, sheath, armour

• Protective measures of high voltage installations, earthing, overvoltage and lightning protection

• Protection of lines and cables against overload and short circuits

• Design: low-/medium-/high-voltage cables, utility/special cables • Production: paper insulated/extruded cables, conductor/core/cable • Quality Management: ISO 9000, quality assurance, type-/sample/routine-test, commissioning test, ageing, lifetime • Accessories: termination, joints, field control, installation techniques

• Protective equipments and their modes of action, protection criteria, fuses, line protective switches and current-operated earth-leakage circuit breakers, relays, surge arresters

• Protection systems for motors, generators and transformers

Institute for High Voltage Technology

• Risk of electric currents, accident statistics, effects of current on human body, safety limits, endangering by high-frequency fields

• Cable Projects: cable route, current carrying capacity, transport, laying • High Power Cables: cable losses, forced cooling, HVDC, gasinsulated cables, cryogenic cables, superconducting cables (LTSC, HTSC)

Seminar Switchgear and Substations (Summer/Winter Term) Lecturer (Summer term): Prof. Dr.-Ing. A. Schnettler, Dr.-Ing. M. Schumacher (ABB AG), together with partners from the industry

A NNUAL R EPORT 2010

85

Institute for High Voltage Technology

Lecturer (Winter term): Prof. Dr.-Ing. A. Schnettler, Dr.-Ing. R. Puffer 5th Semester Bachelor German only

• Grids and components • Overland lines

Content • Switching arcs

• Components: masts, armatures etc • Construction

• SF6 circuit breaker

• Operation

• Vacuum circuit breaker

• Maintenance

• High voltage fuses

• Conductors

• HVDC • Air insulated substations • Gas insulated substations • Medium voltage switchgear and substations • Power transformers • Cables and overhead lines

86

Contents

• Asset management in distribution grids

Business Managment for Engineers (Summer Term) Lecturer: Dr.-Ing. J. Schneider L2, E1 German only

Content • Company structures • Business administration • Finnace

Compare report on page 107

• General Management: Strategic planning, budget, controlling

Overhead lines (Winter term)

• Project management: Risk analysis, cost planning, marketing, quality management, Corporate Governance

Lecturer: Dr.-Ing. R. Puffer L2, E1, ab 5. Semester German only

A NNUAL R EPORT 2010

• Optimization

Teaching

Laboratory in Electrical Power Engineering I (Summer Term) TE4, from 6th Semester German and English

Electrotechnical Laboratory I (Summer term) TE3, from 2nd Semester German only

Content Content The laboratory in Electrical Power Engineering I is carried out by several institutes. Three experiments take place at the Institute for High Voltage Engineering:

The Lab contains 8 basic units: • Current- und voltage sources • Voltage divider • Measurement in linear circuits

• Measurement of alternating voltages / Breakdown measurements

• Simulation of linear circuits

• Generation and measurement of direct voltages

• OP Amp (Measurement)

• Surge voltages

• Measurement of non-elctrical phenomena

Laboratory in Electrical Power Engineering II (Winter Term) TE4, from 7th Semester German and English

• Diode und transistor

• OP Amp (Simulation)

High Voltage Laboratory TE4, from 7th Semester German only

Content • Breakdown measurements

Content The laboratory in Electrical Power Engineering II is carried out by several institutes. Three experiments take place at the Institute for High Voltage Engineering • High Voltage Divider

Institute for High Voltage Technology

Laboratory Courses at IFHT

• Experiment on DC voltage • Partial discharge • EMC Interference Launch • Surge voltages • Transient processes on wires

• Transient Processes on Wires, Travelling Waves

• High voltage divider

• Electromagnetic Coupling

• Dimensioning of a synthetic test circuit

A NNUAL R EPORT 2010

87

Institute for High Voltage Technology

Measurement Laboratory for Mechanical Engineers (SS) 4th

TE4, from Semester Collaboration of the faculties for electrical engineering, mechanical engineering and science and methmatics German only

88

A NNUAL R EPORT 2010

Content The IFHT offers 4 labs in this course: 1. Voltage source and transformer 2. Measurement of electrical phenomena 3. Oscilloscope and rectifiers 4. Digital measurements

Theses

Diploma Theses Altwasser, Dominik Entwicklung optimaler Managementstrategien für Elektrofahrzeuge Both the continuing climate change and increasing instability of commodity markets have ignited a social and political debate over Germany’s future mobility strategy, of which one of the most widely and publicly recognized core elements is electromobility. At this juncture, the integration of electric vehicles into the existing transmission and distribution network constitutes a focal point of recent research in this area. Part of this interest may arise from the ability of electric vehicles to store energy temporarily before it is used later to generate mechanical operating energy, allowing for the flexible control of charging and discharging. In this thesis, a control procedure based on evolutionary algorithms was developed with the aim of exploring the possibilities of flexible charging and discharging. Using these algorithms, both the temporal and spatial arrangement of the charging processes can be selectively modified according to various criteria in a real network model. Possible applications of this procedure include avoiding overloading the existing network infrastructure as well as charging the observed electric vehicles at the lowest possible cost. The functionality of the devised procedure was verified in the course of this thesis by applying it to a developed scenario; as expected, the algorithm arranged the charging processes corresponding to the specified goal function. Anomalies that were observed, especially relating to the spatial component of the op-

timization procedure, could be ascribed to the nature of the vehicle population under consideration. Finally, consequent analyses showed the potential of evolutionary algorithms for additional uses, such as multicriteria optimizations and the application of probabilistic dependent and target variables. Betreuer: Thomas Pollok, M.Sc., Dipl.-Ing. Thomas Helmschrott

Institute for High Voltage Technology

Diploma, Master, Bachelor and Study Theses completed in 2010

Beilin, Michael Investigation on the Impact of Nanofiller on the Properties of Syntactic Foam Syntactic foam is a composite material consisting of hollow glass microspheres and epoxy resin, which is used as a lightweight insulation material for high voltage applications. The viscosity of epoxy resin is increased by inserting microspheres. This results in a maximum filling degree, which is currently reached at 55 vol.-%. The viscosity can be lowered by adding nanoscaled particles - so called nanofillers - to the composite. The focus of this diploma thesis lays on the experimental and theoretical investigation of the thermal, electrical and mechanical properties of syntactic foam with nanofiller. Based on the gained knowledge different models are created to explain the impact of nanofiller on the properties of syntactic foam. As main result of this thesis an improvement of the electrical and thermal properties by the influence of nanofillers can be ascertained. On the other hand the mechanical properties are slightly declined. Supervisor: Dipl.-Ing. Anja Strauchs, M.Sc. Andrey Mashkin

A NNUAL R EPORT 2010

89

Institute for High Voltage Technology

90

Doroudian, Sepideh Szenarioanalyse zur Bestimmung des Einflusses fluktuierender Einspeisung auf die deutschen Stromnetze This thesis analyses the impact of the integration of decentralised energy conversion technologies and the subsequent fluctuation of load and supply on low voltage power grids. Due to political objectives and a growing ecologic consciousness of society an increasing level of the use of decentralised energy conversion technologies grids is expected in the incoming years. In order to examine the Impact of these Technologies on the low voltage power grids at first a simulation model is developed. This model simulates the stochastic behaviour of single decentralised energy conversion technologies. Using this model, the potential for each of the individual conversion technologies is determined. This is done both for the present state and - by means of a scenario analysis - for possible future developments until the year 2040. Subsequently, the specific costs and emissions as well as the self-sufficiency of the grid area are calculated for selected scenarios. The results of these calculations make it possible to draw conclusions on the optimal network operation and on a necessary future development of the portfolio of power plants. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Wirt.-Ing. Eva Szczechowicz Dulisch, Fabian Conception, enhancement and implementation of a test circuit to investigate contact materials in vacuum switching chambers During the last 30 years the vacuum switching technology was established for low and medium voltage applications up to 40.5 kV with increasing market share asserting its position opposite to other switching tech-

A NNUAL R EPORT 2010

nologies as minimum-oil or SF6 circuit breakers. Due to application of modern contact materials and contact geometries vacuum switching chambers are suitable for almost all switching operations. The durability of vacuum switching chambers is significantly determined by the used contact material. Therefore in the context of this diploma thesis a test circuit, which emulates duty with a high number of switching operations, is conceived, enhanced and operated. Furthermore different contact materials are investigated. Supervisor: Dipl.-Ing. Andreas Kurz, Dr. D. Gentsch Filippidis, Stylianos Condition detection and prognosis of equipments inside air-insulated substations The maintenance of the components in the electrical network has a direct influence on the costs and quality of supply. A detection of the condition as well as a condition prognosis allow network operators to achieve a condition-based maintenance, which is seen as cost-effective. A method for condition detection and prognosis was developed in this work, which considers the knowledge of the experts in house of the network operators. Supervisor: Dipl.-Ing. Christian Hille Fuchs, Bernhard Experimentelle Analyse der Stromtragfähigkeit elektrischer Freileitungen For a number of reasons, such as the liberalisation of the energy market, the eastern expansion of the UCTE - grid or the development of windenergy-engineering, the demands on the german transmission grids have changed. The grids should be reconstructed to fulfill the new requirements of power flows in the future. In reality these methods require long licensing procedures (usually >10 years), therefore it is necessary

Theses

Gialinizoglou, Vasileios Entwurf und Aufbau eines Kommunikationssystems zur Anbindung dezentraler Messsysteme und steuerbarer Lasten an eine Leitstelle The phenomenon Climate Changerepresents one of the greatest challenges of our time. Meanwhile Climate Protection demands a dramatic reduction of emissions and thus the shift of the current energy system towards a more efficient one providing moresources of renewable energy. Politics recently have understood that it is the interest ofwhole mankind that certain steps have to be undertaken. Climate Protection has beenidentified as the predominant force for generating economic growth and employment. It is the contribution of this degree dissertation to assist in restricting Climate Changeby providing a better network inte-

gration of renewable energies into the established infrastructures. By this diploma thesis three systems have been developed including a measuring, a communication as well as a data logging system. These systems have been built up and implemented in order to function as the spine of a monitoring system for the surveillance of the network condition. The DIN EN 50160 serves as the basis for the identification of relevant parameters for the description of the network condition. Furthermore - grounded on the measuring method according to DIN EN 61000-4-30 - the implementation of the measuring data registration has been done. As a result all forthcomingly registrated and calculated data will be stored in a data bank within a capable structure where it will be at hand for all coming scientific research. First measurings have shown that the erected measuring system works well and that the measured data has been successfully transmitted over the virtual private network (communication system) and stored in the data base. Concerning the network throughput the performance analysis has shown good results (average network throughput 6 kbit/sec ), while the processor usage proved alarming and worth of improval. Betreuer: Dipl.-Ing. Claas Matrose, Dipl.Ing. Holger Krings Gödde, Markus Technische Analyse des Einflusses verteilter Energiewandlungseinheiten auf Verteilungsnetze mithilfe eines agentenbasierten Investitionsmodells The thesis at hand deals with technical, economic and ecological consequences of an increased integration of distributed energy conversion units on distribution grid infrastructure. The Core of the work is an agent-based investment decision model, that simulates the decision making process of households on energy conversion units

A NNUAL R EPORT 2010

Institute for High Voltage Technology

to look for alternative solutions til then. For these reasons the monitoring of overhead lines plays an important role. The idea is to take advantage of the present weather conditions in order to increase the ampacity of existing overhead lines. Therefore models are used, which describe the thermal behavior of overhead conductors subject to different weather conditions. Using the CIGRE model, a validation should be made by means of experimential investigations. This thesis focuses on constructing a laboratory, in order to investigate the impact of low wind speeds (< 1m/s) on conductor temperature. The construction is based on the analyses of the existing technical requirements and the sensitivity of the physical model. After the description of realisation the breadboarding will be calibrated and analysed. Last but not least, measurements will be carried out in order to analyse the validation of the model. Betreuer: Dipl.-Ing. Martin Scheufen, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs

91

Institute for High Voltage Technology

92

for the example of photovoltaic, microcombined heat and power and small wind turbine units taking into account socioeconomic factors. Under consideration of weather and geographical data the model provides penetration rates for different types of areas until 2030. In addition, electric vehicles are considered as an exogenous factor. On this basis load flow calculations are performed for different distribution systems and the effectiveness of control mechanisms through intelligent local network stations, charging strategies of electric vehicles and interference in the subsidization of energy conversion units were analyzed. Furthermore, a method is presented by which the combination of agent populations and distribution networks can be evaluated for various support strategies and calculatory areas in terms of the specific CO2 emissions (ecologically), subsidies and network expansion costs (economically), and voltage ligament injuries and self-sufficiency (technically) over time until 2030. One result of this work is the heavy stress of distribution networks in rural and sunny areas in southern Germany by the increasing number of photovoltaic installations, leading to locally high grid development costs and relatively low levels of self-sufficiency. In the model, a higher subsidy for combined heat and power and small wind turbines in the context of the technical ecoefficiency leads to better results. It is also noted that intelligent local network stations can significantly contribute to network stability, but can only delay necessary grid development costs minimally. With a moderate penetration of electric vehicles with a small power input the network stability cannot be significantly increased. Overall, it is found that the increased penetration of distributed energy conversion units leads to high supplies of electric energy into the transmission grid infrastructure, especially during summer in southern, populous areas. This will significantly affect the power

A NNUAL R EPORT 2010

plant scheduling in the future. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Wirt.-Ing. Eva Szczechowicz

Güthe, Alexander Thomas Ökologisch-technische Analyse von Betriebsstrategien für Elektrofahrzeuge zur Integration von erneuerbaren Energien ins Übertragungsnetz Through voluntary commitment, such as the Kyoto Protocol, increasing pressure is noticeable in regard to avoid emissions. The objectives of the policy are ambitious and can only be realized with great effort. A significant contribution to achieve these aims could be the integration of new concepts, such as electric vehicles and renewable energy. Because of the ecological potentials of electric vehicles as well as plug-in hybrids these are investigated in the present work. Two possibilities to foster the reduction of the emissions are highlighted the lowering of direct vehicle emissions as well as supporting the integration of renewable energies. For this, appropriate loading strategies are developed and implemented using a simulation model. A distinction of six categories of vehicle permits a both precise and differentiated determination of vehicle emissions, allowing a comparison with conventional vehicles which goes beyond the ordinary averaging. As a rating the CO2 equivalent is used. The results show a clear environmental advantage of electric vehicles over their counterparts with combustion engine. It also shows that the limits of possible support of the integration of renewable energies such as centrally bundled offshore wind farms through electric vehicles are quickly reached. Betreuer: Dipl.-Wirt.-Ing. Eva Szczechowicz, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs

Theses

Szenarienbasierte Bewertung der Entwicklung von Elektromobilität unter Berücksichtigung netztechnischer und wirtschaftlicher Aspekte The focus of this thesis is on the implications of charging infrastructure for electro mobility. Infrastructure is admittedly a crucial aspect for the future technological diffusion of electric cars into the mass market, due to the very limited user willingness of driving an electric car without the opportunity of public charging. None the less, critical aspects regarding the business, economic and ecologic implications are not elaborated yet. In the end, the future charging infrastructure will be defined by the future electric fleet. As well, only limited insights are given in the framework of the future fleet development. Within this thesis an approach is developed that - based on real life motorized journeys - is able to extrapolate future trends within the fleet population by using a multi-agent-system. Based on the penetration of electric cars, policy advices are given concerning subsidies strategies for charging infrastructure. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Ing. Claas Matrose, Dipl.-Kfm. Andreas Pfeiffer Kalverkamp, Frederik Bewertung der elektrischen Eigenschaften von Windparks auf Basis von Modellrechnungen mit Blick auf die Vorgaben der Systemdienstleistungsverordnung The present study deals with the evaluation of electrical properties of power generation systems. Its verification at the point of common coupling is required by The Ancillary Service Directive (SDLWindV) since 2009. Corresponding references are based on an authentic wind farm. At first the underlying grid codes are presented as well as the effective technical guide-

lines of the german research association for wind power and renewable energies, FGW. Moreover the methods to be applied and the arithmetic techniques for the validation of simulation models are explained, which represent the basis for the dynamic behavior of wind farms during grid faults. Furthermore the methodology for evaluation of electrical properties of power generation systems based on these simulation models and further characteristics of wind turbines and additional equipment of the wind farm is identified and demonstrated. The examination of the validation criteria using a concrete manufacturer model of a real wind turbine leads to a validated simulation model to portray the behavior of the whole power generation system during a grid fault. These dynamic stability analyses and additional calculations of the electrical properties of the wind farm - using power flow calculations and approved calculation methods - as well as the verification of certificates of wind turbines and characteristics of further components constitute the basis of the evaluation in terms of its conformity to standards. In addition further technical analyses of different net conditions and -topologies are realized and subsequent evaluations concerning the system stability accomplished. These technical studies are followed by a monetary evaluation of the ancillary service for the grid that has been demonstrated and an effortbenefit-assessment of the steps in process validation and system simulation. The work closes with an outlook of additional open issues in the determined certification approach. Betreuer: Dipl.-Ing. Martin Brennecke, Dipl.-Phys. Bernhard Schowe-von der Brelie, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Kippen, Michael Monetary network evaluation based on free

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Hinrichs, Hauke

93

Institute for High Voltage Technology

available information The worldwide increasing energy market liberalization offers the possibility of network acquisition and sale to network operators. A previous monetary evaluation of the networks is a mandatory requisite for purchasers to judge and estimate the purchase amount. In this work, a method for value estimation of high and highest voltage networks has to be developed based on free available information. For this, it was essential to determine the number of installed cost-intensive equipments. An innovative procedure has been developed and can be found in the current annual report under the topic Synthetic generation of network topologies based on the use of free available data. Supervisor: Dipl.-Ing. Christian Hille

Klosta, Florian

94

Modellierung und Analyse von virtuellen Energieversorgungssystemen mit verteilten Wandlungseinheiten This thesis researches the supply of an investigation area with power and heat. The chosen approach to the topic is a Virtual Energy Supply System with centralized control consisting of the interconnection of dispersed energy conversion, load, storage and net to guarantee an economic, ecologic and save supply with power and heating in the private sector. The base of the developed model of the Virtual Energy Supply System is the analysis of the requested heat and the conversion units for the thermal and electrical feed-in. The model is capable to simulate and evaluate differently composed electrical and thermal conversion unit parks and integrated strategies. The results show that the interconnection to a Virtual Energy Supply System provides economic and ecologic advantages in comparison with the uncontrolled supply with power and heat. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing.

A NNUAL R EPORT 2010

Thomas Dederichs, Dipl.-Ing. Stefan Krengel, Dipl.-Ing. Joachim Funcke (IDG) Koch, Myriam Investigation of the measurement of arc resistance distribution depending on position The electric arc inside high voltage circuit breakers is surrounded by a nozzle and cooled by a quenching gas flow. This gas flow leads to an increase of arc resistance close to the natural current zero crossing. A successful interruption of the current is strongly depending on the increasing arc resistance. Present investigations show not only a strong influence of the total arc resistance but of the spatial distribution on the interruption capability. For investigating this spatial arc resistance distribution over time a new measuring system is developed and established at the Institute for High Voltage Technology. The present diploma thesis performs basic investigations to achieve important characteristics and parameters for dimensioning this new measuring system. Supervisor: Dipl.-Ing. Matthias Hoffacker Lütcke-Lengerich, Christoph Analyse von Umwelteinflüssen auf die Stromtragfähigkeit elektrischer Freileitungen (Freileitungs-Monitoring) mittels charakteristischer Betriebs- und Geodaten Due to the increasing supply of renewable energies in the German high-voltage grid, especially speaking of the more dominating wind energy supply in the north the existing overhead lines come up against their limiting factors. Using overhead line monitoring it is possible to increase the carrying capacity of the overhead line using weather-related cooling effects. To calculate the higher current carrying capacity the Cigré-Model is taken as a basis. Refering to this the influence of the wind speed has the greatest cooling effect. Therefore, in

Theses

Ma, Ming Analyse und Modellierung der Ausgestaltung zukünftiger Marktplätze für energiebezogene Systemdienstleistungen In the present work the control energy market design is investigated in the future to support the rapid integration of renewable energy. The economic and technical influences are evaluated for the modeling of these designs to market. To the provision of ancillary services in the area frequency control (such as the provision of control energy/power) lead the four transmission system operator by the tender on the common Internet-platform to meet their own needs of the control power. This auction mechanism is the current German control power market. This model is faced with the challenge to adapt to future situations and to guarantee the security system ever. For the economic analysis are the criteria of minimizing costs and reducing the burden of consumers in the center of events. The aim of the developed market model is to create scenarios of the future market design, to examine procurement costs and profitability

of the power plant operators in compliance with technical constraints. In the context of the scenario analysis, the factors influencing the control power consumption and price trends are analyzed for the control energy market. The control power demand for 2020 is determined using the first model for different scenarios and view both values of the Economy can be determined using the market model for the scenarios of market design. After examining the results in particular future trading opportunities of the control energy market is recommended. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Wirt.-Ing. Eva Szczechowicz

Institute for High Voltage Technology

this work a method for spatial interpolation of wind speed has been implemented taking shadowing effects of the landscape into account. Using this method, the higher capacity of an exemplary power transmission line was represented on a potential map. This is continued by a bottleneck analysis of transmission line, which helps the transmission system operator to decide where weather measurement technology should be installed. The final evaluation based on a synthetic network shows that by using the overhead line monitoring at the found locations is possibility to increase the capacity of overhead lines on the basis of the implemented procedure. Betreuer: Dipl.-Ing. Martin Scheufen, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs

Meinerzhagen, Ann-Kathrin Szenarioanalyse zur technischen Ökoeffizienzbewertung globaler Energieversorgung This thesis deals with an analysis of influences of global energy supply and potential assessment criteria. To this end, scenario analysis is introduced as the chosen approach. A survey of different methods for evaluating the technical eco-efficiency of the global energy supply and an analysis of the current supply situation form the ground of the scenario analysis. First, global causal chains that affect the energy supply composition are identified by reviewing models of economic, social and ecological processes. Further influences are determined through an appraisal of known scenarios. Based on the findings from these first steps, drivers of the global energy supply are defined and their interconnected dependencies are explicated. Founded on this knowledge, a model of worldwide energy usage is developed. A sensitivity analysis of the model gives further insight into the interactions that rule global energy demand. Thus, scenarios for a further assessment of worldwide energy supply are elaborated and criteria for a technical eco-efficiency assess-

A NNUAL R EPORT 2010

95

Institute for High Voltage Technology

96

ment are devised. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Prof. Dr. Dr. Rafaela Hillerbrand Natemeyer, Hendrik Szenarioanalyse zur ganzheitlichen Bewertung zukünftiger Ausbaumaßnahmen im europäischen Übertragungsnetz The intensive promotion of renewable power generation and the demand for the expansion of the European power trading necessitate the development of the European transmission grid. To create this development ideal, a tool for assessing predefined extension measures becomes necessary. This tool will be developed in this study. Therefore, a model of the grid and a power plant database are developed and assembled into a complete model of the European electricity supply. To calculate this model, the methodology of the OPF (Optimal Power Flow) is used. In addition, this is extended with respect to an ecological approach (Ecological OPF). The methodology of the OPF is explained and validated in comparison to a simulation of the power plant scheduling. On the basis of planning principles in the transmission grid, economical, ecological and technical criteria are created that allow an assessment of development measures. To represent various loads of the grid, scenarios are defined that represent the different situations of electricity demand and the availability of renewable energy sources. Finally, development measures in the European transmission grid are identified fromexternal sources and evaluated in terms of economical, ecological and technical criteria. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Ing. Martin Scheufen Neef, Florian

A NNUAL R EPORT 2010

Entwicklung und Modellierung netzstabilisierender Maßnahmen für Elektrofahrzeuge im Verbundnetz Within this thesis a reduced grid model is developed, which allows to qualitatively investigate the influence of power conversion units on the stability of the European interconnected power system. Electric vehicles with energy storage systems and plugin capability are examined exemplarily for power conversion units. The examination of stability includes aspects of small-signalstability as well as those of transient stability. A restriction is made by considering the mains frequency as the important factor and neglecting the system voltage. Small-signal-stability is investigated by modal analysis and transient stability by RMS-simulations. The effects of different regulation concepts on the energy supply of electric vehicles and its impact on the damping of critical inter-area-oscillations is analyzed. With the help of the developed model it can be shown that the damping of inter-area-oscillations improves by increasing the quantity of adequately controlled electric vehicles. Also the fall off of mains frequency after power plant failures can be attenuated. Nowadays it is not permitted to participate with the electric vehicles in the primary control market, in spite of the technical feasibility. Therefore a reorganization of the statutory framework is recommended to guarantee the grid stability for the future, as a further reduction of rotating mass is forecasted. Betreuer: Thomas Pollok, M.Sc., Dipl.Wirt.-Ing. Eva Szczechowicz Nikolic, Gregor Investigation and modelling of the dielectric strength for hot gases in insulating nozzles Nowadays sulphur hexafluoride SF6 is used in high voltage circuit breakers due to its outstanding dielectric and arc quenching

Theses

Plath, Cornelius Condition assessment of power supply equipments This diploma thesis investigated which established possibilities could be applied to assess the condition of equipments. For this it is possible to employ historical data, as well as the knowledge of manufacturers and the experience of network operators. An appropriate possibility has been implemented and verified by means of an available database. Lastly, possible economical scenarios of the condition were sketched in order to introduce the results of this work into a network evaluation. Supervisor: Dipl.-Ing. Christian Hille Roder, Jochen Spannungsebenenübergreifende Bewertung von Verteilungsnetzen unter Berücksichtigung von Elektromobilität In recent theses at the IFHT the effect of rising penetration by electric vehicles on single voltage levels of distribution grids has been investigated using powerflow calculations. Furthermore different charging

strategies for electric vehicles have been developed and evaluated on low-voltage grids. Based on the already developed methods the coupling of the voltage-levels is involved in this thesis for the first time. The analysis of distribution grids is expanded to the detection of weak spots for directed grid enforcement and to the consideration of redundancy concerning switching operations. In order to simulate controlled loads the loading strategy based on multi-agentsystems is integrated into the developed simulation- and evaluation model. As an interface to this, different control concepts are develoved, which aim for a reduction of component overloads, and which can be used on the low-voltage level as well as the medium voltage-level. Due to the modular character of this control a directed relief of the requested voltage-level can be achieved. Moreover an impact on the voltage-band is implemented using a tap-changer control of the transformers. The effects of rising penetration by electric vehicles with and without load control are compared in simulations, and appropriate suggestions for the application of the controls are given. Summarizing the results penetrations higher than 50% can be admitted by the examined grids using the developed controls, without overloading any components, while uncontrolled charging is leading to significant violations already at a penetration of 8%. After a critical discussion of the developed control concepts is done, possible extensions are introduced. Betreuer: Thomas Pollok, M.Sc., Dipl.-Ing. Claas Matrose Schaefer, Annegret Correlation between maintenance and incentive regulation for different asset management strategies On 1.1.2009 came the Energy Industry Ordinance on Incentive Regulation into effect. The network operators must adjust their an-

A NNUAL R EPORT 2010

Institute for High Voltage Technology

properties as good thermal conductivity and high electronegativity. But as SF6 is a greenhouse gas, the actual debate on climate change makes the search for alternatives necessary. The focus of previous investigations is set on the thermal interruption capability of the possible SF6substitutes. In the same way the dielectric recovery of those gases is of high importance. Preliminary investigations show an interaction between insulating nozzles and quenching gas which can lead to a dielectric breakdown in case of slightly axially blown arcs. In the context of this diploma thesis the dielectric recovery of alternative quenching gases is further investigated. Supervisor: Dipl.-Ing. Daniel Eichhoff, Dipl.-Ing. Andreas Kurz

97

Institute for High Voltage Technology

nual revenue according to the revenue upper limit determined by the Federal Network Agency and therefore should make use of possible efficiency improvements. Furthermore, the incentive regulation has an influence on the maintenance strategies and on the investment behavior of the network operators. In this work, the complete correlation between incentive regulation and maintenance was illustrated and the consequences on the network components were analyzed. Additionally, an asset simulation for the revenue upper limit of the distribution network providers was developed, which is also able to consider modified asset management strategies. Supervisor: Dipl.-Ing. Christian Hille Scheefer, Christian

98

Technische Analyse gekoppelter Gleichund Wechselspannungsübertragungssysteme im europäischen Verbundnetz The increasing volatile generation from renewable sources and the trade of electrical energy inside liberalised electricity markets lead to an overloaded transmission system in Europe. An extension of the European transmission grid is required. High Voltage Direct Current (HVDC) technology seems suitable for an additional overlay network. First applications are in discussion building a connection between North African solar power generation and the European grid (Desertec project). The calculation of interconnected AC/DC-systems requires software tools, that model AC and DC components and implement both in their algorithms. Different software tools are available at the Institute for High Voltage Technology (IFHT). These tools were not tested thoroughly regarding the calculation of AC/DC-systems. This thesis presents an analysis and assessment of software tools available at the IFHT. A rating system is deduced from the theory of Usability Engineering including the assessment of

A NNUAL R EPORT 2010

functionality, usability, visualisation and flexibility. Special attention is given to the calculation of interconnected AC/DCsystems. The second part of the thesis adapts the network model of the UCTE transmission system to allow a first connection of solar power from North Africa to Europe using HVDC. Two topologies are analyzed. For each topology a linecommutated and a voltage source converted HVDC system is implemented and compared. The connection points and transmission capacity is chosen by determining the infeed-potential of each 380 kV node of the European transmission grid. To calculate the interconnected AC/DC-system, the network model has to be converted from Matpower to a program that allows AC/DC calculations. Converter tools are realized to convert power flow and optimal power flow data. For an assessment of the chosen scenarios, technical criteria are identified. The scenarios are assessed using these criteria. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Ing. Martin Scheufen Tomm, Viktor Ökoeffizienzbewertung elektrochemischer Speichertechnologien in Energieversorgungsnetzen The electrochemical storage technologies are a possibility to offer ancillary services, which are indispensable to secure net stability. In the present work the focus lies on the application of stationary, electrochemical storage systems in the power network as well as on the economic and ecological evaluation of its use. Potentially favorable operating strategies are selected and defined under economic and ecological considerations. As a part of a holistic approach numerous environmental effects and economic feedback during the entire life cycle are considered and included in the evaluation. Therefore the present work focuses on

Theses

Walkenfort, Matthias Entwicklung von Managementstrategien für Elektrofahrzeuge zur Integration in die elektrischen Verteilungsnetze Because of anticipated developments in the area of electric mobility, the national grid faces new challenges. At the same time, new possibilities arise for network control given that charging times of electric vehicles are variable within certain limits. Thus, controlled charging and discharging could have a supportive effect to the grid. However, to specifically utilize this potential, mechanisms are needed that have a controlling impact and can therefore employ this potential well-directed. Hence, this thesis is about developing an algorithm that can be used to implement complex strategies. The genetic algorithm applied for this optimization is introduced and discussed. This optimization method is modeled on the natural process of evolution. The process itself is characterized by working with a set of possible solutions. By changing and combining these solutions, they develop themselves iteratively towards an optimal solution. By including interfaces, it is possible to develop an optimizing algorithm without

further knowledge of the simulation process. Finally, a synthetic data set is used to verify both the optimization algorithm itself and particular sub-functions. It has shown that the developed algorithm is well suitable to solve complex problems, and it can be considered a very robust tool. Betreuer: Thomas Pollok, M.Sc., Dipl.-Ing. Claas Matrose Wasowicz, Bartholomäus Entwicklung einer integrierten handelsbasierten Einsatz- und Ausbauplanung des deutschen Kraftwerksparks unter ökologischen Kriterien The focus of this diploma thesis is the determination of the future composition of power plant fleets and the energy generation structure until 2030 with the integration of the present trends in the electricity power system, above all the political framework requirements. The main objective is therefore the development of a model for the longterm expansion planning of the German power plant fleet. For a realistic simulation of investments in power plants, an agent-based approach is developed. This procedure enables the modeling of expansion decisions on a yearly basis with the net present value as the economic criterion. The massive expansion of renewable energies as an exogenous dimension is incorporated by changing residual loads. A central component of the model is a power plant dispatch algorithm taking technical and economic restrictions into account which is employed to calculate revenues of an agent in a liberalized power market. Simulations are exercised for different scenarios based on an analysis of significant influencing variables on the development of the power plant fleet. As result a considerable change in the requirements for the future operation of power plants is identified. Due to sinking and partially negative residual loads with high volatility the demand for flexible

A NNUAL R EPORT 2010

Institute for High Voltage Technology

development of a model which simulates the operation of a battery in a power network depending on the type and operating strategy in accordance with technical requirements. A model of the ancillary market helps to represent the stochastic character of the achievable income. To cover also the emissions of the production, the manufacturing process of the examined Batteries is simulated. Evaluation of results involves comparison of the benefit from the use of the selected operating strategies for different types of batteries as well as determination of economic efficiency and environmental impact of influencing parameters. Betreuer: Dipl.-Wirt.-Ing. Eva Szczechowicz, Thomas Pollok, M.Sc.

99

Institute for High Voltage Technology

100

power plants is increasing. In comparison to 2010, the total thermal installed capacity is sinking in all scenarios until 2030, while the demand for gas-fired power plants is rising in all investigations. In spite of a high integration of renewable energies the fulfillment of the CO2 reduction goals of the German government could only be nearly realized in a scenario with an extension of operating times of nuclear power plants. Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Ing. Martin Scheufen Wei, Yingze Investigation of the influence of the network impedance on the thermal interruption behaviour of a circuit breaker model The interruption capability of circuit breakers is amongst others depending of the network impedance which occurs at the point of current interruption. Hence different switching circumstances - i.e. different grid topologies and different types of errors cause different interruption capabilities of a circuit breaker. The dependency of the thermal interruption capability of a SF6 circuit breaker on the network impedance is given in the literature. It is investigated whether this dependency can be applied to the circuit breaker model at the Institute for High Voltage Technology. Therefore the dependency of the thermal interruption capability of the existing breaker model on the network impedance is investigated and compared with results given in literature. Supervisor: Dipl.-Ing. Andreas Kurz, Dipl.-Ing. Matthias Hoffacker Winter, Sebastian Analyse und Modellierung der Auswirkungen technologischer Innovationen auf die europäischen Energieversorgungsstrukturen This paper analyzes infrastructural devel-

A NNUAL R EPORT 2010

opment processes of the European energy supply system which are driven by innovation. At first the main influence factors are determined including such of socioeconomic character. Subsequently those are transferred to a model of the technological and infrastructural development in Europe. This model is based on the well established theory of experience curves, whereat this theory is advanced in the context of this paper to additionally describe interdependencies between innovation processes. Finally conclusions are drawn from the results of a scenario analysis, which predict a strong expansion of renewable energy power generation and HVDC transmission systems throughout Europe. Betreuer: Dipl.-Ing. Martin Scheufen, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Zhang, Jingxuan Technical and economical Analysis of possible V2G services for electrical vehicles Electric vehicles (EVs) can store or feed energy back to the grid when they are plugged in. For this reason, EVs have the capability to provide ancillary services to the grid in a large population. This thesis focuses on the investigation of technical potential and economic profitability in providing grid regulating services by electric vehicles. At the beginning, technical requirements of German regulating service market and constraints in the distribution grid have been analyzed. Based on this analysis, a technical model was developed, in order to simulate the process of providing grid regulating services by EVs. Then an economic model has been developed by taking into account additional battery degradation cost and bidding strategies on the regulating market, in order to explore maximum profits which can be created by EVs. In addition, service strategies and service modes in providing positive regulating service in both

Theses

Master Theses Chitamara, Nuttawoot Internal Arcs in Electrical Installations - Validity Range of Pressure Calculation Methods Internal arcs cause a sudden pressure rise in electrical installations. This leads to a pressure stress acting on switchgear compartments and switchgear rooms, and could cause building collapse. In order to predict the overpressure, pressure calculations become essential. Nowadays, the pressure calculations can be carried out by two methods. The first method is the CFD calculation, which employs the finite volume method yielding to spatially resolved results. The second methodology is the improved standard calculation (ISC) method providing spatially averaged results. Due to this averaging and under consideration of geometric boundary conditions, the validity range of the ISC method needs to be investigated. For this purpose, the pressure calculation results from both methods for different sets of parameters are compared in this thesis. Supervisor: M.Sc.Kittipong Anantavanich, Dipl.-Ing. Daniel Eichhoff

Hartanti, Hesti Development of the continuous assessment model of transformer solid insulation condition by Fuzzy Logic approach Condition of solid insulation has to be in close consideration because it gives the main effect to the ageing of power transformers. Although various relevant factors of solid insulation can be continuously detected by online monitoring, generally only the thermal factor is considered for the ageing model. The goal of this work is to develop a comprehensive diagnostic model based on all available relevant factors, e.g. thermal, water content and gas in oil. The Fuzzy logic approach is to be applied for assessing the condition of solid insulation of power transformers. Supervisor: M.Sc. Tirinya Cheumchit

Institute for High Voltage Technology

unidirectional and bidirectional way have been presented and compared. Afterwards, required number of EVs according to different requirements and profit range per EV has been calculated. The technical and economic models developed in this thesis can be applied in the planning of V2G services. The calculated profit created by EVs in kinds of regulating services and their required EV numbers would give a reference to the potential V2G service organizer. Betreuer: Dipl.-Wirt.-Ing. Eva Szczechowicz, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs

Lund, Johan Assessment and modeling of switching technologies for application in HVDC-circuit breakers In future electrical power systems based on DC technology appropriate switchgears are essential for switching operations during normal conditions as well as in case of fault conditions. Contemporary AC circuit breakers are unable to perform these operations due to their basic physical process. Furthermore, the unavailability of HVDC circuit breakers for high and medium voltage applications constrains the progress of future electricity networks. To overcome this obstacle, DC circuit breakers need to be developed. Generally, various concepts and technologies for DC switching exist, e.g. semiconductor-based devices or mechanical devices as ultra-fast switches or vacuum switches with additional resonant circuits. Due to the high number of different technologies available, a systematic comparison is essential for a subsequent methodical development of a circuit breaker

A NNUAL R EPORT 2010

101

Institute for High Voltage Technology

102

for HVDC applications. Hence, the first task of this thesis contains an intensive literature study to identify suitable technologies for DC switching. Different devices and concepts, especially hybrid topologies should be evaluated considering both technological and economical issues. Based on the this analysis one breaker-setup will be identified and studied in detail, whereas transient stresses during switching operation are in the scope of these investigations. For this purpose a model approach should be developed and implemented in a simulation environment including a detailed analysis of the transient processes inside the breaker system. Finally, the behavior of the breaker topology should be investigated by means of simulations of different switching operations. The Master Thesis of Mr. Johan Lund contains mainly a theoretical approach on the analysis of DC switching technologies. The focus will be set on the modeling and the simulation of the circuit breaker topology that will be identified during the thesis. Supervisor: Dipl.-Ing. Daniel Eichhoff, Dipl.-Ing. Andreas Kurz

analyzed. Supervisor: M.Sc. Tirinya Cheumchit Muhaimin, Emir Pattern learning approach for monitoring data of power transformers Online monitoring system is supposed to be able to give continuous supervision to the important equipment like power transformer. Important detected and calculated data are stored in time-series as historical data. However, sometimes the communication part of the monitoring system does not work properly and therefore important data in some amounts of time are not stored. The main goal of this work is to develop the approach to learn the pattern of timeseries data from monitoring system in order to fill the vacancies of these data. The different process variables have to be firstly classified and then the pattern of data from sample process variable of each group has to be learned. Moreover the accuracy of synthetic data has to be examined. Supervisor: M.Sc. Tirinya Cheumchit Offermann, Michael

Makhfud, Imam Comprehensive analysis of historical data from monitoring systems of power transformers Comprehensive analyses of data from various process variables and transformers can assist many important tasks for power transformer; e.g. condition assessment. However, these data from monitoring systems are stored with different time stamps and time slots. Therefore it is not possible to directly consider all of these data together. In this work, the approach to manage historical data from monitoring systems for supporting comprehensive analyses has to be developed. Moreover, data from different transformers have to be comprehensively

A NNUAL R EPORT 2010

Indirect investigation of the influence of the network impendence on the interruption capability of an arc arrangement with an alternative blowing process for circuit breakers Circuit breakers are key components in modern power grids. They are designed to switch on and off nominal currents as well as short-circuit currents. The switching element of a circuit breaker is the switching arc which occurs due to the interruption process. To extinguish the switching arc it is cooled by a quenching gas flow in axial direction. By a pressure build-up inside the circuit breaker this gas is forced to flow onto the switching arc at current zero. This process is used in state-of-the-art self blast circuit breakers. The interruption capabil-

Theses

Srisupha, Montri Diagnostic of circuit breaker interrupter by means of a frequency response analysis Condition based maintenance requires detailed information about the condition of all important components of electrical assets. For circuit breakers, the nozzle- and contact-system is a major part for the current breaking ability. Specially the nozzle has still to be checked visually, which requires a time consuming disassembling of each pole. Thus, a fast, non-invasive method was required to increase the maintenance efficiency. A frequency response analysis (FRA) offered the possibility to detect changes in capacitive, inductive or ohmic arrangements. This gave us the opportunity to use it for the mentioned change of maintenance measures. Supervisor: Dipl.-Ing. Christian Hille Wild, Tobias Investigation of arc resistance inside circuit breakers at current zero

The electric arc inside high voltage circuit breakers is surrounded by a nozzle and cooled by a quenching gas flow. This gas flow leads to an increase of arc resistance close to the natural current zero crossing. A successful interruption of the current is strongly depending on the increasing arc resistance. The present master thesis develops and implements a method to investigate the dynamic behaviour of arc resistance at current zero. Particularly not only the total arc resistance is taken into account but the spatial arc resistance distribution along the axis of the arc is of interest. Supervisor: Dipl.-Ing. Matthias Hoffacker

Institute for High Voltage Technology

ity depends on this pressure build-up and additionally on the impedance of the network. In this thesis, an arc arrangement with an alternative blowing process is investigated experimentally. Based on former investigations the influence of the network impedance on the interruption capability is analyzed. Thus, the impedance is changed indirectly by modifying the test object. This change is realized by a connection of two test objects in series. In addition to the interruption capability, by this approach the dynamic behavior of the switching arcs can be observed with a high speed camera. Derived from the scaling law of an axially blown circuit breaker a scaling factor is introduced whereby a comparison of different measurements is possible. Supervisor: Dipl.-Ing. Daniel Eichhoff, Dipl.-Ing. Matthias Hoffacker

Bachelor Theses Adler, Sophie Ganzheitliche Analyse und Optimierung des Erzeugungsmanagements von Windenergieanlagen mittels raum-zeitlicher stochastischer Modelle Betreuer: Dipl.-Ing. Martin Scheufen, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Bennewitz, Florian Szenarioanalyse der Entwicklung des deutschen Hoch- und Höchstspannungsnetzes unter ökologischen Kriterien Betreuer: Dipl.-Ing. Thomas Helmschrott, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Bongers, Tim Dominik Theoretical investigations for the determination of technical requirements for circuit breakers in future DC grids Supervisor: Dipl.-Ing. Daniel Eichhoff, Dipl.-Ing. Andreas Kurz

A NNUAL R EPORT 2010

103

Institute for High Voltage Technology

104

Bredtmann, Christian

Grüneberg, Philipp

Entwicklung adaptiver Multi-AgentenSysteme unter Betrachtung von Elektromobilität Betreuer: Thomas Pollok, M.Sc., Dipl.Wirt.-Ing. Eva Szczechowicz

Ökologische Analyse des Einflusses von Elektrofahrzeugen auf die Städteregion Aachen Betreuer: Dipl.-Wirt.-Ing. Eva Szczechowicz, Dipl.-Ing. Thomas Helmschrott

Erlinghagen, Philipp

Haverkamp, Philip

Entwicklung einer Methode zur topologischen Analyse und Generierung von synthetischen MittelspannungsVerteilungsnetzstrukturen Betreuer: Dipl.-Ing. Thomas Helmschrott, Dipl.-Ing. Martin Scheufen

Szenarioanalyse zum Einfluss der Kommunikationstechnik auf zukünftige Smart Grid Infrastrukturen Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Ing. Markus Jordans

Falke, Tobias

Heinemann, Antonia

Analyse und Modellierung von Betriebsstrategien von Energiespeichern für den Einsatz in intelligenten Verteilungsnetzen

Analyse des Einflusses von Elektrofahrzeugen auf die V2G-Einnahmen am zukünftigen Regelenergiemarkt Betreuer: Dipl.-Wirt.-Ing. Eva Szczechowicz, Thomas Pollok, M.Sc.

Betreuer: Dipl.-Ing. Stefan Krengel, Dipl.Ing. Dipl.-Wirt.Ing. Thomas Dederichs

Heyn, Martin Frankeser, Sophia Johanna Mathematical methods for the determination of the assets amount in electrical networks Supervisor: Dipl.-Ing. Christian Hille

Design and evaluation of different insulation and protection concepts for high voltage battery storage systems for compensation of volatile energy sources Supervisor: Dipl.-Ing. Christoph Roggendorf

Frechen, Henning

Hinnenkamp, Mirco

Development of an algorithm for compensation of the acoustical attenuation in XLPE Supervisor: Dipl.-Ing. Gregor Brammer

Wirtschaftlich-technische Analyse von Lastmanagement in Versorgungssystemen mit dezentraler Energiewandlung Betreuer: Dipl.-Ing. Stefan Krengel, Dipl.Ing. Dipl.-Wirt.Ing. Thomas Dederichs

Gaßmann, Tim Lauer, Anna Katharina Analysis of the damage behavior of highest voltage disconnect switches Supervisor: Dipl.-Ing. Christian Hille

A NNUAL R EPORT 2010

Bewertung von Methoden zur Lebenszyklusanalyse in Hinblick auf deren Anwendung

Theses

Möller, Robert Analyse des Einflusses verschiedener Elektromobiltätsinfrastrukturen auf Kunden und Netzbetreiber im Hinblick auf Aufstellungsort und Kapazitätsbedarf Betreuer: Dipl.-Ing. Thomas Helmschrott, Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs Mühlbeier, Artur Theoretical investigations on the modeling of the dielectric recovery of hot quenching gases in insulating nozzles Supervisor: Dipl.-Ing. Daniel Eichhoff, Dipl.-Ing. Andreas Kurz Reißing, Daniel On the estimation of the asset amount of electrical networks by means of synthetic network planning Supervisor: Dipl.-Ing. Tilman Wippenbeck Shanmugalingam, Suganthan Simulative investigation of a self-blast circuit breaker Supervisor: Dipl.-Ing. Matthias Hoffacker, Dipl.-Ing. Andreas Kurz

Zurmühlen, Sebastian Modellierung des elektromobilen Verkehrs unter Berücksichtigung sozioökonomischer Daten und Analyse der Auswirkungen auf elektrische Verteilungsnetze Betreuer: Dipl.-Ing. Thomas Helmschrott, Dipl.-Wirt.-Ing. Eva Szczechowicz

Study Theses Brust, Martin Preliminary investigations of electrical and mechanical properties of syntactic foam based on flexible formulated epoxy resins Supervisor: Dipl.-Ing. Christoph Roggendorf Buß, Stefan Investigation of the influence of the nozzle geometry of a self-blast circuit breaker model Supervisor: Dipl.-Ing. Matthias Hoffacker, Dipl.-Ing. Ming Tang Frehn, Tobias Investigation of the influence of the test circuit on the interruption limit of a circuit breaker model Supervisor: Dipl.-Ing. Matthias Hoffacker, Dipl.-Ing. Matthias Behle Schneider, Jurij

Siegler, David Analyse des Einflusses ortsbezogener Subventionen auf die Entwicklung erneuerbarer Energieträger Betreuer: Dipl.-Ing. Dipl.-Wirt.Ing. Thomas Dederichs, Dipl.-Wirt.-Ing. Baris Özalay

Institute for High Voltage Technology

in Systemen der dezentralen Energiewandlung Betreuer: Dipl.-Ing. Stefan Krengel, Dipl.Wirt.-Ing. Eva Szczechowicz

Investigation on the impact of alkali-ions on the water absorption of syntactic foam Supervisor: Dipl.-Ing. Anja Strauchs, M.Sc. Andrey Mashkin Vobis, Daniel

A NNUAL R EPORT 2010

105

Institute for High Voltage Technology

Preliminary investigations of electrical and mechanical properties of syntactic foam based on flexible formulated epoxy resins Supervisor: Dipl.-Ing. Christoph Roggendorf

model Supervisor: Dipl.-Ing. Matthias Hoffacker

Projects Iloba, Nnamdi

Weiß, Beatrix Temperature measurement inside the heating volume of a self-blast circuit breaker

106

A NNUAL R EPORT 2010

Influencing factors on the condition of power transformer components Supervisor: M.Sc. Tirinya Cheumchit

Seminar Switchgear and Substations

Voltage

The seminar „High and Medium Voltage Switchgears and Substations“ imparts basic knowledge about design and function of components and substations for electrical energy transmission and distribution. The main focus lays on physical basics and economical aspects of switchgears and substations. This seminar is suitable for engineers and technical staff coming from industry, utilities and universities as well as for students of RWTH Aachen University.

After the great success of the last years, the seminar „High and Medium Voltage Switchgears and Substations“ was also arranged in 2010 on 21th and 22th of June. The seminar is carried out annually in co-operation between the Institute for High Voltage Technology, the VDE Regio Aachen, the Forschungsgemeinschaft für elektrische Anlagen und Stromwirtschaft e.V. Mannheim and the RWTH International Academy GmbH. This year, the event was visited by 60 representatives from industry and utilities as well as 120 scientific employees and students of IFHT. For this two-day advanced training course the in-house auditorium Eph was booked.

benefit from the proficiency and know-how in the following discussions. The pause times provided great opportunities for further questions and interesting discussions and enabled students to establish further contacts to the industry.

107 Abb. 2: Speakers: Dr.-Ing. Martin Schumacher (left), Dipl.-Ing. Frank Hofmann (right)

Abb. 1: Venue auditorium Eph

Reputable speakers presented various topics concerning switchgears and substations and their physical fundamentals as well as their economical aspects. This way, the participants gained a comprehensive and practical insight on these topics and were able to

Institute for High Voltage Technology

Seminar High and Medium Switchgears and Substations

In the following an overview over this year’s seminar programm is given: • Welcome Prof. Dr.-Ing. Armin Schnettler, RWTH Aachen, Institut für Hochspannungstechnik • Introduction Dr.-Ing. Martin Schumacher, ABB Schweiz AG Hochspannungsprodukte, Zürich • Electric Arcs in Circuit Breakers Dr.-Ing. Martin Kriegel, ABB Schweiz AG, Zürich • SF6-High Voltage Circuit Breakers Dipl.-Ing. Nils Werning,

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Siemens AG Energy Schaltwerk Berlin

Sector,

ABB AG Calor Emag Mittelspannungsprodukte, Ratingen

• Vacuum Circuit Breakers Dr.-Ing. Dietmar Gentsch, ABB AG Calor Emag Mittelspannungsprodukte, Ratingen

• Cables and Overhead Lines Dr.-Ing. Frank Merschel, RWE Rheinland Westfalen Netz AG, Essen

• High Voltage Fuses Dipl.-Ing. Johannes-Georg Gödeke, SIBA GmbH & Co. KG, Lünen

• Disturbances of Electricity Supply Equipment Prof. Dr.-Ing. Armin Schnettler, RWTH Aachen, Institut für Hochspannungstechnik

• High Voltage Direct Transmission Dr.-Ing. Jutta Hanson, ABB AG, Mannheim

Current

• Introduction into System Engineering and Gasisolated Switchgears Dr.-Ing. Martin Schumacher, ABB Schweiz AG Hochspannungsprodukte, Zürich • Airinsulated Switchgears Dipl.-Ing. Gerd Lingner, Siemens AG, Energy Sector, Power Transmission Division, Erlangen

108

• Local Network Substations Dipl.-Ing. Gisbert Brüggemann, Viva Electrica!, Wetter • Power Transformers Dipl.-Ing. Frank Hofmann, AREVA Energietechnik GmbH, Mönchengladbach • Medium Voltage Switchgears Dr.-Ing. Stefan Göttlich,

A NNUAL R EPORT 2010

Finally, an excursion to the transformer factory of AREVA Schorch Energietechnik GmbH in Mönchengladbach and to the 400 kV substation of STAWAG Aachen at Aachener Kreuz was provided for all students. We would like to thank again STAWAG Aachen and AREVA Mönchengladbach for the outstanding support during our visit. We also want to thank all speakers for their interesting presentations during the two-day seminar. Due to the positive resonances of the participants, it is planned to offer the seminar „High and Medium Voltage Switchgears and Substations“ again in 2011.

Contact Dipl.-Ing. Anja Strauchs [email protected] +49 241 80-94920

Seminar Protection and control systems

The seminar “Protection and control systems” offered for the first time at RWTH Aachen took place from 14th -18th November 2010 in Vorarlberg, Austria on invitation of OMICRON electronics GmbH. The aim of the seminar is to give students, especially the power engineering ones, a practical insight into the special world of protection and control in electrical power systems.

Reflection by RWTH Aachen Based on the kind invitation of OMICRON electronics GmbH, the RWTH Aachen for the first time offered a seminar on “Protection and control systems” to ten interested students in the winter term 2010-2011,. The application process allowed for selecting ten highly motivated students covering a wide range of fields of interest in power engineering and protection and control systems experience. This group of interested students met Mr. Rainer Luxenburger from OMICRON electronics GmbH, who is a skilled and very experienced speaker on the subject. In the three days of the seminar Mr. Luxenburger gave insight to the theoretical and practical world of protection by presentations, lively discussions and practical demonstrations.Mr. Luxenburger illustrated selected topics of classical power system protection and future challenges caused by the integration of distributed generators and storage devices. In the aftermath of the seminar, reports on further topics will deepen the insight into the field of protection. The visit of the switching station Bürs of Vorarlberger Kraftwerke AG illustrated the seminar content again in the systemic context. The excellent accommodation and food for the seminar group at the seminar house in Laterns offered an excellent physical, atmospheric and social basis for a successful seminar and for the necessary regeneration. Diversified social events in the evenings allowed for refreshment and provided an opportunity for closer acquaintance after in-

tensive seminar days. The students had the opportunity to get to know OMICRON electronics by talking to employees and in a guided tour demonstrating the excellent working conditions as well as the activities of OMICRON electronics. We would like to thank OMICRON electronics GmbH again for the invitation and the successful seminar. Personally, we would like to thank Mr Luxenburger as speaker and Ms. Erika Fimpel for the excellent organization and support on site. We look forward to further cooperation and the next seminar. Our thanks at this point also aims at Vorarlberger Kraftwerke AG for kindly enabling the guided tour in the substation Bürs by two highly experienced specialists.

Reflection by OMICRON electronics GmbH ”We were very pleased that the RWTH Aachen followed our invitation to the seminar “Protection and control systems” to Vorarlberg. Since the cooperation with IFHT means a lot to us, it was our special concern to give a very motivated and interested group of students a closer insight to protection technology in form of practical lecture by our expert Rainer Luxenburger. Besides the intensive course, the students also had the opportunity to examine the protection at substation in Bürs and thereby to see the theory put into practice. We have also tried to give the students an insight into the company OMICRON and hope to have

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Seminar “Protection and control systems”

109

Institute for High Voltage Technology

Abb. 1: Group picture

110

somewhat loosened up the seminar by the evening program. We look forward to welcome and host another equally motivated group of students next time and look forward to continuing our good cooperation with the IFHT.”

A NNUAL R EPORT 2010

Kontakt Dipl.-Ing. Tilman Wippenbeck [email protected] +49 241 80-90015

The Institute

A NNUAL R EPORT 2010

111

Institute for High Voltage Technology

Institute for High Voltage Technology

Chronicle 2010 04.01.10 First working day of Mathias Knaak, Jens Knauel, Daniel Winkel 11.01.10 First working day of Johannes Stüeken 15.01.10 First working day of Thomas Helmschrott, Baris Özalay 15.01.10 Visit of the company CONSENTEC with guided tour of the institute Armin Schnettler and Christian Hille 15.01.10 Attendance at FGLA-Colloquium „Electromobility“ Stefan Federlein, Thomas Smolka, Bernhard Schowe-von der Brelie

112

17.01.10 Last working day of Martin Splettstößer 29.01.10 Last working day of Walter Taeter 28.01. - 03.02.10 DAAD-Preselection meeting for scholarship applicants (research scholarships) from the Russian Federation and selection interviews with applicants for scholarships (master programme) in Moscow Attendant: Gerhard Pietsch 01.02.10 First working day of Stefan Krengel, Tilman Schmidt 03.02.10 Visit of Prof. Dr. Mihai Cernat, University of Brasov, Romania at IFHT 05.02.10 Conference of the project-related committee to the AiF-Intention „Reduction of the compression load in electrical equipment in case of fault arc“ in Aachen

A NNUAL R EPORT 2010

Attendants: Gerhard Pietsch, Kittipong Anantavanich, Daniel Eichhoff 14.02.10 High voltage prom at IFHT 18.02.10 Preselection meeting of the Mummertfoundation (future leadership) for scholarship holders from Eastern European countries in Cologne Attendant: Gerhard Pietsch 20.02. - 02.03.10 Armin Schnettler lectures at TGGS in Bangkok 24.02. - 02.03.10 Christian Hille visits the TGGS in Bangkok 01.03. - 31.03.10 The Ph.D. students Viktor Belka and Yuri Soloview from Saint-Petersburg State Polytechnical University, Russia, attend the institute for a study visit within the LeonhardEuler-Project „Investigations of high voltage technology components“ 05.03.10 Doctoral examination of Kittipong Anantavanich with subject „Calculation of pressure rise due to internal arcs considering SF6-air mixtures and arc energy absorbers“ Reviewer: Gerhard Pietsch, Second reviewer: Hans-Jürgen Haubrich 05.03.10 Doctoral examination of Ming Chark Tang with subject „Widerstandsverteilung von Schaltlichtbögen in Selbstblasleistungsschaltern während der Stromnulldurchgangsphase“ Reviewer: Armin Schnettler, Second reviewer: Michael Kurrat 13.03. - 18.03.10 Selection Azerbaijan governmentscholarship holders (with participation of

Chronicle

19.03. - 27.03.10 CIGRE-Working Group Meeting A3.24 (fault-arc simulation) and CIGRE Technical Seminar „Modelling and Testing of Transmission and Distribution Switchgear“ in Brisbane/Australia Lecture of Gerhard Pietsch 25.03.10 Workshop „Aachen als Modellregion für Elektromobilität“, Tivoli Aachen Lecture of Armin Schnettler 29.03.10 Student internship in mathematicalphysical science area Martin Statz, Einhard-Gymnasium Aachen 27.04.10 DKE conference „Die Zukunft elektrisiert!“ in Offenbach Lecture of Armin Schnettler 29.04.10 Doctoral examination of Stefan Federlein with subject „Modellierung des typspezifischen Störungsaufkommens von Hochspannungs-Schaltgeräten“ Reviewer: Armin Schnettler, Second reviewer: Gerd Balzer 30.04.10 Last working day of Kittipong Anantavanich 30.04.10 Last working day of Ming Tang 03.05.10 First working day of Sebastian Winter 06.05. - 08.05.10 DAAD-Preselection meeting scholarship for alumni of german schools in foreign countries (new applications) in Bonn-Bad Godesberg Attendant: Gerhard Pietsch 10.05.10 First working day of Jenny Jaensson-

Weidner 15.05.10 Mannheim Marathon, Organiser: Christoph Roggendorf Attendants: Jan Becker, Florian Bennewitz, Tim Bongers, Christian Bredtmann, Ilka Bremer, Denys Demenko, Daniel Eichhoff, Andreas Höwedes, Claas Matrose, Robert Möller, Armin Schnettler, Johannes Stüeken, Beatrix Weiß, Daniel Winkel 24.5. - 28.05.10 Whitsun study trip 25.05.10 First working day of Frederik Kalverkamp

Institute for High Voltage Technology

DAAD) in Baku/Azerbaijan Attendant: Gerhard Pietsch

31.05.10 Last working day of Thomas Smolka and Mathias Behle 01.06. - 04.06.10 BMBF - journey Shanghai, China Attendant: Armin Schnettler 03.06.10 Honor of Bartholomäus Wasowicz and Fabian Potratz for the project idea „Plug Into Future“ at the competition for students of the governmental department for education and research „Energie für Ideen“ Supervising tutor: Thomas Dederichs 15.06.10 Prof. Szkutnik, Prof. Kocak at IFHT 21.06. - 22.06.10 Seminar „Hoch- und Mittelspannungsschaltgeräte und -anlagen“ Supervisor: Martin Schumacher 29.06.10 2. congress to smart grids, the internet of energy and the intelligent electromobility, at Eurogress Aachen Lecture of Thomas Dederichs 01.07.10 First working day of Jochen Roder and Michael Kippen 02.07.10 Doctoral examination of Michael Keßler

A NNUAL R EPORT 2010

113

Institute for High Voltage Technology

with subject „Einsatz von elastischen syntaktischen Schäumen in der Hochspannungstechnik“ Reviewer: Armin Schnettler, Second reviewer: Volker Hinrichsen 02.07. - 04.07.10 Sailing weekend of IFHT at Veerse Meer 09.07. - 10.07.10 DAAD-Preselection meeting scholarship for alumni of german schools in foreign countries (extensions) in Bonn-Bad Godesberg Attendant: Gerhard Pietsch 12.07.10 Doctoral examination of Jörg Feldmann with subject „Planungsgrundsätze für Niederspannungsnetze unter Berücksichtigung von Versorgungsqualität und Kosten“ Reviewer: Hans-Jürgen Haubrich, Second reviewer: Armin Schnettler

114

02.08.10 First working day of Sepideh Doroudian and Bernhard Fuchs 18.08.10 IFHT staff day to the Ahrtal, governmentbunker Bad Neuenahr, wine tasting and guided tour at the winegrowers cooperative Mayschoss 21.08.10 Rad am Ring, Nürburgring Attendants: Christian Hille, Matthias Hoffacker, Tobias Küter, Lutz Malchus, Claas Matrose, Mark Meuser, Andreas Nolde, Thomas Pollok, Ralf Puffer, Christoph Roggendorf, Daniel Schacht, Oliver Scheufeld, Armin Schnettler 22.08. - 27.08.10 CIGRÉ Session 2010 in Paris Attendants: Thomas Dederichs, Stefan Federlein, Ralf Puffer, Armin Schnettler 22.08. - 24.08.10 CIGRE-Working Group Meeting A3.24 (fault-arc simulation) in Paris

A NNUAL R EPORT 2010

Attendant: Gerhard Pietsch 01.09.10 First educational day of Volker Lontzen 03.09. - 18.09.10 Excursion to Australia Attendants: Gregor Brammer, Thomas Dederichs, Stefan Federlein, Christian Hille, Matthias Hoffacker, Maximilian Keller, Michael Keßler, Andreas Kurz, Andrey Mashkin, Claas Matrose, Gerhard Pietsch, Thomas Pollok, Ralf Puffer, Christoph Roggendorf, Martin Scheufen, Armin Schnettler, Anja Strauchs, Eva Szczechowicz 06.09. - 11.09.10 „International Conference on Condition Monitoring and Diagnosis“ (CMD 2010) in Tokyo (Japan) Attendant: Tirinya Cheumchit 15.09.10 First working day of Rita Kurth 28.09. - 30.09.10 DAAD-Scholarship seminar for alumni of german schools in foreign countries and language schools in Berlin Lecture of Gerhard Pietsch 30.09.10 Last working day of Jenny JaenssonWeidner 30.09.10 Handover of the E-Fiat 500 to institutes of the RWTH Aachen (two cars to the IFHT) with attendance of Aachen’s mayor 01.10.10 Andrey Mashkin gets the EEIM John Neal Award for the paper „Analysis on the Electrical Properties of Syntactic Foam Under Various DC Field Stresses“ 01.10.10 Conference of the project-related committee to the AiF-Intention „Reduzierung der Druckbeanspruchung elektrischer Anlagen im Störlichtbogenfall“ in Aachen Attendants: Gerhard Pietsch, Daniel Eich-

Chronicle

04.10.10 First working day of Lars Schröder, electronic laboratory 18.10.10 Armin Schnettler signs a cooperation agreement „Clarification of properties of controlling pressure rise due to fault arcs“ with the Central Research Institute of Electric Power Industry (CRIEPI), Japan 25.10. - 28.10.10 DAAD-Preselection meeting for the awarding of scholarships for candidates from the Russian Federation due to the LomonossovProgramme in Moscow Attendant: Gerhard Pietsch 01.11.10 Discussion of the research group CIGRE A3 in Aachen Attendants: Armin Schnettler, Gerhard Pietsch 02.11.10 First working day of Gregor Nicolic 04.11.10 Doctoral examination of Michael Roscher Reviewer: Dirk Uwe Sauer, Second reviewer: Armin Schnettler 08.11.10 First working day of Hendrik Natemeyer 08.11. - 09.11.10 VDE Kongress in Leipzig 2010 Attendants: Armin Schnettler, Stefan Federlein, Sepideh Doroudian, Thomas Helmschrott, Christian Hille, Stefan Krengel, Thomas Pollok, Eva Szczechowicz 09.11.10 Bestowal of the SAG Award (2nd place) to Stanislav Votruba within the eStudend days at the VDE congress in Leipzig

10.11.10 Bestowal of the price of the Otto-JunkerStiftung to Philipp Masmeier for the diploma thesis with subject „Theoretische und experimentelle Untersuchung von Lichtbogenform und Schaltverhalten in Leistungsschalter-Modellanordnungen mit alternativen Düsengeometrien“ / „Theoretical and experimental investigation of the switching behaviour and the form of the electric arc in circuit breaker model arrangements with alternative nozzle geometries“ 12.11.10 Anja Strauchs gets the IEEE DEIS Graduate Award 2010 for outstanding PhD works on the field of electrical insulation systems, dielectrics and high voltage technology

Institute for High Voltage Technology

hoff

13.11.10 „34. Hochwissenschaftliches Kolloquium“ at IFHT with award of the price „Mann mit der ruhigen Kugel“ to Thomas Dederichs 22.11.10 First working day of Markus Gödde 14.12. - 17.12.10 DAAD-Preselection interviews with scholarship applicants (master programme) in Skopje/Macedonia Attendant: Gerhard Pietsch 17.12.10 Doctoral examination of Torsten Wirz with subject „Verfahren zur Optimierung der Ultraschalldiagnostik an polymeren Isoliersystemen“ Reviewer: Armin Schnettler, Second reviewer: Ernst Gockenbach 30.11.10 Last working day of Leo Künzer, electronic laboratory 31.12.10 Last working day of Michael Kippen

A NNUAL R EPORT 2010

115

Institute for High Voltage Technology

116

Publications Kittipong Anantavanich, Gerhard Pietsch, Daniel Eichhoff Importance of SF6-air gas data for pressure calculation due to fault arcs in electrical installations Proc. XVIIIth Int. Conf. on Gas Discharges and Their Appl., Greifswald, 2010 Germany, pp. 122-125 Kittipong Anantavanich Calculation of Pressure Rise in Electrical Installations due to Internal Arcs Considering SF6-Air Mixtures and Arc Energy Absorbers RWTH Aachen University, Dissertation 2010, Aachener Beiträge zur Hochspannungstechnik - Band 14, Verlagshaus Mainz, ISBN 3-86130-677-8 Tobias Blank, Stephan Thomas, Christoph Roggendorf, Thomas Pollok, Ionut Trintis, Dirk Uwe Sauer Design and construction of a test bench to characterize efficiency and reliability of high voltage battery energy storage systems 32nd International Telecommunications Energy Conference (INTELEC), Orlando, FL, USA, 6-10 June 2010 Tirinya Cheumchit, Armin Schnettler, Thanapong Suwanasri Derivation of Aging Characteristics for Power Transformers by Artificial Intelligence Techniques Conference Proceedings of CMD 2010 Stefan Federlein, Bartosz Rusek Praxisnahe Modellbildung zur Bewertung des Störungsaufkommens von Hochspannungs-Schaltgeräten FGH-Fachtagung, Heidelberg, September 2010

A NNUAL R EPORT 2010

Stefan Federlein, Claas Matrose SmartWheels - Intelligente Elektromobilität in der Modellregion Aachen RWTH Themen, Elektromobilität made in Aachen, Ausgabe 02/2010, ISSN-Nr. 0179079X Stefan Federlein Modellierung des typspezifischen Störungsaufkommens von HochspannungsSchaltgeräten RWTH Aachen University, Dissertation 2010, Aachener Beiträge zur Hochspannungstechnik - Band 15, Verlagshaus Mainz, ISBN 3-86130-678-6 Thomas Helmschrott, Devid Perissinotto, Martin Scheufen, Armin Schnettler Integriertes Verkehrs- und Energieflussmodell / Integrated Modeling of Traffic and Power Flows VDE-Kongress 2010, Leipzig, November 2010 Thomas Helmschrott, Devid Perissinotto, Martin Scheufen, Armin Schnettler Integriertes Verkehrs- und Energieflussmodell Internationaler ETG-Kongress, Leipzig, November 2010 Mikimasa Iwata, S. Tanaka, T. Ohtaka, T. Miyagi, T. Amakawa, Kittipong Anantavanich, Gerhard Pietsch CFD Calculation of Pressure Rise and Propagation Depending on Arc Energy in a Closed Container Proc. XVIIIth Int. Conf. on Gas Discharges and Their Appl. 2010, Greifswald, Germany, pp. 146-149 Mikimasa Iwata, Kittipong Anantavanich, Gerhard Pietsch

Publications

Mikimasa Iwata, S. Tanaka, T. Othaka, T. Amakawa, Kittipong Anantavanich, Gerhard Pietsch Influence of electric arc energy on pressure rise and propagation due to high current fault arc in a closed container Records of the 2010 Annual Meeting I.E.E. Japan, paper 6-206, 2010, p. 350 Christoph Kahlen, Holger Krings, Peter Wittlinger, Bernhard Schowe-von der Brelie, Stefan Federlein, Claas Matrose, R. Frings, Thomas Smolka Einflüsse und Auswirkungen von Elektromobilität auf das Energieversorgungsnetz sowie Anforderungen und Voraussetzungen für eine interoperable Netzintegration Internationaler ETG-Kongress, Leipzig, November 2010 Michael Kessler, Armin Schnettler Investigation of the DC Breakdown Mechanism in Elastic Syntactic Foams IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 17, Issue 3, pp. 898905, IEEE, Juni 2010 Michael Kessler, Christoph Roggendorf, Armin Schnettler Behavior of Elastic Syntactic Foams under Impulse Voltage Stress Conference Record of the 2010 IEEE, International Symposium on Electrical Insulation (ISEI), 4 Seiten, San Diego (USA) 6.-9. Juni 2010 Michael Keßler Einsatz von elastischen syntaktischen Schäumen in der Hochspannungstechnik RWTH Aachen University, Dissertati-

on 2010, Aachener Beiträge zur Hochspannungstechnik - Band 16, Verlagshaus Mainz, ISBN 3-86130-679-4 Stefan Krengel, Thomas Dederichs, Thomas Smolka, Armin Schnettler Scenario Analysis about Ecological and Economical Benefits of Dispersed Generation in Future Energy Networks Third International Conference on EcoEfficiency, Egmond aan Zee, Juni 2010 Stefan Krengel, Armin Schnettler, Thomas Dederichs, M. Bertling, J. Biermann, T. Gervens, M. Fiedeldey, E.-P. Meyer Das Virtual Power System Allgäu - der Pilotbetrieb VDE-Kongress 2010, Leipzig, November 2010 Andreas Kurz, Martin Seeger, Daniel Eichhoff, Armin Schnettler Determination of the dielectric recovery inside insulating nozzles concerning gas flow Proc. XVIII International Conference on Gas Discharges and Their Applications, Greifswald 2010, pp. 58-61 Andreas Kurz, Christof Kahlen, Ming Chark Tang, Matthias Hoffacker, Armin Schnettler Investigation of New Nozzle Geometries and their Influence on the Interruption Behaviour for Gas Circuit Breaker Proc. XVIII GD 2010, International Conference on Gas Discharges and Their Applications, Greifswald, Germany 2010, pp. 166-169 Andrey Mashkin Analysis on the Electrical Properties of Syntactic Foam Under Various DC Field Stresses European Electrical Insulation Manufacturers John Neil Award 2010

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Influence of Current and Electrode Material on Fraction kp of Electric Arc Energy Leading to Pressure Rise in a Closed Container During Internal Arcing IEEE Trans. Power Del., Vol. 25, No. 3, 2010, p. 2028 - 2029

117

Institute for High Voltage Technology

118

Andrey Mashkin, Anja Strauchs, Armin Schnettler, Bernd Freiheit-Jensen, Arne Lunding Einfluss der inneren Feldstärkeverteilung auf die elektrische Durchschlagfestigkeit von syntaktischem Schaum / The Impact of the Internal Field Distribution on the Breakdown Strength of Syntactic Foam ETG-Fachtagung: Isoliersysteme bei Gleich- und Mischfeldbeanspruchung; 27.28.9.2010; Köln Andrey Mashkin, Anja Strauchs, Armin Schnettler, Arne Lunding The Impact of Different Alkali Ion Concentrated Hollow Glass Microspheres on the Electrical Breakdown Mechanism of Syntactic Foam 10th IEEE International Conference on Solid Dielectrics (ICSD); July 4th -9th, 2010; Potsdam, Germany Andrey Mashkin, Anja Strauchs, Armin Schnettler, Jörn Podlazly, KarlMika Reß Investigation on Inverse Volume Effect of Syntactic Foam under Uniform dc Field Stress IEEE International Symposium on Electrical Insulation (ISEI) 2010; 6.-9.Juni 2010; San Diego, USA Claus Neumann, Bartosz Rusek, Christian Schorn, Stefan Federlein, Armin Schnettler, Gerd Balzer, Thomas Krontiris Modelling the effect of maintenance on failure occurrence and lifetime management of high voltage circuit breakers CIGRE Session, A3-203, Paris (France) 2010 Thomas Pollok, Eva Szczechowicz, Claas Matrose, Georg Stöckl, Georg Kerber, Martin Lödl, Armin Schnettler, Rolf Witzmann, Petra Behrens Flottenversuch Elektromobilität - Netzma-

A NNUAL R EPORT 2010

nagementstrategien mittels elektrifizierter Fahrzeugflotten VDE-Kongress 2010, Leipzig, November 2010 Ralf Puffer Netzoptimierung durch witterungsabhängigen Freileitungsbetrieb und Hochtemperaturleiter Deutsche Umwelthilfe (DUH) Netzkongress 2010, Erneuerbare ins Netz, Berlin 2010 Ralf Puffer Möglichkeiten zur Erhöhung der Strombelastbarkeit von Freileitungen am Beispiel von Freileitungs-Monitoring und Hochtemperaturleitern 6. ETP-Konferenz Instandhaltung von Freileitungen und Freileitungsmasten, 2010, Frankfurt E. Rijks, G. Sanchis, G. Balzer, Stefan Federlein, G. Ford, A. Gaul, T. Jesson, K. Sand, B. Sander, S. Voronca et al. Transmission Asset Risk Management Cigré Working Group C1.16, TB 422, Paris 2010 E. Rijks, G. Sanchis, G. Balzer, Stefan Federlein, G. Ford, A. Gaul, T. Jesson, K. Sand, B. Sander, S. Voronca, et al. Transmission Asset Risk Management Cigré Working Group C1.16, TB 424, Paris 2010 Christoph Roggendorf, Michael Kessler, Sascha Schulte, Armin Schnettler Accelerated Test Procedures for Hydrothermal Aging Conference Record of the 2010 IEEE International Symposium on Electrical Insulation (ISEI), 5 Seiten; San Diego (USA) 6.-9. Juni 2010 Martin Scheufen, Ralf Puffer, Bartosz Rusek

Publications

Michael Schmale, Hans-Joachim Dräger, Ralf Puffer Implementation and Operation of a Cable Monitoring System in Order to Increase the Ampacity of a 220-kV Underground Power Cable CIGRE Session, B1-113, Paris (France) 2010 Michael Schmale, Ralf Puffer, HansJoachim Dräger Optimierung der Strombelastbarkeit eines 220-kV Kabels / Ampacity Improvement of a 220-kV Underground Cable E.ON, Bayreuth 2010 Michael Schmale, Hans-Joachim Dräger, Ralf Puffer, Reinhold Kliegel Freileitungs-Monitoring - Entwicklung und Einführung eines Systems zur witterungsabhängigen Erhöhung der Strombelastbarkeit von Freileitungen / Development and Implementation of a Monitoring System for Weather Based Dynamic Rating of Overhead Lines E.ON, Bayreuth 2010-09-08 Michael Schmale, Ralf Puffer Freileitungs-Monitoring im 380-kV-Netz Freileitungen sicher betreiben und Reserven nutzen Netzpraxis, Jg.49 (2010), Heft 11 Anja Strauchs, Andrey Mashkin, Armin Schnettler, Alexander Tröger, Bernd Wandler Einfluss von Temperatur und elektrischer Feldstärke auf die elektrische Leitfähigkeit und Spannungsfestigkeit von syntaktischem Schaum ETG Fachtagung, Isoliersysteme bei

Gleich- und Mischfeldbeanspruchung; 27.28.9.2010; Köln Anja Strauchs, Andrey Mashkin, Armin Schnettler, Jörn Podlazly The Impact of Water Absorption on the Dielectric Properties of Syntactic Foam 10th IEEE International Conference on Solid Dielectrics (ICSD); July 4th -9th, 2010; Potsdam, Germany Anja Strauchs, Andrey Mashkin, Armin Schnettler, Jörn Podlazly, Bernd Wandler The Impact of Electrical Field Stress on the Volume Conductivity of Syntactic Foam IEEE International Symposium on Electrical Insulation (ISEI) 2010; 6.-9.Juni 2010; San Diego, USA Anja Strauchs, Andrey Mashkin, Armin Schnettler, Jörn Podlazly, Bernd Freiheit-Jensen Investigations on the Partial Discharge Behavior of Syntactic Foam under Uniform Field Stress IEEE International Symposium on Electrical Insulation (ISEI) 2010; San Diego, USA 6.-9.Juni 2010 Ming-Chark Tang Widerstandsverteilung in Schaltlichtbögen von Selbstblasleistungsschaltern während der Stromnulldurchgangsphase RWTH Aachen University, Dissertation 2010, Aachener Beiträge zur Hochspannungstechnik - Band 13, Verlagshaus Mainz, ISBN 3-86130-676-X M. Wietschel (Herausgeber), M. Arens, C. Dötsch, S. Herkel, W. Krewitt, P. Merkewitz, D. Möst, Martin Scheufen Energietechnologien 2050 - Schwerpunkte für Forschung und Entwicklung Frauenhofer Verlag 2010 (ISSN: 16127455); ISBN: 978-3-8396-0102-0

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Analyse der Strombelastbarkeit von Freileitungen mit Freileitungs-Monitoring bei kleinen Windgeschwindigkeiten GIS-Anwenderforum 2010, Darmstadt

119

Institute for High Voltage Technology

POSTER SESSIONS Tirinya Cheumchit, Armin Schnettler, Thanapong Suwanasri Derivation of Aging Characteristics for Power Transformers by Artificial Intelli-

120

A NNUAL R EPORT 2010

gence Techniques CMD-Konferenz, Tokyo (Japan) 2010

Technical Excursion

An excursion with professors and assistants is conducted at regular intervals of three to four years at the institute for high voltage technology (IFHT). This year’s excursion went to Down Under. After a flying time of approx. 24 h, four stations on the east coast were visited within 12 days. The trip started in Frankfurt, through Singapore as a stopover, to the metropolises like Melbourne, Sydney and Brisbane, but also to the tropical Cairns. During the time, numerous companies, universities and places of interest were visited, where existing contacts have been refreshed and new established.

Melbourne 05.-08.09.2010 The first visiting point was Melbourne located in southeast of Australia, which revealed his diversity of the Australian lifestyle during a sightseeing tour on the first day. Then, the first business related trip started on the next day (Monday) in the morning to Traralgon. There, we got to see the power station Loy Yang as well as the adjacent brown coal open-cast mining. In this context, it was interesting to figure out, that Australia has more than enough installed power plant capacities and raw materials like coal and brown coal available however, the water shortage of the country has a significant impact on management and operation of power stations. In addition, the actual growth of population is an essential challenge in the operation and construction of assets. From the power station it then only was few miles to the Basslink converter station. The Basslink interconnector links South Australia with Tasmania over a HVDC line. Here, even for a distance of 295 km, submarine cables at a voltage of approx. 400 kV are used to transmit 500 MW of electrical power. Olex the largest cable manufacturer of Au-

Institute for High Voltage Technology

Institute excursion to Australia 03.-18.09.2010

stralia gave a comprehensive insight into the cable production on the third day in Melbourne before heading for the first domestic flight to Sydney.

121 fig. 1: Participants of the Australia excursion in the open-cast mining in Traralgon

Sydney 08.-10.09.2010 After arrival in Sydney, the largest aquarium in the southern hemisphere was visited as part of a city tour. The highlight of the city tour was certainly the Sydney Opera House, which was declared as UNESCO world heritage in 2007. On Thursday morning AEMO (Australian Energy Market Operator) gave an insight into the structure of the Australian energy market and the significant ongoing challenges in the planning and operation of assets - primarily driven by the rapid growth of population. In the afternoon the 330 kV GIS Haymarket

A NNUAL R EPORT 2010

Institute for High Voltage Technology

substation in the heart of Sydney, operated by Transgrid (an Australian transmission system operator), has been visited. The substation belongs to one of the largest in the world with a transmission capacity of 1200 MVA. It therefore was a very impressive substation, but with an inconspicuous exterior design. On 10/09/2010, we arrived at AGL (Australian Gas Light Company), the largest Australian electricity and gas supplier at the same time dealing with the highest share of renewable energies in Australia. Among other things, the attractiveness of participating in the Australian energy market through a differentiated range of products was shown. Finally, we moved from subtropical Sydney to tropical Cairns in northern Australia through the weekend.

A 7.5 km long funicular railway through the primeval forest provided a great view over the treetops and the Barron Gorge waterfalls.

fig. 3: Whitewater Rafting at Barron River

Brisbane 13.-17.09.2010

122 fig. 2: Opera House and Harbour Bridge in Sydney

Cairns 10.-13.09.2010 The very special activities in a tropical surrounding during the weekend offered impressive scenery of the Australian rain forest which was complemented with the Whitewater Rafting on Barron River. The highlight of the weekend was a day trip to the Great Barrier Reef from Port Douglas. At three different places fantastic coral formations, colourful fish and even reef sharks have been seen. Finally, a half day tour ended up in the Australian “jungle”.

A NNUAL R EPORT 2010

At the beginning of the second week the last domestic flight went to the east of Australia, Brisbane. Here both campus universities Queensland University of Technology and University of Queensland were visited on Tuesday. During a mixed workshop an exciting scientific exchange between the Australian university and our institute was initiated, which also established new contacts to these universities. The last trip - taking an entire day - was spent together with the company Powerlink , which kindly guided us to a transformer substation equipped with hybrid GIS system. Afterwards, the Wivenhoe power station was visited. The latter spot is a pump storage with two 250 MW turbines, operated by Tarong Energy. The Wivenhoe Dam creates the artificial Lake Wivenhoe, which at the same time also serves as drinking water storage for the metropolitan area of Brisbane. The day trip finished in the Lone Pine Koala Sanctuary, in which, among other things, nume-

Technical Excursion

fig. 5: Shipping tour on Brisbane River fig. 4: Koala in the Lone Pine Koala Sanctuary Park

The headquarters of AEMO, which is located in Brisbane, was the last company to be visited during the Australia trip. Here, the topic was a detailed presentation of the structure and the challenges of the Australian energy market. The big difference of the Australian compared to European grids is reasoning in its huge longitude with an only low meshed structure. Afterwards, typical Australian animals such as various spiders and snakes have been admired in the Queensland museum.

With many new impressions we finally left the interesting and amazing country of Australia with some new experiences on board on Friday.

Institute for High Voltage Technology

rous koalas and kangaroos could be fed.

Coordinators of this excursion and authors of this article: Dr.-Ing. Michael Keßler [email protected] +49 241 80-94912 Thomas Pollok, M. Sc. [email protected] +49 241 80-94958

A NNUAL R EPORT 2010

123

Institute for High Voltage Technology

Student excursion to Hamburg The metropolis Hamburg was this year’s excursion destination for the students of the RWTH Aachen University. From May 24th 2010 to May 28th 2010 Hamburg was the host for the paticipants of the excursion that consist of the chairs assistants, students and Prof. Schnettler. During this week, many companies such as Airbus, RePower and Nexans ranging from electronics to energy systems, made it possible for the students to recieve a valuable insight into their field of activities whereas the range extended from electronis in a plane to furnances of a steel mill.

Journey to Hamburg At 9 am on Whitsun monday the excursion headed to Hamburg. Despite this late departure, the bus ride went well since the students fell back to sleep immediatly. After a lunch break the mood became more livley, because the free night schedule and the room arrangements had to be discussed. At about 5 pm, we arrived at the central located A&O-Hostel in Hamburg.

ask questions. Afterwards, the guided tours started in several groups. The factory was introduced to us starting with the iron ore, over the inspection of the furnance right up to the manufacturing of the products such as steel wire (Fig. 2).

124 Fig. 2: Steel mill by ArcelorMittal

Fig. 1: View onto Hamburg’s harbour

In order to compensate for the long bus ride, the next item on the agenda was the traditional jogging with Prof. Schnettler. The well known outer Alster was the ideal route.

Arcelor Mittal - steel mill On Tuesday, we visited our first station in Hamburg, the steel mill by the company Arcelor Mittal. After a few presentations and a film about the company and the functionality of the steel mill, we were able to

A NNUAL R EPORT 2010

Very impressing were the extremly hot and loud furnances, where the iron ore is melted. Another imposing impression was the furnance racking where the liquid material is drained for the further process.

E.On grid - 110kV-substation After the exciting inspection of the steel mill we drove to the countryside for a little recovery. We visited a 110kV substation by E.On grid which is located in Norderstedt in northern Hamburg. While we had coffee and cake in a restaurant, it was introduced to us how the establishment of a substation is acomplished. Afterwards we

Student Excursion in Kreetslag where the final assembly of the A320 familiy and the structure and equipment assembly of the A380 is located (Fig. 5). Very impressing were the different dimensions of the aircraft families as well as the high quality requirements even on small planes which are checked twice.

Institute for High Voltage Technology

drove to the substation, where we looked at the construction from the outside as well as the controlling and adjustment possibilities. (Fig. 3).

Fig. 3: Group picture at the substation

At about 5 pm we went back to Hamburg where we made acquaintance with the rushhour traffic for the first time. The rest of the evening was used to get to know the inner city of Hamburg as well as the Reeperbahn.

Airbus - Cabin Systems Test Center and guided tour of the site The third day of the excursion started very early. Our first stop was at AIRBUS Operations GmbH in Buxtehude where the parts of the development are placed. The CSC (Cabin Systems Test Center) and the Testrigs for the cabin electronic demonstrated how complex the technology of a plane is and that everything is still based on analog instrumentation. A walk through the A380 was possible in a few parts at least.

Fig. 5: Excursion participants in front of the A380

Vattenfall - power station Moorburg After an exciting insight view into the aircraft construction we turned towards conventional power engineering and visited the new construction of the power plant Moorburg (Fig. 6).

Fig. 6: Construction site of the power station Moorburg

Fig. 4: Presentations at Airbus

After a very technical introduction, we visited the aircraft production from Airbus

After an introduction about the pitfalls of building a new power plant such as including the construction of a fish ladder, we got construction site friendly clothes and started a very informative and also sunny tour through the area. After a following questions-and-answers session we headed back to the inner city at 4.30 pm where the

A NNUAL R EPORT 2010

125

Institute for High Voltage Technology

126

students made a harbour tour or explored the city by foot.

RePower - TechCenter and production managment Dispite nightly activities everyone was right on time and ready for departure on the fourth day of the excursion. We drove northward to the Repower TechCenter in Osterrönfeld where we listened to presentations and technical lectures about wind energy. After a refreshment we continued our tour to the production of RePower in Husum. In this factory, they produce the nacelle and hub of the wind energy plant. It was possible for the students to climb into a finished nacelle to get an impression of the extremly tight spaces inside. After the return to Hamburg, we headed to one of the highlights of the excursion, the dinner with Prof. Schnettler which took place in a small restaurant on the Elbe.

Fig. 7: Excursion participants at Repower in Husum

A NNUAL R EPORT 2010

Nexans and return to Aachen The last day of the excursion started by checking out of the hostel. We drove from Hamburg to Hannover where we had our last station at Nexans, one of the leading companies in the cable industry. After a very interesting technical lecture about the power cable technology by Dr. Meurer, the most interesting power cable projects were introduced. After lunch we started the tour through the factory where we could recover the previously learned in the production process. During the final questionsand-answers session the students were able to ask their open questions. At about 4 pm we went back to Aachen where the students arrived tired but satisfied.

Organisation Dipl.-Wirt.-Ing. Eva Szczechowicz [email protected] +49 241 80-94916

Recreational Excursion

The trip on this year’s annual appreciation day on Wednesday, 18.08.2010 led us into the Ahr Valley where we visited the former bunker of Germany’s federal government. As the Ahr Valley is well known for its wineries we went to the Winzergenossenschaft Mayschoss for a guided tour through the wine production process and wine tasting later in the day. In the evening there was the traditional barbecue at Prof. Schnettler’s home.

Guided tour through the bunker museum When we started our coach trip it was drizzling slightly. Nevertheless, 36 employees of the IFHT were looking forward to an interesting day in the Ahr Valley which is approximately a one hour drive from Aachen. The first stop led us on the traces of the Cold War. During the 1960ties a bunker construction was build in former railway tunnels in order to offer protection for up to 3000 individuals for 30 days against the impact of a nuclear war over Germany.

part of the complex has been preserved and has been converted into a museum called „Dokumentationsstätte Regierungsbunker“. Divided into groups we went on a guided tour though be bunker where the temperature is 12°C during the whole year. On the tour we were able to examine the large number of former interior, machinery and the security installations to learn about the conditions of living and working in a bunker building.

Institute for High Voltage Technology

Appreciation Day 2010

127

Fig. 2: Security installations

Fig. 1: Bunker museum

Winzergenossenschaft Mayschoss

The bunker system consisted of tunnels with an overall length of more than 15 km. Offering accommodation for important representatives of the government it was intended to ensure that most functions of the authorities would stay intact in the case of emergency. With the end of the Cold War, the so called „Ausweichsitz der Verfassungsorgane des Bundes im Krisen- und Verteidigungsfall zur Wahrung von deren Funktionstüchtigkeit“ became useless and the deconstruction started in 1997. A small

Due to the continuing rain we skipped the planned walk on the „Rotweinwanderweg“ and drove directly to the „Winzergenossenschaft Mayschoss“, a winery in the Ahr Valley where we got a guided tour through the production facilities. Starting at the delivery gate for the grapes, we went along the pressing station, tanks for fermentation and storage and the facilities for bottling and labelling to learn the whole process of wine production. Especially the possibility to create individual labels for wine bottles as a

A NNUAL R EPORT 2010

Institute for High Voltage Technology

128

gift for jubilees, weddings or birthdays was of high interest.

Fig. 3: Botteling station

Beside the modern production facilities the winery also has a historic wine cellar where you can find good wines from many vintages back to the 1950ties. We were led though this cellar into a cosy vaulted brick cellar room for the wine tasting. Fortified by a typical lunch we tried a number of six wines starting with a light white wine over a Blanc de Noir and finished with a smooth red wine. Our guide gave not only explanations for every wine concerning the production process and the characteristics, he always hat an interesting anecdote coming with each wine. In the end of the day we learned that there are only two individual types of wine: Type one tastes well, type two not.

Fig. 4: Wine tasting

A NNUAL R EPORT 2010

Barbecue at Prof. Schnettler’s house We ended the day with a barbeque in the garden of Prof. Schnettler’s family. The rain finally stopped and most of our colleagues followed the invitation which has already become a tradition for the IFHT’s appreciation day. Everybody contributed something to the buffet so there was beside the meat a huge variety of salads, side dishes and desserts. Prof. Schnettler lighted the fireplace as is became uncommonly cold for an evening in August and so we got a cosy ending of an interesting day.

Fig. 5: Barbecue at Prof. Schnettler’s house

Organisation Dipl.-Ing. Stefan Krengel [email protected] +49 241 80-93034

Service Portfolio

The Institute for High Voltage Technology has experience in various fields of high voltage measuring and testing technologies and has appropriate facilities available to perform such investigations. Various research and testing services as well as consulting can be offered.

Services A focus of available studies is the determination of electrical and dielectric properties of insulation materials. The main scope are testing procedures for insulation materials such as the determination of the breakdown field strength of insulation materials and the measurement of the dielectric properties (dielectric permittivity ǫr , dissipation factor tan δ and surface / volume resistance) of material samples. Furthermore, partial discharge measurements can be performed at the institute with voltages up to 200 kV at a noise level below 0.3 pC. If a test requires special voltage forms such as the lightning impulse test it is possible to provide these test setups. All tests are carried out according to recognized standards or to the customer’s requirements. The institute offers the possibility to perform tests in a well defined climatic environment in order to identify the influence of temperature and humidity. Apart from the experimental test procedures, there is also the possibility to commission the IFHT with simulations. The focus in this area is on CFD-simulations of physical problems (gas-, plasma- or fluid dynamics) as well as on calculations of electromagnetic fields. For comprehensive investigations on certain tasks a collaboration within R&D projects can be offered.

Test facilities In addition to the classic high voltage test laboratories a number of specialized laboratories and test facilities are available at the IFHT:

• High voltage test lab for tests up to 400 kV AC or 800 kV impulse voltage • High voltage construction kits for modular assembly of test circuits • Synthetic test circuits for the determination of the breaking capacity of circuit breaker models with different insulation gases.

Institute for High Voltage Technology

Service Portfolio at IFHT

• Partial discharge test facilities with test voltages up to 200 kV and a noise level < 0.3 pC • Several climatic test chambers (15 m3 , -40 °C up to 140 °C), furnaces and thermal shock chambers • Test benches and test chambers for the analysis of aging behaviour (e.g. salt spray test) • Material processing laboratory for insulation materials In collaboration with the institute’s workshop, various test setups can be realised to fulfill special test requirements.

Contact Dr.-Ing. Ralf Puffer [email protected] +49 241 80-94950 Dipl.-Ing. Mathias Knaak [email protected] +49 241 80-94909

A NNUAL R EPORT 2010

129

Institute for High Voltage Technology

130

Test Center for Storage Systems and Smart Grid Technologies Future distribution networks will exhibit a long-lasting coexistence of decentralized generating units, conventional and modern consumers and the rising share of information and communications technologies. Especially the integration of new energy converters, storage systems and electric vehicles leads to new technical requirements, which are not known in detail yet. For example examinations of equipment in different stress cases (normal and overload), new protection concepts and contributions to system services by electric vehicles or batteries are largely unknown at present. In order to identify these requirements, real measurements are just as important as simulations of those systems. For this, a new laboratory for storage systems and smart grids is being built by RWTH Aachen University.

Introduction The greater integration of renewables leads away from conventional energy supply structures to highly dynamic systems. A secure energy supply in the future needs intelligent network operations, taking bi-directional power flows, decentralized power generation, strong load fluctuations and the integration of electric vehicles into account. This kind of power system needs an intensive communication between electrical energy sources, consumers, storage systems and network components, in order to know the status of the whole system.

Fig. 1: Test Laboratory: Aachen

Hüttenstraße 1-9,

There is a need to study technical requirements and concepts of grid management, which such an energy system will entail. To achieve this, grid impacts, protection concepts, as well as the behavior of components both in normal operation or in the event of faults have to be researched sufficiently.

A NNUAL R EPORT 2010

In order to carry out such metrological tests and investigations a test bench for battery storage systems and Smart Grid technologies is currently being built by RWTH Aachen.

Test Laboratory The test bench, which is sponsored by the Federal Ministry of Economics and Technology, is implemented by the Institute for Power Electronics and Electrical Drives (ISEA) as well as the Institute for High Voltage Technology (IFHT). The collaboration in the domain of energy supply and storage systems makes it possible to identify and evaluate problems, which depend on both themes. This makes investigations of specific problems of grids, storage systems and power units and their intersections possible. These are for example • Interactions between distribution networks and storage systems or power units • Behaviour of network equipment in the normal and fault case • The analysis of potentials for system services in distribution networks (for example the supply of control power

Test Center

• Effects of system services on storage systems and power units (for example the impact on the aging of storage systems)

for electric vehicles, information and communication technology and network management systems can be achieved.

• Tests of new communication networks for Smart Grids These and other investigations define specific requirements for information and communication technologies, power electronics, network technologies and network management systems, which allow to derive technical and economic optimized operating strategies. Thereby, some principle questions must be taken into account. These are for example: • How do accounting systems work inside a distributed energy supply structure? • How can be ensured that services for distribution networks do not lead to disadvantages for the users of electric vehicles? • What are the solutions for incentive pricing models for electrical energy? • Which safety requirements must be delivered by communication networks (prevention of abuse and manipulation)? Furthermore the Test Laboratory will play a very important role in the standardization and certification of new concepts for Smart Grid technologies and electric vehicles. The equipment allows standardized tests for harmonic currents, voltage fluctuation and tests of interference resistance in case of voltage dips. All components can be tested in normal and fault case of distribution grids. Thus, both the development as well as the verification of new grid codes

Fig. 2: Technical Structure of the Test Laboratory

Figure 3 shows the construction of the Test Center at the RWTH Aachen. The location, used for the construction of the laboratory, has hall areas of about 1200 m2 and open spaces of more than 2000 m2. On the outdoor areas there will be 22 single and three double laboratory containers with technical equipment for the investigation of battery cells and battery packs under different electrical and mechanical conditions. The equipment includes for example circuits with automatic charging and discharging units for synthetic testing of battery packs. Moreover there are temperature chambers (-40 - 130°C) and an impedance spectroscope, which allows to analyze electric and dielectric material properties.

The laboratory for the investigation of new structures of distribution networks will be built up in the hall area. Figure 2 shows the structure of the Laboratory. There will be a grid simulator for generating variable voltage profiles for low voltage networks up to peak power above 200 kVA per phase. The system allows the measurement of harmonics and flicker, so that the physical behavior of each test item can be analyzed. There are many test procedures, which can run automatically. These test procedures allow the measurement of

A NNUAL R EPORT 2010

Institute for High Voltage Technology

by electric vehicles and distributed power units)

131

F NS

iNS

NS iNS

NS iNS

workshop

MS NS iNS

NS

Medium voltage switch gear Local network substation Intelligent local network substation Network simulator

MS

LVRT - testsystem MV - cable LV - cable Concrete channel

HV MV STAWAG

Fig. 3: Structure of the Test Laboratory

• Harmonic currents which are lead back to the low voltage network (IEC 61000-3-2, IEC 61000-3-3), • Voltage fluctuation and flicker which are lead back to the low voltage network (IEC 61000-3-3),

132

office

iNS

Institute for High Voltage Technology

NS

• Interference resistance tests against voltage dips, short term interruptions and voltage fluctuations (IEC 610004-11) Current concepts show that communication between grid components, storage systems and power units will become more important in the future. In order to test these systems for distribution networks, the Test Laboratory will have four ”intelligent” network substations of different power (400 kVA, 630 kVA, 1250 kVA) and suppliers. These substations are equipped with electrical measurement and communication technology and allow reproducing real low and medium voltage grids and the test of different communication technologies (Ethernet, PLC, GSM...) for Smart Grids. As with real network types in densely populated areas, the substations are connected in a ring network (Figure 4). Each substation is equipped with six power take offs, which can be used for any kind of

A NNUAL R EPORT 2010

low voltage networks with cable lengths up to two kilometers (NAYY-J 4x150 mm2 and NAYY-J 4x240 mm2). Furthermore there is test equipment to reproduce low voltage ride through tests for medium voltage grids (up to 4 MVA at 10kV). Using an inductive voltage divider, voltage dips can be realized for residual voltages between 0% and 100% (up to several minutes). It allows the investigation of components in the case of grid faults, without negatively influencing the TSO’s medium voltage grid. These kind of tests were actually developed in order to verify grid connection conditions for wind energy plants (IEC 61400-21). The increasing development of decentralized electricity generation will expect this kind of examinations for future applications (for example battery storage systems, large-scale PV installations, cogeneration units...).

Test Center

Medium Voltage Switchgear

STAWAG

Measurement LVRT

Ring-1

Ring-2

110 kV/10 kV

Hall

LVRTContainer

Power Supply 1250 kVA

Power Supply 1250 kVA

400 kVA

630 kVA

630 kVA

1250 kVA

Grid Simulator

Battery Test Bench

630 kVA Power Supply

Institute for High Voltage Technology

Transformer Station

Regen Unit

Flexible Connections

Fig. 4: Medium voltage network in the test bench

Contact

Dipl.-Ing. Bernhard Fuchs [email protected] +49 241 80-93038

Dr.-Ing. Stefan Federlein [email protected] +49 241 80-94959

Dipl.-Ing. Mathias Knaak [email protected] +49 241 80-94909

A NNUAL R EPORT 2010

133

Institute for High Voltage Technology

EEIM John Neal Award 2010 Andrey Mashkin, M. Sc. has received the European Electrical Insulation Manufacturers’ John Neal Award 2010. The 5 000-euro prize is given for outstanding investigation accomplished in the field of electrical insulation materials and systems. His paper entitled "Analysis on the Electrical Properties of Syntactic Foam under Various DC Field Stresses"was judged by the award committee as the best application overall to meet the award criteria for technical content, practical application, originality and clarity of presentation. Investigation focuses on the analysis of the electrical behavior of syntactic foam, a new low-weight insulation material, which has been developed in the co-operation between RWTH Aachen University and Philips Medical Systems (PMS).

Syntactic Foam

134

The design of several high voltage applications requires an exceptionally efficient and compact insulation system. Especially in the rotating or mobile equipment the requirements on light insulation materials cannot be fulfilled by conventional dielectrics. Syntactic foam, a special kind of foam-like material, provides the best combination of compactness, light-weight and high breakdown strength for this kind of applications. In syntactic foam the voids are not made by blowing agent but exist inside of hollow micro filler, which is embedded in the polymeric basis. The Investigation focus on syntactic foam, consisting of microscalic hollow glass microspheres, which are mixed into an epoxy resin matrix. The resulting compound is cured to rigid syntactic foam, which density is reduced by 40% compared to epoxy resin. The mechanical and electrical properties of the syntactic foam can be tailored for each kind of application by varying the filler size, type

A NNUAL R EPORT 2010

and volume percentage. Fig. 1 shows a Scanning Electron Microscopy (SEM) picture of a cutting area of the material.

Fig. 1: SEM picture - inner structure of syntactic foam

Aim of the Study The electrical Breakdown process in syntactic foam under short time ac and dc field stress has been analyzed already at IFHT in the previous investigations. The determined high breakdown field strength, which ranges 20-40 kV/mm for ac and dc stress, enabled the use of syntactic foam in high voltage applications. To broad the dc application area of syntactic foam, several further processes in the material have to be known. In many applications the high electrical dc field is applied as a step function. This causes transient processes inside the insulation system. Furthermore, during long term stress an ageing process develops, which has not been investigated jet. The awarded study focuses on the further use of syntactic foam in dc components and applications. For characteristic periods- at the begin and during the continuous stress of the material the field stress of each material component as well as eclectic properties of the whole material are analyzed. Finally, the consideration of the microscopic field distribution and the gained breakdown data are used to extend the breakdown mechanism models. The award-winning paper

EEIM Award 2010

Download The synopsis of the study is available for download on www.ifht.rwth-aachen.de/ EEIM JNA2010 synopsis.pdf.

The Ceremony

Results overview In the experiment, different types of syntactic foam are stressed with step function voltage, causing transient processes to take place inside the material. The materials’ conductivities are determined by measuring the volume current. Additionally, the dependency of the breakdown field strength on the transient properties of syntactic foam is determined. The consideration of the microscopic field distribution inside syntactic foam during the transient process is used to analyze and extend the breakdown mechanism model of syntactic foam. The breakdown field strength in the long term test is investigated for durations up to 5000 h. The time to breakdown for field stresses of 16, 18 and 20 kV / mm is measured. The combination of the determined conductivities provides all the required data to analyze the field distribution in insulation systems containing syntactic foam. The breakdown properties of syntactic foam enable the long term use of the material in high voltage components. The determined values can be used as design criteria to buildup lightweight and efficient insulation systems.

The award was presented on October 1, 2010 in Aachen on the occasion of the 90th anniversary of Rogowski-Institute, in the presence of the dekan of faculty, Award Committee members, a representative of the generator design team from Philips Medical Systems, professors and IFHT alumni. On behalf of his team, the award winner thanked the award Committee for the high honor.

Fig. 2: Award Committee members and awardwinning research team: f.l.t.r.: Berno Habertür (EEIM), Michael Chapman (EEIM), Anja Strauchs (IFHT), Andrey Mashkin (IFHT), Patrick Veluzat (EEIM), Nobert Rudat (EEIM), Jörn Podlazly (PMS), Prof. Armin Schnettler (IFHT)

Contact M.Sc. Andrey Mashkin [email protected] +49 241 80-94939

A NNUAL R EPORT 2010

Institute for High Voltage Technology

provides a basis for the use of syntactic foam in the broad field of dc applications, which ranges from high-voltage, direct current (HVDC) components to high voltage DC test and lab equipment.

135

136

Institute for High Voltage Technology

A NNUAL R EPORT 2010

Institute for High Voltage Technology

Staff of IFHT 2010

137

A NNUAL R EPORT 2010