48V Recuperation Storage Based on Supercaps for Automotive Applications

Transcription

1 EVS28 KINTEX, Korea, May 3-6, V Recuperation Storage Based on Supercaps for Automotive Applications Andreas Baumgardt 1, Dieter Gerling 1 1 Universitaet der Bundeswehr Muenchen, Werner-Heisenberg-Weg 39, Neubiberg, Germany, andreas.baumgardt@unibw.de Abstract This paper suggests and investigates a 48V recuperation storage for hybrid vehicles based on supercaps. A recuperation system composed of 48V components provides a low-cost solution with a promising proportion of fuel saving and system costs. First, a complete system model of a HEV is presented in order to simulate the fuel consumption. The model includes a supercapacitor model as well as a battery model. The storage models are validated in a laboratory setup. Next, both storage technologies are characterized and investigated. Finally, an extensive comparison of both technologies is made regarding a specific energy storage device each. Both storage devices are subjected to a load which derived from the power flow that occurs within the driving cycles NEDC and WLTC. The supercap solution acquires promising results, considering fuel savings and costs in particular. Keywords: recuperation, supercapacitor, ultracapacitor, energy storage, 48V power net, hybrid electric vehicle 1 Introduction Due to decreasing CO 2 limits, the automotive industry gives more and more priority to the development of battery electrical vehicles (BEV) and particularly hybrid electrical vehicles (HEV). Withal it is important to hold down costs and keep vehicle prices affordable. A recuperation system based on 48V components provides a low-cost solution to save CO 2 emissions. Due to a system voltage beneath 60V there is no need for isolated high voltage components which are expensive and complex in comparison. With a four times higher recuperation power than a conventional 12V recuperation system, a perfect proportion of CO 2 savings and system costs can be reached. To absorb the recuperated energy, several storage technologies are in discussion. Presently, batteries and supercapacitors are the most promising ones [1]. As batteries already cover a wide range of applications, automotive applications in particular, supercapacitors as energy storage however can be found in some special applications such as offshore wind farms, buses or motorsport. In addition, research on utilization in hybrid or fuel cell vehicles can be found [2, 3]. This paper suggests and investigates a recuperation storage for hybrid vehicles based on supercapacitors in comparison to battery systems. First, a complete system model of a HEV is presented in order to simulate the fuel consumption. In particular, the automotive power net demand is considered in simulations. The model includes a supercapacitor model as well as a battery model. Both models are validated in a laboratory setup. The temperature behaviour of the storages is considered in addition. Next, both storage technologies are characterized and investigated. Finally, one specific storage of each EVS28 International Electric Vehicle Symposium and Exhibition 1

2 technology is regarded and compared to the other one. 2 Simulation models 2.1 Powertrain configuration HEVs include an internal combustion engine (ICE) as well as an electrical drive (ED). A parallel configuration is chosen for the investigations conducted in this paper, as proposed in [4] and presented in Figure 1. The electrical machine is placed between ICE and transmission, which allows a wide range of operating options. Connecting ED and ICE while the transmission is disconnected, the ED can operate as generator or as ICE starter. On the other hand, the ED can be used for recuperation and electric driving with closed transmission clutch and opened ICE clutch. Figure 1: Powertrain configuration of a parallel hybrid electric vehicle (copied from [4] with permission of the author) Basically, hybrid electric vehicles can be simulated using forward-facing models (dynamic modelling) or backward-facing models [5]. Using the forward-facing method, a driver model including a vehicle speed control loop is necessary. The model is a more precise replication of reality and preferred for hardwarein-the-loop applications or control strategy development. Backward-facing models are much faster and include more simple models such as efficiency maps. Figure 2 presents an overview of the model assembly. As the efficiency data of the main components are available and the focus is not on the control strategy, the backward-facing method is chosen for the outer model, including the vehicle model and the ICE-model. As the operation of the recuperation system is dependent on state variables like state of charge of the storage (SoC) or temperature of the components, the recuperation system is realized with a forward-facing model including a control strategy. Velocity Vehicle model P el,pn M a n Recuperation system Control strategy I st U oc, SoC, T st + M ed Electrical drive P el,ed + P el,st Energy storage ICE Model Figure 2: Overview of vehicle model and recuperation system model 2.2 Vehicle model The vehicle model accounts for the losses occurring in the vehicle except the driving unit losses, i.e. recuperation system and ICE losses. A velocity profile representing a driving cycle is given as input to the vehicle model. This allows the calculation of translational and rotational energy of the vehicle. The sum of both is defined as E vehicle. The first derivate with respect to time represents the current power demand. Adding the aerodynamic drag losses P aero, the rolling friction losses P roll, the transmission losses P trans and the power consumption of auxiliaries P aux, the power demand as seen from the driving unit P a (represented by the torque M a and the shaft speed n) can be calculated as: Pa Evehicle Paero Proll Ptrans Paux (1) The parameters such as weight, drag coefficient and rolling resistance used for energy and power loss calculation define the vehicles characteristics. Thus different vehicle types such as small cars or SUVs can be used for simulation. In the following, a mid-size executive car is considered. This paper considers two driving cycles. First, the New European Driving Cycle (NEDC) is the current European standard cycle for CO 2 determination including urban and interurban driving up to 120km/h. The velocity profile and the calculated power demand P a are presented in Figure 3. Moderate velocity slopes result in a wellarranged power profile. While a positive power demand describes a need of power from the ICE or the recuperation system, a negative value allows energy recuperation. Second, the Worldwide harmonized Light duty Test Cycle (WLTC) is a driving cycle more close to reality and is under development in order to M ice n Fuel consumption EVS28 International Electric Vehicle Symposium and Exhibition 2

3 power nets [7]. It consists of an electrical drive including an electrical machine and the corresponding power electronics, an energy storage device and a control strategy. Each component is described in detail in the following. Figure 3: NEDC velocity profile (top) and calculated power demand P a seen from the driving unit (bottom) Figure 4: WLTC velocity profile (top) and calculated power demand P a seen from the driving unit (bottom) replace the NEDC by a new global test procedure [6]. Figure 4 presents the corresponding velocity and power demand profile. Higher dynamics of velocity as well as power are recognizable. 2.3 Recuperations system The recuperation system converts mechanical power into electrical power (generator operation) and vice-versa (motor operation). In a parallel hybrid configuration, the electrical drive power can be simply added to the ICE power. On the electrical side, the power demand of the automotive power net P pn has to be covered in addition. The power can be either supplied by the energy storage or by the electrical machine. The power net demand includes all electrical loads in the vehicle, e.g. electrical power steering and comfort systems. There are two models included, each representing a recuperation system for different applications: The first model includes a 12V automotive generator as a simple recuperation system covering the automotive power net demand P pn. This model is used to generate reference simulation results of a conventional car. The second model describes the 48V recuperation system in detail. The system is based on a voltage of 48V, which is to be introduced in automotive Electrical drive This component includes an induction machine with concentrated windings and the corresponding power electronics. With a voltage level of 48V an electrical machine power of 10kW to 20kW is feasible. The electrical machine is presented in [4]. The model includes an efficiency map, covering motor and generator operation including the power electronic losses. The efficiency map is the result of a finite element method calculation. The maximum power of the electrical drive also constitutes the upper power limit of the recuperation system Control strategy A control strategy is necessary to control the torque of the electrical machine in order to keep the components in its limits and minimize the fuel consumption. There is a lot of research published about optimizing the control strategy in view of the fuel consumption [8]. The focus of this paper is on the energy storage investigation, whereas a simple, rule based control strategy is sufficient and chosen. The main rules are: Recuperate as much power as possible if the vehicle power demand P a is negative, i. e. velocity is decreasing. Supply power demand of the automotive power net. Support ICE with a defined power if possible (can also be set to zero). As the energy storage is the main examination object, it is assumed that the variables of the electrical drive fit into their limits. Therefore the control strategy only ensures the storage compliance. This includes a current, voltage and temperature limitation. A block diagram of the control strategy is presented in Figure 5. The electrical machine model converts the vehicle power demand P a, presented as the torque M a and the shaft speed n, into electrical power. Adding the automotive power net demand P el,pn, the resulting power is the potentially available electric power P el,av as seen from the storage device. EVS28 International Electric Vehicle Symposium and Exhibition 3

4 P el,pn M a n Electrical drive model I st U esr I st U esr P el,av Voltage calculation Voltage limitation Current limitation I st U dc U oc U dc + - U oc U oc or SoC Figure 5: Control strategy for current controlled operation With the open circuit voltage of the storage U oc as additional input, the target current for the storage I st is calculated considering the voltage, current and temperature limits. If U oc is not available, e.g. in case of a SoC determination based on current integration, the SoC can be used as input value to determine the storage power and the target current I st. In that case, SoC limits have to be defined. The temperature rise is only dependent on the power losses. This implies a dependency on the current. Thus, the temperature limitation can be integrated in the current limitation. The energy storage is driven in current controlled operation. The determination of I st based on the storage model is presented in the next chapter Energy storage Supercaps and lithium ion batteries are the most promising storage technologies for usage in hybrid and electric vehicles [1]. In the following, a simplified model for each storage technology is presented. Both models are sufficiently accurate for use in driving cycle based simulations and allow an easy implementation. Basically, storage can be modelled as a current integrator. The internal losses are described with an equivalent series resistance. This is valid for both, supercap and lithium ion battery, merely the voltage response is different (described hereafter). As regarding driving cycles with durations wide less than one hour, the selfdischarge, usually represented by a parallel resistance, can be neglected. Figure 6 presents the simplified equivalent circuit diagram of a supercapacitor (left) and a lithium ion battery (right) as described. The voltage response of supercapacitors can be easily modelled utilizing the differential equations of the equivalent circuit. This includes an ideal capacitor and the series resistance. The modelling of the dynamic behaviour of batteries T st Figure 6: Equivalent circuit diagram of a supercapacitor (a) and a lithium ion battery (b) is more complex in comparison. [9] presents a dynamic battery model utilizing a measured voltage curve and a PT1-element. This model is used for investigations in this paper. Regarding both storage models, the requested power for charging and discharging can be described as followed: Udc Uoc Prequested Udc I Udc ( ) (2) R The quadratic equation gives two solutions for the desired voltage U dc, whereas only one is reasonable. U dc 2 Uoc Uoc 4 Resr Prequested (3) 2 Now, the calculated voltage can be limited to minimum or maximum values of the storage, which results in U dc,lim. The current can be calculated as followed: I U (a) U dc,lim oc calculated (4) R esr Adding a current limitation to I calculated, the resulting current is the given current of the control strategy I st in current regulated operation. The storage s power loss P st,loss is the input parameter for the thermal model, which is presented next. P R I (5) 2 st, loss esr st The temperature of the storage is estimated using a thermal model to limit the temperature rise. A simple thermal network is built up, including thermal resistance and thermal capacity of the respective storage. The parameters of both supercap storage and lithium ion battery are esr (b) EVS28 International Electric Vehicle Symposium and Exhibition 4

5 experimentally determined. A precise calculation of temperature behaviour in and around the storage would require a complex thermal network. But the results of the simple model are accurate enough for a temperature estimation in order to detect and prevent overheating of the storage device. 2.4 ICE The internal combustion engine (ICE) has to provide the sum of vehicle and recuperation system power demand. The more power is covered by the recuperation system, the lower is the resulting fuel consumption. The ICE is modelled with an efficiency map of torque and motor speed. A typical mid-size ICE for mid-size cars is used. As only regarding relative consumption values, detailed engine parameters and model accuracy are not important. The recuperation system is also able to increase the load of the ICE in order to operate the same with a better efficiency. This is part of a control strategy optimization and can be investigated in future work. 3 Validation of storage models 3.1 Experimental setup The models of both supercap and lithium ion storage are validated with an experimental setup. A bidirectional DC-source is connected to the device under test. The DC-source is able to drive a test current of ±600A and provides high dynamics with a 10% to 90% current rise in 3ms. This allows an accurate replication of a timecurrent profile. The voltage and current of the storage device are measured with a power meter. Regarding temperature measurement, up to eight thermocouples are attached to the device under test. 3.2 Supercap storage The supercap storage is realized as a series connection of several high capacity cells suitable for automotive applications. First, a dynamic power profile is applied to the storage for 200s. Figure 7 presents the current curve, the simulated and measured voltage response as well as the relative deviation of measured to simulated voltage. The deviation can be explained by an inaccuracy of the output current of the DCsource first and an inaccuracy of the upper voltage limitation second. Nevertheless, the Figure 7: Validation of supercap model with a dynamic power profile for 200s relative deviation doesn t exceed 1.5% in a time period of 200s. Common driving cycles cover a time range of 10 to 60 minutes. Furthermore, the energy storage is drained to the lower voltage or SoC limit several times during common driving cycles. Thereby the relative deviation of simulation and reality is reset to zero. As a result, the duration without a deviation reset extremely rarely exceeds 200s considering common driving cycles. Figure 8 presents the power profile of the NEDC and the corresponding voltage response comparison. The lower voltage limit is set to 36V. An inaccuracy of the lower voltage limit in the experiment is recognizable. Except for some peaks as a result of system inertia, the relative deviation stays below 1.5%. Figure 8: Validation of supercap model with NEDC EVS28 International Electric Vehicle Symposium and Exhibition 5

6 Next, the temperature model is investigated. Figure 9 shows both simulated and measured temperature of the NEDC profile. The measured temperature reacts more slowly than corresponding simulation. This can be explained by the method used for temperature measurement. As a supercapacitor cell is a closed device, it is hardly possible to measure the temperature inside the cell. The thermocouples are attached to the cell case. This leads to the observed delay. Nevertheless, the simulation is accurate enough to estimate the temperature behaviour of the storage device. Figure 9: Validation of temperature model with NEDC Considering the validation of voltage and temperature behaviour, the supercap model is seen as accurate enough for the present application. 3.3 Battery Storage The regarded battery storage consists of several cells connected in series. The nominal voltage lies far below 48V, but can be easily scaled up to the voltage of 48V required for simulation. A current profile with corresponding voltage validation is presented in Figure 10. The dynamic behaviour modelled with a PT1-element fits well to the measured voltage. Beyond, this can be optimized by a more precise parameter adjustment. Yet, the relative deviation remains below 3%. Further measurement results for a model validation are presented in [9]. The temperature of the battery is modelled with the same model as presented for the supercap storage, fed with the measured thermal resistance and capacitance of the battery device. 4 Investigation To give a basis for the comparison of two storage technologies and support the dimensioning of a recuperation storage, some essential investigations based on the presented simulation model are made. The results are presented as fuel Figure 10: Validation of battery model with a current profile for 200s savings, which are calculated by the simulated fuel consumption of a conventional car fc conv (see chapter 2.3) and the simulated fuel consumption of the same car with integrated 48V recuperation system fc recu : Fuel savings fc fc fc conv recu (6) conv For all investigations, the storage s SoC at the start and the end of simulation is equal. First of all, the maximum power of the recuperation system is considered. Figure 11 presents the fuel savings dependent on the power assuming an ideal storage with infinite capacity and without losses. This allows a determination of the upper limit of possible fuel savings. A saturation of the fuel savings can be seen for both driving cycles. An increase of the maximum system power above 20kW has merely no influence on fuel consumption improvement. 48V systems are typically located in the area between 10 and 20kW [7]. The dynamics and velocities of the NEDC are much lower in comparison to the Figure 11: Saturation of fuel savings for both driving cycles NEDC and WLTC assuming an ideal storage EVS28 International Electric Vehicle Symposium and Exhibition 6

7 WLTC. This leads to lower losses and thus higher possible fuel savings. The rippled curve of the NEDC simulation below a power of 8kW can be explained by the artificial composition of this driving cycle. The first part of the NEDC represents urban driving and is repeated four times (see Figure 3). As a result of the homogenous construction of this part, the power steps have a four times higher impact as with inhomogeneous cycles and induce the curve in the way indicated. In the following a system power between 10 and 20kW is considered. The recuperated energy is either used for the coverage of the power net demand or utilized as support for the ICE in order to reduce fuel savings. A percentage splitting of both is presented in Figure 12. As under NEDC test conditions only indispensable power net loads are activated, the amount of power net demand is lower in comparison to the WLTC. 100% Figure 12: Available recuperation energy of a 48V recuperation system in respect of the automotive power net demand In the following, both storage technologies are characterized and investigated. 4.1 Supercapacitors Based on the principles of conventional capacitors, supercapacitors show a fundamental different behaviour than batteries. The differences are discussed in [1] and [10]. Summarized, supercaps provide the following characteristics: Very high power density. High currents for discharging and especially for charging are feasible. Lower energy density compared to lithium ion batteries. A lower energy content of one decade can be estimated. Lower internal series resistance compared to batteries, which results in lower losses for equal currents. 80% 60% 40% 20% 0% NEDC WLTC Remaining energy Power net demand Lower cell voltage, which results in a higher number of series connected cells. Lower potential risks lead to a higher safety. Higher cycle lifetime compared to batteries. Considering the application in HEVs, high power has to be absorbed to increase fuel savings. This makes supercaps despite of their low energy density a feasible alternative to batteries. The usable energy content of the supercap storage is the limiting parameter considering the current application. Figure 13 presents the fuel savings in dependency on the energy content. Capacity and internal series resistance are adapted to the corresponding energy content according to the characteristics of commercially available high capacity cells. The maximum possible fuel savings for an ideal storage are plotted in addition (dotted lines). The fuel savings at the WLTC are saturated earlier in comparison. By contrast, the fuel savings of the NEDC increase linear for energy content above 60Wh. The last slope of velocity regarding the NEDC velocity profile (Figure 3) provides significantly more energy than all other slopes. To absorb the available energy, large energy content is necessary. Figure 13: Fuel savings versus energy content of a supercap storage and the corresponding maximum value for an ideal supercap storage Energy content between 10Wh and 20Wh represents a convincing compromise of costs, weight and space on the one hand and fuel savings on the other hand. The voltage range of supercapacitor storages covers a wider area in comparison to battery storages. The usable energy of a capacitor rises with increasing voltage range. Assuming equal load current, the resulting power of a supercap storage is dependent on the voltage, i. e. the SoC. Thus, the power losses decrease with increasing SoC. Considering this, the control strategy can be optimized to improve efficiency. 4.2 Lithium ion batteries Lithium ion batteries cover a wide range of applications with different cell parameters. The EVS28 International Electric Vehicle Symposium and Exhibition 7

8 Fuel savings (%) parameters such as energy density, power density, safety and calendric and cycle lifetime can be optimized for different applications. However it is hardly possible to optimize all parameters at the same time. For the usage in HEVs, high power cells are preferred [1]. Yet, the bottleneck is the charging power, i.e. the charging current. Usually, the maximum currents for charging and discharging are given as a multiple of the available charge, the c-rate. High power cells provide maximum discharge currents of 60C or more. However the maximum charging current rarely exceeds 4C. Enforcing higher charging currents, the anode is not able to accommodate the lithium ions quick enough and lithium plating occurs. The available charge and lifetime will decrease rapidly [11]. According to that, Figure 14 presents the fuel savings in dependency on the c-rate of the charging current for a battery pack of less than 1kWh. The curves saturate to the drawn maximum fuel savings for an ideal storage. A charging current of 4C admits only about two thirds of the maximum possible fuel savings. To improve fuel savings of the battery solution, it is necessary to increase charging current either by a parallel connection of battery packs or an increase of nominal charge. than the battery storage. The cost comparison is gathered for small quantities. First the fuel savings of the mentioned storage packs are calculated. The results are presented in Figure 15. The fuel savings for the NEDC are similar for both storages. Regarding the WLTC, the supercap storage is more economical than the battery storage because of the more dynamic profile resulting in higher power occurrence NEDC WLTC Supercap Battery 4C Figure 15: Fuel savings comparison of energy storage packs for NEDC and WLTC Next, the state of charge of both storages during WLTC is calculated and plotted in Figure 16. The energy content of the supercap storage is widely used. In contrast, a very small part of less than 3% of the battery storage is in use. This is a result of the limited charging current. The lower SoC limit is set to 50% for this simulation. The battery storage seems widely oversized. However, choosing a smaller battery storage would result in a lower charging current and accordingly in a similar usage behaviour and lower fuel savings in addition. Figure 14: Fuel savings versus battery s maximum charging current for both driving cycles 5 Comparison of specific storage packs After regarding both storage technologies in general, two specific energy storage devices are compared to each other. First, a high power battery pack is chosen. The usable energy content is a little less than 1kWh. As specified by the manufacturer, a maximum charging current of 4C is feasible. Next, a supercap storage is fitted to the same weight as the battery storage. The usable energy content lies between 10 and 20Wh. The supercap storage dimensions exceed the battery pack dimensions by about 50% whereas the price of the supercap storage is 20% lower Figure 16: State of charge for supercap (top) and battery (bottom) storage during driving cycle WLTC To get an overview of the temperature behaviour of both storages, the temperature curve during the WLTC is presented in Figure 17. Regarding thermal characteristics, the thermal capacity of the supercap storage is higher compared to the battery pack, whereas the thermal resistance is equal. The supercap storage has to deal with higher currents on the one hand, but profits by its lower internal resistance on the other hand. As a result, both storages show a similar behaviour. The EVS28 International Electric Vehicle Symposium and Exhibition 8

9 temperature stays below critical values of 60 C in both cases. References [1] A. F. Burke, "Batteries and Ultracapacitors for Electric, Hybrid, and Fuel Cell Vehicles," Proceedings of the IEEE, vol. 95, pp , [2] D. Rotenberg, A. Vahidi, and I. Kolmanovsky, "Ultracapacitor Assisted Powertrains: Modeling, Control, Sizing, and the Impact on Fuel Economy," Control Systems Technology, IEEE Transactions on, vol. 19, pp , Figure 17: Temperature behaviour for supercap and battery storage during WLTC Finally, both storage packs gather convincing results operating as recuperation storage. Although lithium ion batteries are already established in automotive applications, the presented results show that a supercap solution is on par with a battery solution and thus represents a promising alternative to lithium ion batteries in HEVs. The supercap solution is more economic considering fuel consumptions under similar conditions, especially in more dynamic driving cycles more close to reality. However, the battery storage provides a huge energy content, which can be used to enable additional features, e.g. electrical driving for slow velocities or a plug in hybrid upgrade. 6 Conclusion This paper suggests and investigates a supercap storage for a 48V recuperation system. A complete system model is presented to simulate the fuel consumption of HEVs. The model includes an energy storage model for a supercap storage as well as for a lithium ion battery. Both storage models are validated with an experimental setup. The mentioned storage technologies were characterized and investigated in general. The results could serve as support for design and dimensioning of similar recuperation systems. Finally, an extensive comparison of both technologies is made regarding a specific energy storage device each. The supercap solution achieves promising results particularly considering fuel savings. Future work will concentrate on the construction of a test bench including the complete 48V recuperation system for further investigations and model validation. Furthermore, additional storage technologies such as lithium ion capacitors or mixed storages will be regarded. [3] C. Ashtiani, R. Wright, and G. Hunt, "Ultracapacitors for automotive applications," Journal of Power Sources, vol. 154, pp , [4] A. Patzak and D. Gerling, "Design of an automotive 48 V integrated starter-generator on the basis of an induction machine with concentrated windings," in Electrical Machines and Systems (ICEMS), th International Conference on, 2014, pp [5] G. Mohan, F. Assadian, and S. Longo, "Comparative analysis of forward-facing models vs backwardfacing models in powertrain component sizing," in Hybrid and Electric Vehicles Conference 2013 (HEVC 2013), IET, 2013, pp [6] C. Bruenglinghaus and J. Winterhagen, "CO2 Limits Determine Future Directions," ATZ worldwide, vol. 113, pp. 4-7, [7] T. Doersam, S. Kehl, A. Klinkig, A. Radon, and O. Sirch, "The New Voltage Level 48 V for Vehicle Power Supply," ATZelektronik worldwide, vol. 7, pp , [8] P. Pisu and G. Rizzoni, "A Comparative Study Of Supervisory Control Strategies for Hybrid Electric Vehicles," Control Systems Technology, IEEE Transactions on, vol. 15, pp , [9] A. Baumgardt, F. Bachheibl, and D. Gerling, "Utilization of the battery recovery effect in hybrid and electric vehicle applications," in Electrical Machines and Systems (ICEMS), th International Conference on, 2014, pp [10] J. M. Miller, "Energy storage technology markets and application's: ultracapacitors in combination with lithium-ion," in Power Electronics, ICPE '07. 7th Internatonal Conference on, 2007, pp [11] J. Fan and S. Tan, "Studies on Charging Lithium-Ion Cells at Low Temperatures," Journal of The Electrochemical Society, vol. 153, pp , EVS28 International Electric Vehicle Symposium and Exhibition 9

10 Authors Andreas Baumgardt was born in Muenchen, Germany, in He received the M.Sc. degree in Mathematical Engineering from the Universitaet der Bundeswehr Muenchen in Since then he is working at this university on his Ph.D. thesis in the fields of automotive power nets. Dieter Gerling was born in Menden, Germany, in He received the Dipl.-Ing. degree in electrical engineering from Technical University in Aachen (RWTH) in He worked for many years in the Philips research laboratory in Aachen, Department of Electrical Drives and Power Electronics, and with Robert Bosch GmbH in Buehl, Germany, where he was a department manager for advance development of New Systems and Electrical Drives. He received his Dr.-Ing. Degree from RWTH in 1992 and since 2001, he is a professor at the Chair of Electrical Drives and Actuators at Universitaet der Bundeswehr Muenchen. EVS28 International Electric Vehicle Symposium and Exhibition 10