35th International Electric Vehicle Symposium and Exhibition (EVS35)

Oslo, Norway, June 11-15, 2022

Integrated Optimization of Fuel Cell Drive System Design

and Energy Management
Adrian Braumandl1, Guillaume Paris1, Katharina Bause1, Albert Albers1
1
IPEK – Institute of Product Engineering at Karlsruhe Institute of Technology (KIT), Kaiserstr. 10,
76131 Karlsruhe, Germany, [email protected]

Summary
This work describes a method for the automated design of fuel cell drive systems, enabling developers to identify
ideal drive system configurations as well as suitable energy management strategies using multi-objective
optimization. The authors focus on the integration of the energy management strategy optimization in the drive
system design specification process and derive suitable approaches. These approaches are then evaluated
regarding their optimization performance and the optimization results.

Keywords: fuel cell, fuel cell vehicle, optimization, research, simulation

1 Introduction
Legislators in multiple countries passed regulations aiming at a reduction of the greenhouse gas emissions of the
transport sector [1, 2]. In addition to a lower overall traffic volume, an increased vehicle electrification is an
important aspect regarding the reduction of emissions. Battery-electric vehicles (BEV) as well as fuel cell electric
vehicles (FCEV) offer locally emission-free mobility. While BEVs to this day suffer from low perceived range
and comparably high recharge duration, FCEVs combine the advantages of low refuelling duration and high
operational range with the efficiency of electric traction motors. Thus, fuel cell drive systems are especially
advantageous in applications for long-distance operations like semi-trucks [3, 4] although the first mass-produced
and commercially sold FCEV has been introduced with the passenger car Toyota Mirai in 2014.
Proton exchange membrane fuel cells, the type of fuel cell commonly used in mobile applications, have some
insufficiency regarding dynamics and cold start behavior. Furthermore, they do not enable regenerative braking.
Therefore, fuel cell drive systems are hybridized by integrating a rechargeable electrical energy storage system
(REESS), typically a battery. Because of this, fuel cell drive systems have some degrees of freedom during
operation, e.g. the distribution of power and energy deployment between fuel cell and REESS, which are
controlled by an energy management strategy (EMS). A properly optimized EMS ensures a high efficiency of the
drive system and thus reduces the energy consumption of the vehicle. This paper presents and compares different
approaches towards the integration of the EMS optimization during the fuel cell drive system design. The EMS
used is an equivalent consumption minimization strategy (ECMS).

35th International Electric Vehicle Symposium and Exhibition 1

2 State of the Art
This chapter introduces the fundamentals and current state of the art of mathematical optimization methods,
metrics to evaluate optimization results, electric drive system design approaches and energy management
strategies.

2.1 Optimization
A multi-objective optimization problem is characterized by the presence of several objective functions, which
must either be minimized or maximized. Usually, each objective is contradictory with at least one of the other
objectives. Pareto optimality describes a state, when no individual solution can be further optimized towards one
objective function without decreasing at least one other objective function. These solutions are also described as
non-dominated, see Figure 1.

Figure 1: Pareto front, non-dominated and dominated solutions [5]

The simultaneous fulfilment of multiple objectives in one problem gives rise to a variety of such non-dominated
solutions. Classical optimization methods suggest transforming the initial multi-objective optimization problem
into a simple single-objective optimization problem, which minimizes the computation effort but requires the
introduction of more subjective information to bias the solutions. More sophisticated approaches aim to find all
solutions that belong to the Pareto front. These approaches are usually devised based on the imitation of biological
or natural processes. Examples are Particle Swarm Optimization [6] and evolutionary algorithms, most notably
genetic algorithms (GA) [ISS 7-10].

2.2 Optimization Quality

For the comparison of the multi-objective optimizations, a criterion is needed that objectively determines the
quality of the results. The hypervolume is a suitable quality indicator as no prior knowledge of the Pareto front
is needed. The hypervolume is calculated by comparing the Pareto front to a pre-determined closed hypervolume
space. This hypervolume space is defined by two reference points spanning either a plane or three-dimensional
space. An arbitrary number of sample points is uniformly scattered in the hypervolume space, forming a net.
Each sample point is checked whether it is dominated by the Pareto front. The hypervolume is the ratio of sample
points dominated by the Pareto front compared to the total number of sample points, see Figure 2. A higher value
typically indicates a higher quality of the Pareto front as long as the reference points and number of sample points
have been chosen properly for the given optimization problem. [12]

35th International Electric Vehicle Symposium and Exhibition 2

 Figure 2: Hypervolume [11]
Other possibilities to compare and evaluate multiple Pareto fronts consist of a direct visual comparison (only
suitable for two-dimensional Pareto fronts), the comparison of the number of individual solutions on the Pareto
front as an indicator of convergence as well as the comparison of the crowding distances of the individuals on
the Pareto front to evaluate their distribution.

2.3 Drive System Design

Because the components in fuel cell drive systems, aside from the fuel cell, are comparable to those in battery
electric drive systems, this chapter first provides an overview of the state of research on design and optimization
methods for battery electric drive systems before focusing on FCEV drive systems. As discussed in Chapter 2.1
of this work, numerous optimization methods can be used to perform single-objective or multi-objective
optimization. [13]
For the design of battery electric vehicles, multiple single-objective optimization approaches have been
developed. Aside from [14] and [15], the Electric Vehicle Identification (EVID) released in [16, 17] is a prominent
methodology to determine the best drive topologies, component types and component characteristics of BEV.
Several design parameters are modified and adjusted in order to enable customer-oriented design of electric drive
systems in the early stage of product engineering. After that, the design objectives are assessed using a weighted
sum function, with weighting variables determined from specific usage scenarios. [13] presents a multi-objective
optimization-based holistic strategy. The advantages of multi-objective optimization are used to define a method
for assessing the potential of several battery electric vehicle designs in terms of energy efficiency, driving
dynamics, and cost.
In recent years, a large variety of design methods for FCEVs and fuel cell drive systems have been published.
The authors of [18] describes one of the first systematic design methodologies for FCEV drive systems. Further
research used several optimization methods, such as particle swarm optimization or dynamic programming, to
focus on component sizing or cost-reduction in FCEV drive systems, and compared different optimization
methods [19, 20]. Previously published design approaches for FCEV drive systems were enhanced by including
the optimization of the energy management strategy in [21]. The design of fuel cell drive systems and associated
energy management strategies is still a major focus of contemporary research. New methodologies for conceptual
design of FCEV drive systems are given in [22, 23] utilizing full factorial design to identify optimal solutions.
The author of [24] optimizes the concept design of electrified drive systems, including fuel cell drive systems,
with a focus on affordability and a mix of driving performance and low energy usage. While multiple gearbox
types were considered, the optimization procedure only looked at one drivetrain topology with a single electric
machine. A framework for optimizing FCEVs employing various energy storage systems is provided in [25]. A
number of approaches to enhancing the power management strategy were investigated. Furthermore, fuel cells
and electric energy storage systems were dimensioned in detail, also taking into account battery aging and
lifespan costs. In [26], an approach synthesizing and evaluating drivetrain topologies automatically is presented,
supporting developers of hybrid drive systems.

35th International Electric Vehicle Symposium and Exhibition 3

2.4 Energy Management Strategies
As introduced in Chapter 1, fuel cell drive systems are usually hybridized. In hybrid drive systems the energy
management strategy (EMS) determines the power deployment from the different installed energy storage
systems. EMS can be classified in the two main categories rule-based, which contain deterministic and fuzzy
based control strategies, as well as optimization-based, which contain global optimization and real-time
optimization approaches. The different approaches come with different advantages, disadvantages as well as
other implications on the EMS development process which have been presented in [27, 28]. The common goal
is typically to increase the efficiency and performance of the drive system.
The optimization of EMS has become increasingly relevant since the significant rise of hybrid electric vehicles
being developed and has also been focused by researches in the field of fuel cell drive systems [29]. The authors
of [29] already introduced a real-time capable equivalent consumption minimization strategy (ECMS) for FCEV
in 2002. The formulation of the minimization problem proposed in [30] is the following:
∑ 𝑀𝑖𝑛 𝑚̇𝑓𝑒𝑞𝑢 (𝑡), ∀𝑡 (1)
with
𝑚̇𝑓𝑒𝑞𝑢 = 𝑚̇𝑓𝐹𝐶 + 𝑓𝑝𝑒𝑛 ∗ 𝑚̇𝑓𝑅𝐸𝐸𝑆𝑆 (2)
with the fuel flow into the fuel cell 𝑚̇𝑓𝐹𝐶 and the equivalent fuel flow of the REESS 𝑚̇𝑓𝑅𝐸𝐸𝑆𝑆 , estimated by
weighing the electrical energy transferred over the boundaries of the REESS with the efficiency of the fuel cell
and 𝑓𝑝𝑒𝑛 as penalty factor. The penalty factor is a cubic function depending on the state of charge (SOC) of the
REESS – it is more expensive to utilize the battery when the SOC is low in order to reduce the deviation from
the desired SOC. A comparable approach is used in this work.
The challenge of integrating the optimization of the EMS in the drive system design process has been approached
by researchers in multiple ways. In [31], the parameters for the drive system specification of a plug-in hybrid
electric vehicle and the parameters for the fuzzy-based control responsible for power-split and gear shifting have
been optimized together in a single loop. The authors of [32] proposed a method focusing on the sizing of
ultracapacitors and batteries in FCEVs. Optimizing the utilized EMS was a part of the single-loop optimization
problem aimed at reducing initial cost, running cost, and cost associated with degradation. In [33], the EMS was
preliminary developed. The drive system design was optimized focusing on efficiency with the EMS being
applied to suit the installed component sizes.
Other researches proposed an approach based on two optimization loops. In [34], a hybrid electric propulsion
system for ships was optimized utilizing two integrated layers of optimization. The authors of [35] researched
the applicability of the different optimization methods GA, sequential quadratic programming, particle swarm
optimization and pattern search for the outer loop and dynamic programming for the inner loop when optimizing
the drive system (outer loop) and EMS (inner loop) of a hybrid heavy duty vehicle. In [36] two integrated GA
loops were used to optimize a hybrid electric drive system and the corresponding EMS.

3 Methods
This chapter introduces the drive system design approach and the different approaches towards the integration of
the EMS integration in the design approach considered in this work.

3.1 Design Approach

The authors presented an approach towards the automated design specification of FCEV drive systems in [37].
This approach considers the vehicle class for which the drive system is developed as well as further performance
requirements indicated by the expected usage profile and is depicted in Figure 3.

35th International Electric Vehicle Symposium and Exhibition 4

 Figure 3: Design approach (schematic) [37]

A multi-objective optimization heuristic based on the NSGA-II is coupled with a scalable techno-economic
FCEV model to enable developers to identify Pareto optimal configurations of FCEV drive systems satisfying
customer demands. Furthermore, the impact of e.g. modified requirements or changes to the cost of drive system
components can be assessed. The evaluation objectives considered during the optimization in this work are the
hydrogen consumption of the vehicle (determined by the drive system efficiency in different driving cycles), the
acceleration capability of the vehicle, and the drive system cost. [37, 38]

3.2 Integration of the EMS optimization

To integrate the optimization of the EMS into the design method, different approaches were elaborated. The basic
optimization of the design parameters without optimizing the EMS is referred to as A0. The first integration
approach – referred to as A1 – simply expands the design parameter set optimized with the ECMS parameters.
Thus, both parameter sets are optimized simultaneously in a single optimization loop. A flowchart of the
approaches A0 and A1 is depicted in Figure 4.

Figure 4: Single-loop optimization approaches A0 (left) and A1 (right)

Another possible way to integrate the EMS optimization is depicted in Figure 5. It consists of two consecutive
optimization loops and is referred to as A2S. While the design parameters are optimized in the first optimization
loop, another optimization loop follows which optimizes the ECMS parameters for each solution that is a part of
the design Pareto front. Thus, an ECMS Pareto front will be returned for each solution of the design Pareto front.

35th International Electric Vehicle Symposium and Exhibition 5

 Figure 5: Two-loop approach A2S
The third approach considered in this work consists of two nested loops and is referred to as A2P, see Figure 6.
The EMS optimization loop is nested in the design optimization loop. For each parameter set in the design
optimization loop an ECMS parameter set is optimized before the design parameter set is evaluated. On the
contrary to A2S, this approach only returns a single Pareto front consisting of solutions with an optimized set of
design and ECMS parameters.

Figure 6: Two-loop approach A2P

35th International Electric Vehicle Symposium and Exhibition 6

The computational time necessary to run the optimization is dependent on the size of the population (P) and the
amount of generations (G) of the GA. However, the presented approaches also differ regarding their complexity
which also impacts the computation time. The resulting complexities are given in Table 1.
Table 1: Complexity of the different approaches
Approach Complexity
A0 GP
A1 GP
A2S GP²
A2P (GP)²

4 Evaluation
The presented approaches towards the integrated optimization of drive system design specification and EMS
have been run several times in different configurations to determine robustness and provide data for an evaluation.
The design parameters considered in the optimization are the drive system topology (amount and positioning of
electric traction motors and transmissions), the fuel cell rated power, the number of battery cells and their
configuration, the electric traction motor rated power, hydrogen storage capacity as well as the transmission ratio.
The ECMS parameters optimized are the parameters defining the cubic penalty factor 𝑓𝑝𝑒𝑛 introduced in
Chapter 2.4. The chassis model used to represent the vehicle is an upper-middle class sedan.
The optimizations were run with two different settings for the GA: Either with a population of 250 for 200
generations or with a population of 125 for 400 generations. The development of the hypervolume as well as the
amount of non-dominated solutions for the approaches A0 and A1 are depicted in Figure 7 and Figure 8.

Figure 7: Development of the hypervolume (left) and the amount of solutions in Pareto front (right) – Approach A0
It can be concluded that, in both cases, the optimization seems to converge. While there is only a minor difference
regarding the hypervolume it is clear that the setting with a higher population leads to more solutions on the
Pareto front. The distribution of the solutions is more even after more generations ran (as has been also confirmed
by analysing the crowding distances). The approach with a population count of 250 running for 200 generations
is used for further evaluations. The approach A1 manages to reach same levels of hypervolume and is converging
just as quickly as the approach A0. No disadvantage could be identified despite increasing the number of
parameters that are being optimized. It is concluded that the basic parameter set of the ECMS, that is utilized for
all drive systems in the approach A0, delivers good results regarding acceleration performance of the vehicle and
efficiency.

35th International Electric Vehicle Symposium and Exhibition 7

 Figure 8: Development of the hypervolume (left) and the amount of solutions in the Pareto front (right) – Approach A1
A comparison between all approaches introduced in Chapter 3.2 has been made. The Pareto fronts are depicted
in Figure 9. To enable a visual comparison, only the two objectives acceleration capability and hydrogen
consumption (determined by the drive system efficiency) have been considered.

Figure 9: Pareto fronts for the different approaches

For the presented comparison the complexity has been normalized. Thus, A2P had a lower population and a
reduced number of generations compared to the other approaches. A2S was only completed for a single design
solution. It can be concluded, that, with a normalized complexity equalling a population of 250 and 200
generations for A1, the approaches with two loops are not able to provide results of the same quality as A1. A2P
has not converged yet while the result of A2S shows, that, after converging, the A1 Pareto front already includes
the most efficient configuration of the optimized singular vehicle.
To further differentiate the performance of A0 and A1, drive systems were also optimized considering cost and
hydrogen consumption. The results (costs normalized, cost model based on [39, 40, 41]) are presented in Figure
10. A1 performs much better considering the drive system cost. It seems the simultaneous optimization of drive
system design and ECMS parameters yields better results. A more thorough check of the single solutions showed
much smaller components being used in the A1-solutions, resulting in lower costs while still being on par or
slightly better regarding acceleration performance and efficiency.

35th International Electric Vehicle Symposium and Exhibition 8

 Figure 10: Pareto fronts comparing A0 and A1

5 Summary
In this work, an overview over optimization approaches utilized in drive system design and EMS development
for fuel cell and other hybrid drive systems is given. A method for the automated design of fuel cell drive systems
is presented. The optimization of the EMS, based on an ECMS, is integrated in this method in three different
approaches. These approaches are then evaluated regarding their optimization performance and optimization
results. It is concluded that A1 is the most suitable approach for the present challenge. Due to the lower
complexity of this single-loop approach it requires a comparably low computational effort while delivering better
results that the optimization of the drive system specification without optimizing the EMS. Therefore, it is evident
that the EMS has to be considered in the drive system design process in the presented method. None of the
presented approaches with two loops could deliver drive systems with superior performance regarding
acceleration, efficiency and costs. The next development steps for the presented approach will be concerned with
further decreasing the computational effort and increasing the level-of-detail of the optimization by considering
more drive system design parameters and more ECMS parameters. Also, only drive systems for passenger cars
have been considered. Thus, the scope will be expanded towards trucks and other commercial vehicles as fuel
cell drive system adoption is expected in these vehicles.

Acknowledgments
This paper presents excerpts of the results of the project „Methoden zur arbeitsteiligen, räumlich verteilten
Entwicklung von H2-Brennstoffzellen-Fahrzeugen in Kooperation mit China - MorEH2“. The authors are
grateful to the German Federal Ministry of Education and Research for funding this project (Funding No.:
16EMO0316).

References
[1] Regulation (EU) 2019/631 of the European Parliament and of the Council of 17 April 2019 setting CO2 emission
performance standards for new passenger cars and for new light commercial vehicles, and repealing Regulations
(EC) No 443/2009 and (EU) No 510/2011
[2] Bundes-Klimaschutzgesetz vom 12. Dezember 2019 (BGBl. I S. 2513), das durch Artikel 1 des Gesetzes vom 18.
August 2021 (BGBl. I S. 3905) geändert worden ist
[3] NPM – Nationale Plattform Zukunft der Mobilität: Arbeitsgruppe 2: Roadmap – Markthochläufe alternativer
Antriebe und Kraftstoffe aus technologischer Perspektive. Berlin, (2021)
[4] NPM – Nationale Plattform Zukunft der Mobilität: Arbeitsgruppe 4: Positionspapier „Brennstoffzelle“. Berlin,
(2021)

35th International Electric Vehicle Symposium and Exhibition 9

 [5] M. Kumar et. al., Advanced Pareto Front Non-Dominated Sorting Multi-Objective Particle Swarm Optimization
for Optimal Placement and Sizing of Distributed Generation, MDPI Energies 2016, 9, 982;
doi:10.3390/en9120982
[6] J. Kennedy, R. Eberhart, Particle swarm optimization. in Proceedings of ICNN'95-International Conference on
Neural Networks. 1995. IEEE
[7] J.D. Knowles, D.W. Corne, Approximating the nondominated front using the Pareto archived evolution strategy.
Evolutionary computation, (2000). 8(2): p. 149-172.
[8] E. Zitzler, L. Thiele, Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto
approach. IEEE Transactions on Evolutionary Computation, (1999). 3(4): p. 257-271.
[9] E. Zitzler, M. Laumanns, L. Thiele, SPEA2: Improving the strength Pareto evolutionary algorithm. TIK-report,
(2001)
[10] K. Deb. et al., A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE transactions on evolutionary
computation. (2002). 6(2): p. 182-197.
[11] W. Tang et al., Fast hypervolume approximation scheme based on a segmentation strategy. Information Sciences.
Volume 509, 2020, p. 320-342.
[12] J. Bader, E. Zitzler, HypE: An algorithm for fast hypervolume-based many-objective optimization. Evolutionary
computation. 2011, p. 45-76
[13] S. Moses, Optimierungsstrategien für die Auslegung und Bewertung energieoptimaler Fahrzeugkonzepte.
Technische Universität Berlin, Dissertation. AutoUni-Schriftenreihe: Vol. 62. Berlin: Logos-Verlag. (2014)
[14] B. Wang et al., Energy consumption analysis of different BEV powertrain topologies by design optimization.
International Journal of Automotive Technology (2018), 907–914.
[15] P. Reupold, Lösungsraumanalyse für Hauptantriebsstränge in batterieelektrischen Straßenfahrzeugen. (2014)
[16] M. Eghtessad et al., Antriebsstrangoptimierung von Elektrofahrzeugen. (2015)
[17] M. Egthessad, Optimale Antriebsstrangkonfigurationen für Elektrofahrzeuge. Technische Universität
Braunschweig, Dissertation. Schriftenreihe des Instituts für Fahrzeugtechnik: Vol. 35. Aachen: Shaker. (2014)
[18] Y. Gao, M. Ehsani, Design of Fuel Cell Powered Hybrid Vehicle Drive Train, SAE Technical Paper Series 2001-
01-2532, (2001)
[19] O. Hegazy, J. Van Mierlo, Particle Swarm Optimization for Optimal Powertrain Component Sizing and Design of
Fuel Cell Hybrid Electric Vehicle. 12th International Conference on Optimization of Electrical and Electronical
Equipment (2010)
[20] L. Xu et al., Optimal sizing of plugin fuel cell electric vehicles using models of vehicle performance and system
cost. Applied Energy 103 (2013) p. 477-487
[21] M. Johannaber, Auslegung und Energiemanagement von hybriden Brennstoffzellenfahrzeugen. Schriftenreihe
Automobiltechnik, Intitut für Kraftfahrzeuge, RWTH Aachen, (2010)
[22] I. L. Sarioglu, Conceptual Design of Fuel-Cell Vehicle Powertrains. Schriftenreihe des Instituts für
Fahrzeugtechnik TU Braunschweig, Band 36, Shaker Verlag, (2014)
[23] I. L. Sarioglu et al., Optimum design of a fuel-cell powertrain based on multiple design criteria. Journal of Power
Sources. (2014). 266: p. 7-21.
[24] F. Weiß, Optimale Konzeptauslegung elektrifizierter Fahrzeugantriebsstränge. AutoUni-Schriftenreihe, Band 122,
(2017)
[25] F. Odeim, Optimization of Fuel Cell Hybrid Vehicles. (2018)
[26] S. Ruoff, M. Busch, K. Bause, Evaluation of new hybrid electric vehicle drivetrain topologies, Presented at EVS
31 & EVTeC 2018, Kobe, Japan, October 1 - 3, 2018

35th International Electric Vehicle Symposium and Exhibition 10

 [27] D. Wu, S. S. Williamson, Performance Characterization and Comparison of Power Control Strategies for Fuel
Cell Based Hybrid Electric Vehicles, Department of Electrical and Computer Engineering, Concordia University,
2007
[28] X.C. Zhao, G. Guo, Survey on Energy Management Strategies for Hybrid Electric Vehicles, ACTA AUTOMATICA
SINICA, 42(3): 321-334, 2016
[29] A. Sciarretta, M. Back, L. Guzzella, Optimal Control of Parallel Hybrid Electric Vehicles, IEEE
TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, VOL. 12, NO. 3, MAY 2004
[30] G. Paganelli, Y. Guezenne, G. Rizzoni, Optimizing Control Strategy for Hybrid Fuel Cell Vehicle, SAE
International, 2002-03-04
[31] S. F. da Silva et al., Multi-objective optimization design and control of plug-in hybrid electric vehicle powertrain
for minimization of energy consumption, exhaust emissions and battery degradation. Energy Conversion and
Management. Volume 234, 2021, p. 113909
[32] A. Prasanthi et al., Optimization of hybrid energy systems and adaptive energy management. Energy Conversion
and Management. vol. 243, 2021, p. 114357
[33] M. Sorrentino, V. Cirillo, L. Nappi, Development of flexible procedures for co-optimizing design and control of
fuel cell hybrid vehicles. Energy Conversion and Management. 2019, S. 537-551
[34] J. Zhu et al., Bi-level optimal sizing and energy management of hybrid electric propulsion systems. Applied Energy.
vol. 260, 2020, p. 114134
[35] E. Silvas et al., Comparison of bi-level optimization frameworks for sizing and control of a hybrid electric vehicle.
2014 IEEE Vehicle Power and Propulsion Conference (VPPC). IEEE, 2014, p. 1-6
[36] Z. Dimitrova, F. Maréchal, Techno-economic design of hybrid electric vehicles using multi-objective optimization
techniques. Energy. 2015, p. 630-644
[37] A. Braumandl, K. Bause, F. Schlüter, Methodik zur Bewertung von Hybridisierungskonzepten und –grad von
Wasserstoff-Brennstoffzellen betriebener Fahrzeuge. Esslingen, 7. Kolloqium Future Mobility, (2019)
[38] A. Braumandl, K. Bause, Automated Design of Fuel Cell Electric Vehicle Drive Systems. Stuttgart, 22.
Internationales Stuttgarter Symposium Automobil- und Motorentechnik, (2022)
[39] M. Fries et al., An Overview of Costs for Vehicle Components, Fuels, Greenhouse Gas Emissions and Total Cost
of Ownership Update 2017”. München (2017)
[40] Mauler et al., Battery cost forecasting: a review of methods and results with an outlook to 2050. In Energy &
Environmental Science, London, (2021)
[41] S. Satyapal, U.S. Department of Energy Hydrogen and Fuel Cell Technologies Office and Global Perspectives,
Innovation for Cool Earth Forum (ICEF) Website,
https://www.icef.go.jp/pdf/2020/program/concurrent_session/Sunita_Satyapal_P.pdf, last accessed 2022/05/13

35th International Electric Vehicle Symposium and Exhibition 11

Authors
Adrian Braumandl
Dipl.-Ing Adrian Braumandl studied mechanical engineering at the Karlsruhe Institute of
Technology (KIT) and graduated in 2016. After working in the electronics development department
of an industrial trucks manufacturer he joined the Institute of Product Engineering at KIT in 2018.
He is Team Manager of the Research Group Drive Systems, focusing on systems, methods and
processes in the development and validation of conventional, electrified and electric drive systems
and their components.

Guillaume Paris
Guillaume Paris, M. Sc. studied mechanical engineering in a dual graduate program at the
Karlsruhe Institute of Technology (KIT) and the Institut National des Sciences Appliquées de
Lyon. He specialized in Development and Construction as well as Mobile Machinery and
graduated in 2022.

Katharina Bause
Dipl.-Ing. Katharina Bause studied mechanical engineering at the Karlsruhe Institute of
Technology (KIT) and graduated in 2011. Since then she is working as a research associate at the
Institute of Product Engineering at KIT in the field of electric drive systems and validation. She
has been Team Manager of the Research Group Drive Systems from 2016 to 2017 and is Head of
the Research Departments Drive Systems, Clutches and Tribology systems since 2018.

Albert Albers
Prof. Dr.-Ing. Dr. h. c. Albert Albers was born in 1957 and has been Professor and Head of IPEK –
Institute of Product Engineering at the Karlsruhe Institute of Technology (KIT) since 1996. He is a
founding member and former chairman of the Scientific Association for Product Engineering
(WiGeP), member of the National Academy of Science and Engineering (acatech), member of the
Advisory Board of the Design Society and president of the Allgemeiner Fakultätentag (AFT e. V.).

35th International Electric Vehicle Symposium and Exhibition 12