Definition and optimization of the drive train topology for electric vehicles

Due to the limited range of battery electric vehicles, a low energy consumption is more desirable, than it is in conventional vehicles. To accomplish this objective the paper focuses on an increased efficiency of the drive train, its topologies and its components, as this is one of the most promising approaches. With a set of basic characteristics of the desired vehicle (such as maximum speed, acceleration, climbing ability, class and range) an optimal fitted drive train according to the energy consumption should be found. This includes number, type and power of electric machines, transmission ratios, dynamic running radius, axle load distribution and battery capacity. The general approach uses a method consisting of a developed optimization routine and a specific simulation model. The developed optimization algorithm reduces the value ranges or even the design parameters to minimize the number of iterations. This intelligent algorithm is compared to conventional optimizers like pattern search or genetic algorithms. For the vehicle model valid results are important. To ensure validity for all possible topologies, vehicle and power classes an appropriate method is presented. Each relevant component model and its respective scaling concept are validated. After validation of a vehicle model with these component models, the scalability is transferable to the entire vehicle model. Some exemplary results of the model are shown, such as the influence of axle load distribution, choice of high-energy or high-power cells and potential of longitudinal torque-vectoring for multi-motor topologies.