Optimization of Electric Vehicle Concepts Based on Customer-Relevant Characteristics 2012-01-0815

Electric vehicles differ from conventionally powered vehicles in terms of many characteristics that are directly relevant to the customer. The most evident ones are the total driving range, which is limited by the battery capacity, and the different acceleration behavior, which is directly influenced by the electric motor's torque characteristics. Furthermore, there are many other vehicle characteristics, such as lateral dynamics, that are also strongly influenced by electrification. For all customer-relevant vehicle characteristics, it is important to know the necessary and optimal fulfillments in order to plan and evaluate new electrified vehicle concepts. Correlation functions can be used to convert values for technical characteristics to normalized customer satisfaction fulfillments. To evaluate the quality of a vehicle concept during the development process, a parametric cost function is defined. The aim of the optimization loops during the concept stage of the development is to achieve the best possible fulfillment of this cost function.

A simulation model has been implemented in Matlab/Simulink to automate this optimization process. The inputs for the model are the weightings and target values of the vehicle characteristics according to the desired positioning. Using this model, technical parameters such as battery capacity or motor power can be varied in order to find the best solution for the cost function and thus for the customer-relevant characteristics. The output setting for the technical parameters is the best solution in terms of customer demands.

Simulation results for a small electric vehicle are presented. Two different weighting assumptions - one more focused on driving range and the other one on cost efficiency - and the resulting optimized concept parameters show the high potential of the concept development approach demonstrated.

Compared to other development processes, the method shown offers the possibility of “analytical prototyping” with various advantages such as reductions in costs and time. In general, the approach could also allow it to be applied to any type of vehicle and indeed to any type of product.