Multi-objective optimization of hybrid electric vehicles considering fuel consumption and dynamic performance

In this paper a new concept for the layout of hybrid-electric-powertrains is developed that includes optimization of the component-sizes as well as control strategies. In contrast to most existing publications, the approach explicitly considers the conflicting goals of low fuel consumption and high vehicle longitudinal dynamics and the trade-off is quantified. Two multiobjective optimization subproblems are solved for one example with a parallelized genetic algorithm (NSGA-II) using the Condor software framework. The analysis of the solutions (Pareto front) shows that combinations exist which improve the fuel consumption with only a slight deterioration of the dynamic performance. So the designers are supported in their decision for a configuration which is attractive for the customers.