Multi-objective parameters optimization of electric assist control strategy for parallel hybrid electric vehicle

To manage the drivestrain and power flow of hybrid electric vehicle (HEV) and achieve balanced performances on fuel economy, emissions, grade ability and acceleration ability, it is necessary to develop a control system with suitable parameters by solving a multi-objective optimization problem. Based on the proof of the relationship between dominating number and diversity in objective-space, a dominating number based multi-objective genetic algorithm is proposed to enhance the diversity of Pareto-front and has been applied to optimize parameters of electric assist control strategy for parallel HEV under standard test procedure provided in Advisor. Simulations under four drive cycles demonstrate availability and efficacy of the proposed algorithm.