Fuzzy Energy Management Optimization for a Parallel Hybrid Electric Vehicle using Chaotic Non-dominated sorting Genetic Algorithm

This paper presented a parallel hybrid electric vehicle (HEV) equipped with a hybrid energy storage system. To handle complex energy flow in the powertrain system of this HEV, a fuzzy-based energy management strategy was established. A chaotic multi-objective genetic algorithm, which optimizes the parameters of fuzzy membership functions, was also proposed to improve fuel economy and HC, CO, and NOx emissions. The main target of this algorithm is to escape from local optima and obtain high quality trade-off solutions. Chaotic initialization operator, chaotic crossover and mutation operators, chaotic disturbance operator, and chaotic local search operator were integrated into non-dominated sorting genetic algorithm II (NSGA-II) to form this new algorithm named chaotic NSGA-II (C-NSGA-II). Simulation results and comparisons demonstrated that chaotic operators can enhance searching ability for optimal solutions. In conclusion, C-NSGA-II is suitable for solving HEV energy management optimization problem.