Comparison of architecture and adaptive energy management strategy for plug-in hybrid electric logistics vehicle

Highlights: • An IGA-BP neural network driving pattern recognition model is proposed. • Comparisons with three hybrid powertrain systems are made. • Comparisons with two adaptive energy management strategies are made. This paper deals with the comparison of architecture and adaptive energy management strategy (EMS) for hybrid powertrain system (HPS), including one or two electric motor, an engine and a battery, for a plug-in hybrid electric logistics vehicle (PHELV). The most attractive advantage deriving from HPSs is the possibility of reducing emission and improving fuel economic. For comparison purposes, the series, parallel, and series-parallel hybrid powertrain system are examined by dynamic programming (DP) algorithm using the same vehicular parameters. The approach of adaptive EMS is driving pattern recognition (DPR) to obtain optimum estimation of EMS parameters under different driving cycle. A back propagation (BP) neural network DPR optimized model by an improved genetic algorithm (IGA) has been proposed. Taking the costs of fuel consumption, the parameters of the fuzzy logic controller (FLC) and equivalent consumption minimization strategy (ECMS) are optimized. The comparative results show that the series PHELV fuel economy improvements are 7.60% and 6.53%, compared with parallel and series-parallel PHELV. The difference between the optimal fuzzy energy management strategy and the global optimization is 4.74%, and the ECMS is 4.66%.