Energy optimization of multi-mode coupling drive plug-in hybrid electric vehicles based on speed prediction

Abstract By integrating the functions of centralized drive and distributed drive into a series-parallel hybrid system, the multi-mode coupling drive system can greatly improve the fuel economy of a plug-in hybrid electric vehicle (PHEV). However, some simple energy management strategies do not give full play to the advantages of the drive system. In order to get better fuel economy, after the system working principle analysis and modeling, a vehicle speed prediction model combining Markov and BP neural network algorithm was developed to predict the speed of the next 5s, and an adaptive equivalent consumption minimum strategy (AECMS) based on the combined vehicle speed prediction is proposed to optimize the drive modes selection and power distribution. The vehicle speed prediction accuracy was verified by the actual vehicle road test and the energy management effect was verified by the simulation. The research results show that, the prediction accuracy of the combined vehicle speed prediction can be improved by 27.9% compared with the ordinary single speed prediction, and the proposed control strategy improves the energy consumption of 3.7% for the PHEV under the same driving cycle condition when compared with the rule-based optimization strategy.