Optimal energy management strategy for parallel plug-in hybrid electric vehicle based on driving behavior analysis and real time traffic information prediction.

Conclusive evidence has justified great importance of energy management strategies in the performance and economy of plug-in hybrid electric vehicles (PHEVs). This article pays attention to improve adaptive equivalent consumption minimization strategy (A-ECMS) for parallel PHEV based on driving behavior recognition and real time traffic information prediction. Three main efforts have been made to distinguish our work from exiting research. Firstly, a hierarchical driving behavior model is constructed, providing in-depth knowledge about behavior generation, transmission, and consequence. Secondly, an online driving behavior classification method is designed. The proposed method is the coefficient result of offline driving behavior study based on self-report driving behavior questionnaire (DBQ) and online driving behavior discrimination by BP neural network. Thirdly, an improved adaptive equivalent consumption minimization strategy (IA-ECMS) is formulated based on identified driving behavior and predicted real time traffic information. The IA-ECMS can realize equivalent factor tuning instantaneously and reasonably. The simulation results indicate the proposed energy management strategy holds potential in fuel economy improvement than A-ECMS. [ABSTRACT FROM AUTHOR]