基于工况识别的P2.5构型PHEV能量管理策略OA北大核心CSTPCD

Appropriate energy management strategy effectively improves the range of hybrid vehicles.This paper investigates the energy management strategy for condition identification of dual power source plug-in hybrid electric vehicle (PHEV) with P2.5 configuration by constructing the whole vehicle model through Matlab/Simulink.Nineteen domestic and international typical cycling conditions are selected and classified into three categories by hierarchical cluster analysis according to the characteristics of the conditions,and a support vector machine condition identification model is built and optimized by using the whale algorithm.Our simulation results show the identification accuracy of the optimized condition identification model rises to 97.905％ from 76.259％ recorded with the pre-optimization model,up by 21.646％.Combined with the online working condition recognition model,the power allocation results of the dynamic planning energy management strategy under different working condition categories are learned through neural networks,and the offline learning results are applied to the online control to formulate the energy management strategy based on working condition recognition.Our simulation results show compared with the Charge Depleting-Charge Sustaining (CD-CS) strategy,tthe energy management strategy based on operating condition recognition improve its economy by 7.62％.