Research On Predictive Energy Management Strategy Of Parallel Hybrid Electric Vehicle

Against the background of environmental pollution and energy scarcity,due to the advantages of energy saving and emission reduction and relative mature technology,hybrid electric vehicles(HEV)have developed rapidly in the global automotive industry in recent years.Energy management strategy is the core technology to improve the fuel economy of HEV,which can reduce the fuel consumption of the whole vehicle by optimizing the distribution of force source power or torque on the premise of satisfying the demand of vehicle driving power.Among them,the energy management problem of parallel HEV involves the optimal control of gears,which needs to take into account both fuel economy and driving comfortness.The calculation requirements of the control strategy are relatively high,which increases the difficulty of designing the algorithm.Therefore,to develop efficient and real-time optimization control strategies for the parallel HEVs are very necessary.This paper takes parallel HEV as research object.By applying the equivalent consumption minimum strategy(ECMS)to the model predictive control(MPC)framework and fusing the prediction information of vehicle speed,generate an efficient and adaptable MPC energy management strategy to achieve optimal distribution control of parallel HEV torque and gear in the same time.The specific research work is as follows:First,modeling each component of the parallel HEV system,using the method of fitting test data to construct the continuous function model of engine and motor respectively,build an equivalent circuit model of the battery based on the internal resistance model,and then establish a longitudinal dynamic model of the vehicle.According to power balance relationship on the shaft,a transmission system model is generated.By constructing models for each part,a backward simulation model paralleled to HEV is established,which lays the foundation for the study of control strategies.Then,a model of vehicle speed prediction based on BP neural network is designed to obtain the vehicle speed information in the limited time zone of the parallel hybrid electric vehicle.By selecting representative standard city working conditions as training samples for learning and training,and afterwards,two city working conditions are selected as test samples for validation,and then comparison and research on prediction errors under different parameter settings are done to finally select synthetically the parameter values with smaller prediction errors.The network structure of the vehicle speed prediction model is determined,which provides a prerequisite for the realization of the subsequent model prediction control strategy.Finally,a model prediction(ECMS-MPC)energy management strategy based on the minimum equivalent fuel consumption is proposed for the parallel hybrid vehicle.By combining the ECMS algorithm and the model prediction control strategy,the power consumption of the parallel HEV battery is now equivalent to fuel consumption,and then takes into account driving comfortness and fuel economy to achieve optimal control of torque distribution and gear switching.In order to ensure the driving comfortness,the gear shift penalty function is introduced into the objective function to achieve the optimal control of the gears,avoiding frequent gear shifting behavior.In addition,in order to verify the effectiveness of the ECMS-MPC control strategy,a comparative simulation analysis is performed with the dynamic programming(DP)control strategy and the dynamic prediction-based model prediction(DP-MPC)control strategy.The optimization results show that,compared with the other two control strategies,the calculation efficiency based on the ECMS-MPC control strategy is higher,and the fuel consumption level is close to the global optimal DP control strategy,which proves the optimal performance of the control strategy.Finally,the sensitivity of the optimized performance of the ECMS-MPC control strategy to the parameters is discussed to determine the appropriate parameter value to obtain the optimized performance of the ECMS-MPC.