Research On Energy Management Strategy Of Plug-in Hybrid Electric Vehicles Based On Predictive Control

Energy management system(EMS)is a critical technology in control system of HEV/PHEV.The control target of EMS is to improve vehicle fuel consumption performance and dynamic performance,which is also the advantage of HEV/PHEV.With the development of V2X and Intelligent Transportation Systems(ITS),it is possible for On Board Unit(OBU)to obtain road and traffic information(traffic conditions,speed limits,road slopes,etc.).The predictive energy management technology using road and traffic information is expected to significantly improve the fuel economy of PHEV,but it also brings problems such as heavy calculation burden and high calculation cost.Therefore,it is very important to design a predictive energy management control method with good control effect and high calculation efficiency.The proposed predictive energy management strategy is divided into two layers: A upper level algorithm periodically incorporates macroscopic traffic flow and topographic information of the intended route from ITS to predict the future driving trip,which is then clustered for optimal global So C reference trajectory planning.The lower level MPC algorithm predicts the short-term vehicle speed and determines the optimal power split while tracking the above reference So C trajectory and satisfying system constraints.The characteristics of the global optimal So C trajectory under different driving cycle combinations and road topographies was explored.Then an algorithm which apply the road topography and traffic information of ITSs to quickly plan a So C reference trajectory close to the global optimal solution is proposed.The lower level employs a dual-model Model Predictive Control(MPC)method and Forward Dynamic Programming(FDP)algorithm to handle battery power constraint with reduced computational complexity,where 1st order RC battery model and Rint model are fused in an elegant manner.In order to further improve the calculation efficiency,the firstorder RC model is reduced from the 2-state model to the equivalent 1-state model based on strict mathematical derivation.Finally,Model-in-Loop(MIL)test platform was built to simulate and verify the control effect of the EMS algorithm.The results show that compared with the traditional method,the proposed strategy has better performance in fuel economy and More rigorous constraint.