Research On Driving Cycle Prediction And Energy Management Strategy Optimization Of Plug-in Hybrid Electric Vehicle

In the context of the national double carbon target and vigorously promote energy saving and emission reduction,the automotive industry is accelerating the transition from traditional fuel vehicles to new energy vehicles.Plug-in hybrid electric vehicle(PHEV),as a new energy vehicle model,has two power sources: engine and electric motor,which makes up for the shortcomings of both traditional fuel vehicles and pure electric vehicles while integrating the advantages of both,so its key technology,the vehicle energy management strategy,has become the focus of research and development by major car companies and research institutes.As the core technology of plug-in hybrid electric vehicles,intelligent energy management strategy can largely improve the fuel economy of the vehicle,which is important for the improvement and optimization of PHEV performance.The main research contents and results of this paper are as follows:(1)The rule-based energy management strategy is constructed and validated by simulation.The P2 configuration parallel PHEV is used as the research object,and the vehicle dynamics model is established in Cruise software according to the basic parameters of the vehicle,and the rule-based energy management strategy is established in Matlab/Simulink software to control the vehicle model,and then the effectiveness of the established strategy model is verified through joint simulation,which lays the foundation for the following energy management strategy optimization research.(2)The study of the PHEV energy management strategy optimization problem and the equivalent fuel consumption minimization strategy(ECMS)is carried out.Based on the Pontryagin’s principle of minimal value,an equivalent fuel consumption minimization algorithm is derived,and the ECMS-based energy management strategy is constructed according to the optimization objective and additional constraints,and then the effectiveness of the strategy and its superiority in fuel economy are verified through 6-fold UDDS driving cycle.Finally,the ECMS strategy is compared and analyzed in two cases: the same driving cycle with different equivalent factors and the same equivalent factor with different driving cycles.(3)A RBF neural network driving cycle prediction method incorporating prospective information is proposed.Firstly,based on the virtual urban traffic environment established in SUMO software,the driving data of the host vehicle and traffic information such as road speed limit and speed of the vehicle ahead are obtained;secondly,based on the comparison of three prediction methods such as exponential prediction and BP neural network prediction,the prediction method based on RBF neural network is determined;then,by combining real-time traffic information,the prediction model is improved and the RBF neural network driving cycle prediction method incorporating forward-looking information.The online prediction effect of the model under two different driving cycles shows that the prediction error of vehicle speed is mostly kept within ?5km/h,which indicates that the established prediction model has good prediction accuracy and robustness.(4)An adaptive ECMS energy management strategy is constructed by combining the optimization strategy and the driving cycle prediction model.First,based on the driving cycle prediction,the optimal initial equivalence factor of the driving cycle in the predicted time domain is solved by a simplified ECMS algorithm;then,with the deviation of the actual SOC from the reference SOC as the input,a PI controller is established to correct the initial equivalence factor in order to achieve the full utilization of the battery power during the whole trip;finally,the simulation analysis of the energy consumption of the three energy management strategies under two different tested driving cycles shows that the established adaptive ECMS strategy saves 12.02% and 12.83% of fuel consumption under the two driving cycles,respectively,compared to the rule-based strategy,which saves 7.24% and 8.37% of fuel consumption compared to the ECMS strategy.This indicates that the established adaptive ECMS strategy has better fuel savings and driving cycle adaptiveness than the other two strategies,which can effectively improve the fuel economy of PHEV.