Energy Management Strategy For A Extended Range Electric Vehicle Based On Velocity Prediction

With the adjustment of the new energy vehicle industry structure,extended rang electric vehicle has become a very good transitional model in the development of pure electric vehicles,and has broad development prospects.It will not cause anxiety due to battery life during longdistance driving,because the EREV equipes with an auxiliary power system on the basis of pure electric vehicles.Traditional EREV uses rule-based energy management strategies,which cannot be adjusted to make optimal control in real time,resulting in lower energy utilization and higher fuel consumption.In order to reduce the fuel consumption during the driving process of the EREV,optimize the power distribution between the two power sources of the power system and improve the real-time performance of the power distribution,this dissertation researched energy management strategies based on vehicle speed prediction,relied on the Shaanxi Province key industry innovation chain(group)project:the key technology of independent driving extended range passenger car(2020ZDLGY16-01),and the key technology research of plug-in /extended-range hybrid commercial vehicles(2019ZDLGY15-02).Based on the analysis of the structure of the EREV’s power system,according to the driving cycles of Chinese passenger vehicles(CLTC-P)and the performance parameters of the prototype vehicle,the key component parameter matching and modeling of the extended-range electric vehicle power system were completed.The comprehensive performance is verified under the energy management strategy,which provides a model basis for the development of the energy management strategy in this paper.Aiming at reducing fuel consumption,according to the principle of Dynamic Programming(DP),a global optimal energy management strategy was constructed and the solution process was studied.Based on CLTC-P,the global optimal characteristics under the DP algorithm strategy are analyzed.The energy management strategy based on DP algorithm provides the direction for the energy management strategy based on vehicle speed prediction in this paper.Based on test data,a multi-scale single-step speed prediction model based on MarkovMonte Carlo method and a speed prediction model based on wavelet neural network(WNN)are constructed respectively.These models were simulated at Xi’an EV driving cycles,and the rootmean-square error was used to analyze the prediction effects.Finally,a real-time vehicle speed prediction method based on the WNN network was selected.Based on predictive control theory,using WNN network vehicle speed prediction as a prediction model,and combining with the DP algorithm,an energy management strategy based on vehicle speed prediction for EREVs is established.The simulation of the energy management strategy based on vehicle speed prediction under the typical driving cycles of EVs in Xi’an was completed.The simulation results were compared with the fixed-point energy management strategy for gasoline engines and the global optimal energy management strategy based on DP algorithm.Compared with the fixed-point energy management strategy,the energy management strategy can make fuller use of the electric energy in the power battery,reduce the fuel consumption and the economic cost.The real-time control is improved.