Predictive Energy Management For A Plug-in Hybrid Electric Vehicle Based On Real-time Traffic Information

Developing the plug-in hybrid electric vehicles(PHEV)is an effective way for energy saving and low emission,which releases the high pressure from energy shortage,global warming and environment pollution,and becomes the research spot and main direction of new energy vehicle development.PHEV has have a larger battery with a SOC that can be used in a wider range and the battery can be charged from the engine or power grid,which combines the advantages of the pure electric vehicles and hybrid electric vehicles.The energy management strategy is the key technique of PHEV control system,which directly affects the fuel economy.To realize the near optimal global energy distribution of PHEV,this thesis develops a predictive energy management strategy based on real-time traffic information.The main contributions are as follows:(1)The hybrid system structure and working mode of parallel PHEV are analyzed,and the key components are modeled by the experimental modeling method.Based on the dynamic programming(DP)theory,a mathematical model is established for the optimization of PHEV energy managemen,and a global optimization energy management strategy based on dynamic programming is proposed.To balance the computational complexity and the results of the DP,the influence of state variables and control variables discrete precision on the results of DP is researched.The global optimization results are analyzed and the SOC descending rule under the repeated driving cycle are researched,which lays the foundation for establishing theoretical reference SOC.(2)Based on the model predictive control(MPC)theory,the mathematical model of PHEV rolling optimization is established,and the DP algorithm is used to solve the optimization problem in prediction horizon,and a PHEV energy management strategy based on MPC is proposed.The influence mechanism of the exponential prediction model with different decay factors is studied,and an exponential prediction model based on support vector machine(SVM)driving cycle recognition is proposed.The research on the MPC based PHEV energy management strategy is carried out.The necessity of the reference SOC trajectory is verified,and the validity of the rolling optimization method for each time constraint is verified without considering the influence of the prediction model.The MPC simulation based on the SVM recognition exponential prediction model is conducted under the comprehensive driving cycle,the economy of MPC can be improved by the optimization of the prediction model.At the same time,the MPC is greatly influenced by the accuracy of reference trajectory,the theoretical reference SOC has the the driving cycle limitation.(3)The real-time traffic information data obtained from the virtual traffic simulation environment established by the VISSIM platform is analyzed.In order to obtain the possible driving cycle of the target vehicle in the future,the average speed of traffic flow in the front section of the road is obtained,and the trip model of the target vehicle is established.In order to ensure the real-time performance of information in the update cycle,a rapid SOC planning algorithm based on simplified DP is proposed,and the effectiveness of simplified DP algorithm is verified under different types of driving cycle,whcih greatly reduces the SOC planning time and ensures good accuracy.(4)The application principle of real-time traffic information in MPC framework is analyzed,and a PHEV predictive energy management strategy based on real-time traffic information is proposed and the simulation research is carried out.The influence of reference SOC planning delay time on MPC is analyzed,and the effectiveness of the simplified DP fast planning algorithm is further verified.The effect of MPC with and witout real-time traffic information is analyzed,the MPC-traffic can reach the 92.83% optimality as DP.Compared with the MPC based on the theoretical reference SOC,the economy is improved by 6.18%,and the effectiveness of the proposed energy management strategy is verified.The MPC control effect under different update cycle is analyzed,and the influence rule of update cycle on the proposed energy management strategy is researched.