Velocity Prediction And Hierarchical Energy Management For Plug-in Hybrid Electric Vehicles

Plug-in hybrid electric vehicles(PHEVs)are popular among consumers because of their excellent characteristics such as low fuel consumption and long driving range.In this paper,energy management strategy(EMS)related research is carried out with the power distribution PHEVs as the research object.Based on adaptive particle field optimization(APFO)algorithm and model predictive control(MPC)realize PHEV realtime power management online.The specific research contents are as follows:1)Analyze the dynamic structure of the power-split PHEVs,and its key components are modeled and analyzed.Aiming at the structure and lever principle of planetary gear group,power structure of power-split PHEVs,power demand,engine,power battery group and so on,the modeling and detailed analysis are carried out,and the working modes of five different energy supply combinations are analyzed.2)The off-line control strategy based on dynamic programming(DP)algorithm is studied,the optimal state of charge(SOC)discharge curve is designed,and the off-line strategy is simulated and verified in the four selected driving conditions.Analyzing engine power,battery power,and fuel consumption,The verification results show that the DP EMS scheme has better economic advantages than the CD-CS scheme,which also provides a reference for the verification of the designed online EMS scheme.3)Predict the working conditions of the vehicle in the short time domain.In order to better reflect the authenticity of future short-term velocity prediction and improve the control effect,In this paper,I mainly studied Back Propagation(BP)Neural Network(NN),Genetic Algorithm(GA)optimization of BP neural network(GA-BP)and Extreme Learning Machine(ELM),Three velocity(working condition)prediction methods are studied.The working condition selected in this paper is verified by simulation,It is verified that the GA-BP scheme has obvious advantages of higher prediction accuracy and better prediction effect.4)The APFO algorithm is combined with MPC and the velocity prediction is integrated to design online hierarchical EMS of APFO-MPC scheme.Simulation verification is carried out in four different driving conditions of different combinations.The simulation results show that compared with EMS of CD-CS and DP schemes,the studied APFO-MPC strategy has better fuel economy,makes the engine work in the high efficiency zone,and makes the battery power fluctuate in a relatively small range.