Research On Collaborative Optimization Energy Management Strategy For Connected And Automated Plug-in Hybrid Electric Vehicle

The matter of petroleum supply and environmental pollution problem have always been coupled up with the development of China's automobile industry,and has become a shadow mistily covers the vigorously developing automobile market.Plug-in hybrid electric vehicle(PHEV)is more mature than fuel cell electric vehicle(FCEV)on technology level,meanwhile,it has longer travel distance endurance and lower dependence on charging facilities than electric vehicles(EV).Therefore,the development of connected and automated PHEV becomes an effective way for new energy vehicles to realize further unleash of energy saving and emission reduction potential by means of automated driving and connected vehicle technologies.At present,there merely exists systematic research on PHEV collaborative optimization energy management strategy,in which velocity curve is autonomously planned with power source torque real-time distribution controlled collaboratively.In this study,dynamic programming is adopted to comprehensively study the energy management strategy of collaborative optimization for connected and automated PHEV from three aspects: solution method,terminal states constraints and simplification.The main contents are as follows:(1)The structure,operating mode and parameters of the research object are analyzed and introduced.The core power components model of PHEV is established by combining theoretical and experimental method.The global optimal energy management strategy for PHEV based on dynamic programming is designed to study the shift characteristics,power source torque distribution and SOC changes tendency.The difference between CD-CS strategy and global optimization energy management strategy in energy management is compared,and the global optimization energy management strategy can be used as the basis for the follow-up study of collaborative optimization energy management strategy.(2)Based on the principle of dynamic programming and the different kinds of speed constraints,the energy management strategy of PHEV collaborative optimization based on dynamic programming time domain calculation and the energy management strategy of PHEV collaborative optimization based on dynamic programming space domain calculation are designed respectively.The speed-limited driving cycle are defined and studied and its similarities and differences are compared with deterministic driving cycle.A design method of space-domain non-uniform discrete driving cycle is proposed by directly converting the its time-domain driving cycle.The effectiveness of collaborative optimization energy management strategy is demonstrated in three use scenarios of PHEV: AER mode,HEV mode and hybrid mode.The results indicate that both in time domain or in space domain,the energy management strategy can achieve the optimal energy consumption cost under the constrains of speed and acceleration.However,the time domain calculation can not restrict the terminal travel distance,while,space domain calculation cannot restrict the terminal travel time.The interpretation of this phenomenon is descried and the study of terminal state constraint methods for the strategy will be conducted in the following chapter.(3)The three means: state variable constraints based terminal state restriction,timespatial domain calculation based terminal state restriction and adjustment factor based terminal sate restriction.Based on the principle of dynamic programming,three kinds of constraint method are designed and validated.Finally,the third one is proven to be the most suitable for connected and automated PHEV to restrict the terminal state.The calculation verification of the constraint method on terminal state restriction is conducted under the three use scenarios of PHEV.The results reflect that both the time domain solution and the space domain solution can significantly improve the overall energy consumption economy of PHEV with terminal state constraints.The analysis can provide a systematic framework for the research of terminal state constraints in dynamic programming based collaborative optimization energy management problems.(4)The simplification and optimization the of the strategy algorithm are conducted in four aspects: model simplification,dimensionality reduction of control variables,simplification of state variables and optimization of MATLAB program.The difference of energy management performance and computing time between the three strategies of fast planning,collaborative optimization and global optimization are compared under local planning and global planning.Result shows that the fast planning can significantly reduce the computing time on the premise of optimization performance ensured,which proves that fast planning has the potential of online application.The study lies a preliminary exploration for the online application of collaborative optimization in the follow-up research.