Research On Energy Management Strategy Of Serial-parallel Plug-in Hybrid Electric Vehicle Based On Predictive Control

Plug-in Hybrid Electric Vehicles have the advantages of both traditional fuel vehicles and electric vehicles.According to the working characteristics of different power sources,the advantages are complementary and coordinated to make them have good fuel economy and emissions.The formulation of the energy management strategy is the key to determining the energy-saving and emission-reduction performance of the Plug-in Hybrid Electric Vehicle(PHEV).The core is to determine the mode switching and power distribution of each component of the powertrain during driving.Aiming at the series-parallel plug-in hybrid configuration,this paper focuses on the application of energy management strategies based on Model Predictive Control in series-parallel PHEVs to improve vehicle fuel economy.First,the configuration and working mode characteristics of the series-parallel PHEV power system are analyzed.In the Matlab/Simulink simulation platform,the components of the power transmission system and the longitudinal dynamics model are modeled,and the series-parallel PHEV vehicle backward simulation model is built.According to the structure of the target model and the working characteristics of each mode,an energy management strategy for Charge Depleting and Charge Sustaining based on logic thresholds is established to verify the effectiveness of the simulation model.Then,the power distribution problem of the different power components of the series-parallel PHEV is described as a global optimization problem that can be solved by the Dynamic Programming algorithm,and a global optimization energy management strategy based on the Dynamic Programming algorithm suitable for the hybrid structure is constructed.Through reasonable discretization,reverse solution and forward optimization,the numerical solution of this strategy is completed.Perform simulation verification for specific driving cycle and analyze the simulation results to provide theoretical support and evaluation criteria for the subsequent formulation of energy management strategies based on Model Predictive Control.Finally,combining Dynamic Programming with Model Predictive Control,the series-parallel PHEV energy management strategy based on model predictive control is proposed.The focus is on the application of dynamic programming algorithms in the rolling optimization process and the formulation of the state reference trajectory,and the calculation of the feasible region of the state of charge(SOC)of the power battery pack reduces the amount of calculation in the optimization process.Through simulation analysis,the effectiveness of the energy management strategy based on Model Predictive Control is verified.On this basis,a one-step Markov prediction model,a multi-step Markov prediction model and a Radial Basis Function(RBF)prediction model are established.By comparing the prediction effects of various models,a combined prediction model based on online identification of working conditions is proposed.The above prediction models are applied to energy management strategies based on Model Predictive Control to explore the influence of different predictive models and prediction effects on Model Predictive Control.