Research On Energy Management Strategy Of Dual Motor Coupling Drive PHEV Based On Model Prediction

With the increasingly serious problems of environmental pollution and energy shortage in China,the promotion and application of new energy vehicles become increasingly important.As a new energy vehicle,PHEV can effectively reduce energy consumption and avoid the mileage anxiety caused by pure electric vehicle.However,the structure of its multi power source also brings many problems in its energy management strategy.Therefore,how to reasonably develop the appropriate energy management strategy has become an important research topic of PHEV.First of all,this paper proposes a dual motor coupling drive PHEV power system configuration,which realizes a variety of working modes and meets the high efficiency operation of the proposed configuration under different working conditions.For each working mode,the dynamic characteristics of the system are analyzed based on lever analysis method.At the same time,the matching method based on the combination of power performance,economy and condition statistical analysis is used to match the core power system components to reduce the power redundancy of power components,so as to achieve better vehicle economy.Finally,the particle swarm optimization algorithm is used to further optimize the parameters of the power system with the minimum energy consumption as the objective function,so that the energy consumption is better.Compared with before optimization,the energy consumption is reduced by 8.4%.Secondly,complete the modeling of vehicle power system,and establish the mathematical models of engine model,motor model,power battery model,driver model and other core components.Based on MATLAB/Simulink/Stateflow platform,the whole vehicle simulation model is built,which provides the basis for the implementation of the follow-up energy management strategy.Then,a rule-based energy management strategy is proposed to verify the effectiveness of the model,and the efficient operation of PHEV driven by dual motor coupling under various conditions is realized;a global optimal energy management strategy based on dynamic programming is established for the proposed configuration,which achieves the optimal energy consumption under working conditions.Compared with the rule strategy,the fuel consumption is reduced by 18.1%.The above two energy control strategies establish an effective evaluation standard for subsequent control strategies.Finally,the energy management strategy based on model predictive control is established,and the exponential predictive model,Markov predictive model and RBF neural network predictive model are studied.The prediction accuracy of different prediction models in different prediction horizon is compared and analyzed,and the optimal solution in the prediction horizon is obtained by combining with the dynamic programming theory.The local optimal and global approximate optimal energy management strategy in the finite prediction horizon is realized.Compared with the regular energy management strategy,the vehicle economy is effectively improved and the driving cost is reduced by 12.6%;compared with the global optimal energy management strategy based on dynamic programming,the single step calculation time is 0.269 s,which has a good real-time application potential,which provides the possibility for the application of control strategy in real vehicle.