Research On Adaptive Equivalent Fuel Consumption Minimum Energy Management Strategy Of PHEV Based On Driving Condition Recognition

Reasonable formulation of energy management strategy can greatly optimize vehicle economy and emission performance.However,the existing energy management strategies don’t consider the effect of actual driving conditions on the control strategy.So,this paper starts from the instantaneous optimization of energy management strategy to study the driving condition adaptability of energy management strategy in order to improve the fuel economy of the vehicle.Firstly,referring to the parameters of the power system of a plug-in hybrid electric vehicle,based on ADVISOR software platform,the models of the components of the vehicle were established.Based on the original structure of the two-axle parallel hybrid electric system,the power system was redeveloped,thus the whole model of the plug-in hybrid electric vehicle was established,which was used as the simulation platform for further research.Secondly,19 typical driving conditions were selected and five types of driving conditions were obtained by cluster analysis method.Five representative driving conditions were selected as standard from these five types of driving conditions.Then,based on five standard driving conditions,BP neural network algorithm was used to identify the type of driving condition.In addition,aiming at the poor convergence of training results caused by the initial randomization of weights and thresholds of BP neural network,the PSO algorithm was proposed to preprocess the initial weight threshold of the neural network,and finally optimize the recognition effect of the BP neural network.Then,based on the theory of equivalent consumption minimization strategy,the energy management strategy of Plug-in Hybrid Electric Vehicle was formulated.In view of the real-time variation of equivalent fuel factor,the optimal SOC sequences of batteries under various standard driving conditions were solved by DP algorithm under the constraints of battery power balance,and the optimal SOC sequence was taken as the reference trajectory,and the equivalent fuel factor sequences of five standard driving conditions were obtained by simulation of vehicle control strategy.Through the real-time identification of the driving condition identification model,the corresponding effective fuel factor could be solved immediately,thus realizing the real-time application of the energy management strategy.Finally,the simulation results of rule-based motor-assisted energy management strategy,equivalent fuel consumption minimal energy management strategy without considering driving condition identification and adaptive equivalent fuel consumption minimal energy management strategy based on driving condition identification were comprehensively compared and analyzed.From the aspects of fuel economy improvement,integrated battery balance ability and adaptability to driving conditions,results shown that the adaptive equivalent fuel consumption minimization strategy based on condition identification had obvious advantages.