Energy Management Strategy Of Extended-Range Electric Vehicle Based On Driving Pattern Recognition

With the advantages of simple configuration,high fuel economy and long range,Extended-Range electric vehicle has become one of the important research directions of new energy vehicles.Energy management strategy determines the economy,emissions and power performance of incremental electric vehicles to a certain extent.In this paper,the multiobjective optimization of energy management strategy for incremental electric vehicles based on condition identification is carried out around the torsional vibration of the range extender shafting,energy distribution between power sources and condition self adaptability.The main research contents are as follows:(1)The 8-DOF torsional vibration model of the range extender shafting is established,and the torsional vibration map is calculated and constructed.Based on matlab/simlink,the whole vehicle simulation model of Extended-Range electric vehicle is established.The model includes the whole vehicle power model,drive motor model,power battery pack model,and range extender model combined with torsional vibration map and fuel consumption model.The model is verified by torsional vibration test and fuel consumption test,which provides a basis for the follow-up multi-objective optimization of energy management strategy.(2)A genetic algorithm optimized BP neural network(GA-BP)driving pattern identification method is proposed.This method takes the characteristic parameters of the speed range as input and the type of driving pattern as output,and establishes GA-BP driving pattern recognizer and BP working condition recognizer respectively.Compared with its recognition effect,GA-BP driving pattern recognizer has significantly improved in error,maximum error and recognition accuracy.On the basis of traditional strategy and driving pattern identification,the fuel economy of incremental electric vehicles can be further improved.(3)Aiming at the problem that the research on energy management strategy under fixed driving patterns is difficult to adapt to complex driving conditions,a multi-point energy management strategy of range extender based on driving pattern identification and fusion of working modes is proposed.A multi-objective optimization function considering torsional vibration and equivalent fuel consumption is constructed by means of weight summation.Taking the key parameters of energy management strategy as optimization variables,the improved drosophila algorithm is used for global optimization.Analyze the optimization results of different weight distribution before and after driving pattern identification.The simulation results show that the overall optimization objective,equivalent fuel consumption and fuel consumption are improved.(4)The proposed energy management strategy is verified,and a hardware in the loop test system based on real-time simulation hardware platform including simulated vehicle model and Huahai rapid prototype vehicle controller is established.Rapid prototype vehicle controller is the real controller of the vehicle,real-time simulation platform is the carrier of the vehicle model,and Chinese working condition is taken as the input to evaluate the feasibility of multi-objective optimization working condition identification energy management strategy.The results show that the error of hardware in the loop test results in cost,power consumption,fuel consumption,comprehensive torsional vibration and emission is small,the SOC curve is basically consistent,and the energy management strategy has high real-time and reliability,and can meet the design requirements.