Research On The SOC Estimation Method For Lithium Ion Battery Of EV

State of charge of battery is an index to measure the remaining available power of electric vehicles.Accurate SOC estimation is conducive to driving safety and is one of the important indicators in the energy management system of electric vehicles.In order to estimate the state of EV lithium ion battery more accurately and quickly and apply to different types of lithium ion battery,a WNN(Wavelet Neural Network)SOC estimation method based on principal component analysis and genetic algorithm optimization is proposed in this paper.Paper analyzes the lithium ion battery’s characteristics,detailedly.Lithium ion battery history data according to the NASA Ames research center,according to PCA,the parameters that affect the state of battery charge are analyzed by principal component analysis.Principal component analysis to extract the principal component in the effective retention of raw data contains information at the same time,greatly reduce the complexity of the follow-up data processing model.The SOC estimation model of wavelet neural network was established,and genetic algorithm was introduced for optimization.The processed data were used to estimate SOC of lithium ion battery with the wavelet neural network model optimized by genetic algorithm.Finally,the results was compared with BPNN,WNN,and GA-WNN,and the RMSE index was used to evaluate the prediction results of the whole model.The results show that the optimized wavelet neural network based on principal component analysis and genetic algorithm can estimate SOC of electric vehicle battery more accurately,good convergence,and prevent the model from being trapped in local secondary advantages.