The SOC estimation and simulation of power battery based on self-recurrent wavelet neural network

Abstract

Improving the accuracy of power battery state of charge (SOC) estimation is an important basis for formulating and optimizing the control strategy of hybrid electric vehicle. Aiming at the problem of the traditional SOC large estimation error, a SOC estimation method based on self-recurrent wavelet neural network (SRWNN) is proposed. Firstly, the equivalent model of power battery and its SOC estimation model are established. Then, The SOC estimation algorithm based on SRWNN is designed in detail. At last, The SOC estimation method based on SRWNN is compared with other SOC estimation methods from SOC estimation error and vehicle performances in the Matlab simulation environment. The simulation results show that its SOC estimation accuracy is highest and vehicle performances by using SRWNN is best.