基于预测静置开路电压法的锂电池SOC估算OA

The state-of-charge(SOC)estimation of batteries generally uses the principle of ampere-hour integration to construct SOC models,which are further combined with filtering algorithms.To address the problems of a slow convergence speed and complex calculation,a predictive static open circuit voltage(PSOCV)method is proposed based on the open circuit voltage principle and second-order RC circuit models.The feasibility of discrete-time forward and backward difference state-space equations was discussed,and the parameter accuracy was also compared between linear interpolation(LI)fitting and least squares(LS)fitting.Experimental results show that LI fitting is generally better than LS fitting.The average absolute error of SOC estimation obtained using the LI-PSOCV algorithm under HPPC and UDDS operating conditions is below 1％.Compared with the cubature Kalman filter(CKF),the LI-PSOCV algorithm can converge instantly when the initial SOC value deviates and run faster.