State of charge estimation for lithium-ion batteries based on improved barnacle mating optimizer and support vector machine

Abstract

Precise state of charge (SOC) estimation is essential for battery management systems. The improved barnacle mating optimizer-support vector machine (IBMO-SVM) model is proposed and used for SOC estimation of lithium-ion batteries. (1) The cubic chaotic mapping, hyperbolic sinusoidal conditioning factor, and Gauss-Cosey variation are introduced to improve the barnacle mating optimizer (BMO) to obtain the improved barnacle mating optimizer (IBMO); (2) The convergence performance of IBMO is verified by comparing with other five intelligent optimization algorithms under several test functions; (3) IBMO-SVM is created by using IBMO to optimize the search for support vector machine (SVM) parameters; and (4) IBMO-SVM is used for SOC estimation of lithium-ion batteries and the estimation results are analyzed by multiple evaluation indexes. The proposed model's root mean squared error (RMSE) and mean absolute percentage error (MAPE) are 0.0042 and 0.61 %, and its R-square (R2) is 0.9994, outperforming the four comparison models. The SOC estimation methodology proposed in this study is highly accurate and reliable, and it provides advantages for improving the battery management system.