State Estimation of Lithium Battery Based on Least Squares Support Vector Machine

In order to improve the accuracy of battery SOC, this paper presents a novel least squares support vector machine(LSSVM) framework based on machine learning. Put the current, voltage and temperature at the current moment and the SOC estimated at the previous time are used as input vectors of the model to estimate the SOC at the current time. The experimental results show that the proposed model can achieve better SOC estimation accuracy than the LSSVM model with limited data samples.