State of Charge Estimation for Lithium-Ion Battery Based on Particle Swarm Optimization Algorithm and Multi-Kernel Relevance Vector Machine

Abstract

Lithium-ion batteries are key components of energy storage systems and electric vehicles, and their accurate State of Charge (SOC) estimation is important for battery energy management, safe operation and extended service life. In this paper, Multi-Kernel Relevance Vector Machine (MKRVM) and Particle Swarm Optimization (PSO) are used to estimate the SOC of Li-ion batteries under different operating conditions. PSO is used to automatically adjust and optimize the weights and kernel parameters of MKRVM to improve estimation accuracy. The proposed method is validated on three battery operation experiment under different operating conditions. The test results show that the proposed PSO-MKRVM can precisely estimate the battery SOC under different operating conditions with an accuracy higher than 0.99 and its maximum average error does not exceed 2%.