State-of-Charge Estimation for Lithium-ion Battery Using Improved Particle Filters

In the ever-evolving landscape of energy storage systems, the accurate estimation of battery state of charge stands as a critical component. This paper introduces a comprehensive approach to achieving highly accurate battery state of charge estimation. The foundation of our method is the single particle model with electrolyte dynamics, coupled with a particle filter algorithm, which has proven to be a powerful solution for accurately estimating SOC. To address challenges such as particle degeneracy and the loss of particle diversity, we introduce the particle swarm optimization algorithm, further augmenting the particle filter's performance. The battery is tested under two different types of working conditions, constant current discharge conditions and real-world operational scenarios. The experiment results presented in Section 4 provide compelling evidence of the superior SOC estimation capabilities of our proposed approach.