Optimization of the SOC-based multi-stage constant current charging strategy using coyote optimization algorithm

This study presents a new strategy which optimizes the multi-stage constant current (MSCC) charging algorithm with state-of-charge (SOC)-based transition conditions (MSCCSOC) by searching for the optimal values of transition state-of-charge (SOC) and charging currents using the coyote optimization algorithm (COA). The paper firstly uses the electrochemical impedance spectroscopy (EIS) analysis to construct the equivalent circuit model (ECM) of the lithium-ion battery, and particle swarm optimization (PSO) is utilized to determine the parameters of the battery ECM within each 1 % SOC. The study employs the COA for the first time to tackle the challenging multi-objective MSCC optimization problem, which involves nine parameters. By not relying on multiple charging experiments and not restricting the search range of SOC transition and charging current values, the proposed approach can identify the global optimal solution, thus being advantageous over existing methods. The proposed method considers both shortening the charging time and reducing the charging losses. The experimental results show that compared with the traditional 1C CC-CV charging method, the proposed strategy can improve the average temperature rise, charging time, and maximum temperature rise by 17.6 %, 34.0 %, and 26.0 %, respectively. Furthermore, the proposed method outperforms other state-of-the-art MSCC charging algorithms and optimization techniques with limited searching range. Therefore, the proposed strategy provides a promising solution for obtaining the optimal setting for MSCCSOC, which can lead to reduced charging time and charging losses, thereby improving the battery's performance and lifespan.