<p>The State of Charge (SOC) of lithium-ion batteries is one of the important parameters in Battery Management Systems (BMS). Accurate and rapid SOC estimation is crucial for enhancing the safety of battery applications. However, lithium-ion batteries are highly complex nonlinear time-varying systems. This complexity increases the difficulty in SOC estimation. We proposed a SOC estimation method which combines the Improved Least Squares Support Vector Machine (ILSSVM) with the Tornado Optimizer with Coriolis force (TOC), which enables rapid and accurate SOC estimation. To begin with, we use principal component analysis (PCA), Pearson correlation analysis, and Spearman correlation analysis to screen the factors that affect SOC. Following this, we improve the Least Squares Support Vector Machine (LSSVM) by introducing a truncation factor which achieves sparsity in a more direct and rapid manner, reduces the impact of boundary samples on the algorithm, thereby improving the algorithm’s running speed. Subsequently, we use the Tornado Optimizer with Coriolis force to optimize the kernel function parameter and regularization factor of the ILSSVM which further enhances the estimation accuracy. Specifically, the average running time of the algorithm is reduced by 41.7% and 27.1% compared with traditional SVM and LSSVM. The RMSE under HPPC, UDDS, DST and FUDS condition are 0.48%, 0.18%, 0.38% and 0.17%. Compared with mainstream baselines including EKF, LSTM, and XGBoost, its accuracy is improved by more than 30%.</p>