High Precision State of Health Estimation of Lithium-ion Batteries Based on PSO-LSSVM AdaBoost Model

• Published in: 2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) pp. 570 - 574
• Main Authors: Feng, Renjun, Wang, Shunli, Yu, Chunmei, Cui, Yixiu, Zhou, Yifei
• Format: Conference Proceeding
• Language: English
• Published: IEEE 21.07.2024
• Subjects: Adaboost algorithm; Battery management systems; Estimation; Feature extraction; health features; Kernel; least squares support vector machine; Lithium-ion batteries; NASA; Particle swarm optimization; Prediction algorithms; Predictive models; state of health; Support vector machines
• DOI: 10.1109/CCSSTA62096.2024.10691790

The accurate estimation of the state of health (SOH) of lithium-ion batteries is paramount for the maintenance and longevity of battery management systems (BMS). To enhance the precision and robustness of SOH prediction models for lithium-ion batteries, this article introduces a novel approach that utilizes a particle swarm optimization (PSO) algorithm to optimize the least squares support vector machine (LSSVM) model for battery SOH estimation. The study focuses on three individual batteries from the NASA battery dataset as the primary research subjects. The PSO algorithm is then employed to optimize the regularization parameters and kernel parameters of the LSSVM, ensuring that the model is tailored to the specific characteristics of the battery data. Furthermore, the AdaBoost algorithm is integrated into the framework to train the PSO-optimized LSSVM, aiming to further boost the model's predictive capabilities. The experimental results demonstrate the effectiveness and reliability of the proposed model, indicating its potential for accurate SOH estimation in lithium-ion batteries.