摘 要：针对锂离子电池健康状态(SOH)精确估计的难题，本文提出一种融合健康特征参数的PSO-BP算法。通过分析电池充放电过程中的电压与温度曲线变化规律，从电压、温度数据中提取出两个与电池老化高度相关的健康特征参数(HF1与HF2),并采用皮尔逊、斯皮尔曼和肯德尔相关系数验证其与SOH的高度相关性。使用NASA 18650锂电池数据集进行验证，结果表明，该模型预测结果与真实值拟合度高，显著提升了SOH估算精度，为电池管理系统提供了有效技术支撑。
关键词：锂离子电池；健康特征参数；粒子群优化(PSO)；BP神经网络
中图分类号：TM912.9 文献标志码：A DOI：10.15917/j.cnki.1006-3331.2025.05.003
Research on SOH Estimation of Lithium Batteries Using PSO-BP Algorithm with Integrated Health Feature Parameters
LEI Jianxin,ZHOU Hang,ZHONG Xinhao,XIAO Ye,WANG Fan
Abstract: Aiming to address the challenge of accurately estimating the State of Health(SOH)of lithium-ion batteries,this paper proposes a PSO-BP algorithm integrated with health feature parameters.By analyzing the variation patterns of voltage and temperature curves during battery charging and discharging,two health feature parameters(HF1 and HF2)strongly correlated with battery aging are extracted from the cor-responding data.Pearson,Spearman,and Kendall correlation coefficients are employed to verify the strong association between these features and SOH.Validation conducted on the NASA 18650 lithium-ion battery dataset shows that the model's predictions achieve a high goodness-of-fit with the true values,significantly improving the accuracy of SOH estimation and offering effective technical support for battery management systems.
Key words: lithium-ion battery; health feature parameters; PSO; BP neural network
