储能科学与技术

• •    

基于电化学阻抗谱几何解析的锂离子电池健康状态评估

严芷涵(), 王学远(), 魏学哲, 戴海峰   

  1. 同济大学汽车学院，上海 201804
  • 收稿日期:2025-07-09 修回日期:2025-08-14
  • 通讯作者: 王学远 E-mail:whhityzh@126.com;7wangxueyuan@tongji.edu.cn
  • 作者简介:严芷涵（1999—），女，硕士研究生在读，研究方向为动力电池健康状态评估，E-mail：whhityzh@126.com；
  • 基金资助:
    无

State of Health Assessment for Lithium-Ion Batteries Based on Geometric Analysis of Electrochemical Impedance Spectroscopy

Zhihan YAN(), Xueyuan WANG(), Xuezhe WEI, Haifeng DAI   

  1. School of Automobiles, Tongji University, Shanghai 201804, Shanghai, China
  • Received:2025-07-09 Revised:2025-08-14
  • Contact: Xueyuan WANG E-mail:whhityzh@126.com;7wangxueyuan@tongji.edu.cn

摘要：

锂离子电池（Lithium-Ion Batteries, LIBs）的健康状态（State of Health, SOH）评估对电动汽车（Electric Vehicles, EVs）和储能系统的安全性与经济性至关重要。针对传统SOH估计方法依赖全频段电化学阻抗谱（Electrochemical Impedance Spectroscopy, EIS）数据、计算复杂度高且跨工况鲁棒性不足的问题，本文提出了一种基于EIS几何解析与分段特征提取的锂离子电池SOH评估方法。通过弛豫时间分布（Distribution of Relaxation Times, DRT）分析识别电池极化过程，构建分段等效电路模型（Equivalent Circuit Model, ECM），在高、中、低频段分别提取欧姆内阻（Ro）、电荷转移阻抗（Charge Transfer Resistance, Rct）和扩散斜率（β）等9维特征参数，显著降低了数据存储与计算需求。为消除温度与荷电状态（State of Charge, SOC）对特征参数的干扰，设计多层感知机（Multilayer Perceptron, MLP）模型将多工况特征映射至标准工况（25℃, 60% SOC），并结合随机森林（Random Forest, RF）算法建立SOH预测模型。实验结果表明，该方法在多种工况下的SOH评估平均绝对误差（Mean Absolute Error, MAE）和平均绝对百分比误差（Mean Absolute Percentage Error, MAPE）分别低至0.00662和0.251735%，优于梯度提升决策树（eXtreme Gradient Boosting, XGBoost）等对比算法。特征重要性分析显示，Rct、Ro和β对SOH预测贡献显著，且单特征参数的独立评估误差仍低于0.42%，验证了该方法的工程适用性。本研究为嵌入式电池管理系统（Battery Management System, BMS）提供了高精度、低复杂度的SOH评估方案，尤其适用于有限硬件资源的实车应用场景。

关键词: 锂离子电池, 健康状态, 电化学阻抗谱, 几何解析, 随机森林

Abstract:

The accurate assessment of the state of health (SOH) for lithium-ion batteries (LIBs) is critical to ensure the safety and cost-effectiveness of electric vehicles (EVs) and energy storage systems. To address the limitations of conventional SOH estimation methods, such as high computational complexity and poor cross-condition robustness due to reliance on full-frequency electrochemical impedance spectroscopy (EIS) data, this study proposes a novel SOH evaluation framework based on geometric analysis and segmented feature extraction of EIS. By employing distribution of relaxation times (DRT) to decouple overlapping polarization processes, a piecewise equivalent circuit model (ECM) is constructed to extract nine-dimensional features from high- frequency, medium- frequency, and low-frequency EIS segments, effectively reducing data storage and computational overhead. A multilayer perceptron (MLP) model is designed to normalize features under varying temperatures and state of charge (SOC) to a standard condition (25°C, 60% SOC), followed by a random forest (RF)-based SOH prediction model. Experimental results demonstrate that the proposed method achieves a mean absolute error (MAE) of 0.00662 and a mean absolute percentage error (MAPE) of 0.251735% across diverse operating conditions, outperforming comparative algorithms like eXtreme Gradient Boosting (XGBoost). Feature importance analysis reveals the dominant roles of Rct, Ro, and β in SOH estimation, while standalone features maintain errors below 0.42%, confirming its practicality for resource-constrained battery management systems (BMS). This work provides a high-precision, low-complexity solution for embedded SOH monitoring in real-world applications.

Key words: lithium-ion batteries, State of Health, Electrochemical Impedance Spectroscopy, geometric analysis, Random Forest

中图分类号: