电气工程学报 ›› 2022, Vol. 17 ›› Issue (4): 32-40.doi: 10.11985/2022.04.005

• 特邀专栏:电化学储能系统安全管理与运维 • 上一篇    下一篇

扫码分享

基于双高斯模型的锂电池剩余使用寿命预测方法*

李彦梅1(), 刘惠汉1(), 张朝龙1,2(), 罗来劲1()   

  1. 1.安庆师范大学电子工程与智能制造学院 安庆 246011
    2.金陵科技学院智能科学与控制工程学院 南京 211169
  • 收稿日期:2022-08-06 修回日期:2022-09-30 出版日期:2022-12-25 发布日期:2023-02-03
  • 通讯作者: 张朝龙,男,1982年生,博士,教授。主要研究方向为动力电池测试技术,故障诊断和预测。E-mail:zhangchaolong@126.com
  • 作者简介:李彦梅,女,1973年生,教授。主要研究方向为动力电池性能检测技术。E-mail:liym@aqnu.edu.cn
    刘惠汉,男,1997年生,硕士研究生。主要研究方向为动力电池管理技术。E-mail:luihuihan@163.com
    罗来劲,男,1998年生,硕士研究生。主要研究方向为动力电池健康状态估计。E-mail:laijinluo@163.com
  • 基金资助:
    *国家重点研发计划(2020YFB0905905);国家重点研发计划(2016YFF0102200);国家自然科学基金(51637004);安徽省级质量工程“四新”研究与改革实践(2021sx092)

Lithium-ion Battery RUL Prediction Method Based on Double Gaussian Model

LI Yanmei1(), LIU Huihan1(), ZHANG Chaolong1,2(), LUO Laijing1()   

  1. 1. School of Electronic Engineering and Intelligent Manufacturing, Anqing Normal University, Anqing 246011
    2. College of Intelligent Science and Control Engineering, Jinling Institute of Technology, Nanjing 211169
  • Received:2022-08-06 Revised:2022-09-30 Online:2022-12-25 Published:2023-02-03
  • Contact: ZHANG Chaolong, E-mail:zhangchaolong@126.com

摘要:

准确的剩余使用寿命预测对锂离子电池的性能最大化和维护是至关重要的。为了对锂离子电池的剩余使用寿命进行精准预测,提出一种新颖的双高斯模型用于描述锂离子电池老化过程。首先对常用的几种电池容量衰减经验模型进行分析与评价,并提出性能更优的双高斯模型。随后,基于历史容量数据,利用粒子滤波(Particle filter,PF)技术建立双高斯老化模型,同时引入拟合相关系数与均方根误差评估模型。最后,根据实验室的单体电池老化数据和美国国家航空航天局的电池老化数据,进行剩余使用寿命预测试验,以验证所提出的老化模型的有效性。试验结果表明,所提出的老化模型可以准确地预测锂电池剩余使用寿命,与其他模型相比,预测误差得到明显改善。

关键词: 锂离子电池, 剩余使用寿命, 双高斯模型, 粒子滤波算法

Abstract:

For the performance maximization and maintenance of lithium-ion batteries, accurate remaining useful life(RUL) predictions are essential. To accurately predict the RUL of lithium-ion batteries, a novel double Gaussian model is proposed to describe the aging process of lithium-ion batteries. Specifically, several popular empirical models for battery capacity degradation are analyzed and evaluated, and a double Gaussian model with better performance is proposed. Afterward, a double Gaussian aging model is established utilizing the particle filter(PF) technique, based on the historical capacity data. The fitted correlation coefficient and root mean square error are also introduced to assess the model. Finally, the RUL prediction experiments are conducted to verify the verification of the proposed aging model based on the battery aging data from the laboratory’s battery cells and the National Aeronautics and Space Administration(NASA) Ames Prognostics Center of Excellence. The experimental results demonstrate that the proposed aging model can predict the RUL accurately, and the prediction error is significantly improved compared to other models.

Key words: Lithium-ion battery, RUL prediction, double Gaussian model, particle filter(PF)

中图分类号: