基于数据预处理和VMD-LSTM-GPR的锂离子电池剩余寿命预测

doi:10.19595/j.cnki.1000-6753.tces.230210

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (437 KB)
输出: BibTeX | EndNote (RIS)

摘要

锂离子电池的剩余使用寿命（RUL）是健康管理中重要参数,其准确评估对于保证电池设备的安全稳定运行非常重要。本文提出一种数据预处理联合变分模态分解（VMD）、长短期记忆网络（LSTM）和高斯回归过程（GPR）的预测框架。首先选取充放电循环过程中的信息作为间接健康因子（HI）,并通过核主元分析方法（KPCA）实现间接HI的特征提取,完成数据预处理;其次通过VMD-LSTM方法实现健康因子的分解、预测和重构,并将重构得到的数据应用于RUL预测的GPR模型,完成预测模型搭建;最后以NASA锂电池数据集作为算法测试数据,结果表明,所提取的健康因子能够准确跟踪锂电池的退化过程;所提预测方法能够准确的估计电池的剩余寿命,同时具有较高的可靠性和稳定性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	李英顺
	阚宏达
	郭占男
	王德彪
	王铖

关键词 ：锂离子电池, 剩余寿命, 健康因子, 变分模态分解, 高斯回归过程, 长短期记忆

Abstract：

The performance of lithium-ion battery will deteriorate with the decrease of capacity and the increase of impedance during the continuous charging and discharging process, which can lead to equipment and system failures and even catastrophic losses. Therefore, achieving accurate and reliable prediction of the remaining useful life (RUL) of lithium-ion battery is crucial in scientific research and practical application of battery management system design. However, onthe one hand,probability prediction could not be supported and point prediction could be only supported in previous prediction methods, which did not provide a clear mathematical representation of the confidence level of the prediction results. On the other hand, the noise and capacity rebound phenomena in the original data was ignored. To address these issues, this paper proposed a prediction framework based on data pre-processing combined with variational mode decomposition (VMD), long and short-term memory network (LSTM) and Gaussian regression process (GPR).
Firstly, the indirect health indicator(HI) reflecting the life degradation trend of lithium-ion battery was extracted from the charge/discharge curve. Pearson and Spearman correlation coefficients were used to verify the correlation between the extracted indirect HI and capacity. The kernel principal element analysis(KPCA)method was used to remove the redundant components of indirect HI to transform into fusion HI to reduce the computational complexity. Then, the VMD decomposition method was applied to decompose the obtained fusion HI into multiple modal components. Based on the central frequency of the modal components and the capacity-related coefficient, multiple modal components are divided into three parts: global attenuation, local regeneration and other noise. Further, the time series prediction of the global attenuation and local regeneration components were performed separately using LSTM neural network to obtain the predicted values after the prediction starting point, and the values of each component were summed up correspondingly to obtain the reconstructed HI. Finally, the reconstructed HI and capacity were used as the input and output of the GPR model respectively to achieve the RUL prediction of lithium-ion battery.
NASA lithium-ion battery public data set was used for experimental validation, and different prediction starting points were set for each battery, with the starting point set at about 40% of the total cycle time. The experimental results show that the predicted values of capacity can track the battery aging trend well, and also have good estimation ability for the local regeneration phenomenon of capacity regeneration during the aging process. The true value of capacity basically falls in the 95% confidence interval of the predicted value, while the prediction effect of the model gets better as the training data increases. Among the evaluation indexes of capacity prediction results, the maximum values of root mean square error, mean absolute error and mean absolute percentage error are 2.98%, 2.34% and 1.81%, respectively, and the errors of remaining useful life prediction are all within 2 cycles.
The following conclusions can be drawn from the experimental analysis: (1) The KPCA algorithm can remove the redundant information between indirect health indicatorto reduce the data complexity and achieve the data pre-processing. (2) The VMD decomposition method can mine the intrinsic information in the data and capture the long-term downward trend, local regeneration and noise component implied in the fusion HI, so as to remove the implied noise component in the data. (3) The GPR model can support probabilistic prediction and give confidence intervals for the capacity prediction results.

Key words： Lithium-ion battery remaining useful life health indicator variational mode decomposition Gaussian regression process long and short-term memory

收稿日期: 2023-02-22

PACS:

TM912

基金资助:

辽宁省‘兴辽英才计划'项目资助（XLYC1903015）

通讯作者: 阚宏达, 男,1997年生,硕士,研究方向为火控计算机的故障诊断与锂电池的寿命预测。E-mail：1780623034@mail.dlut.edu.cn

作者简介: 李英顺, 女,1971年生,博士后,教授,研究方向为火控系统的故障预测与健康管理。E-mail：1982215089@qq.com

引用本文:

李英顺, 阚宏达, 郭占男, 王德彪, 王铖. 基于数据预处理和VMD-LSTM-GPR的锂离子电池剩余寿命预测[J]. 电工技术学报, 0, (): 9017-17. Li Yingshun, Kan Hongda, Guo Zhannan, Wang Debiao, Wang Cheng. Prediction of Remaining Useful Life of Lithium-Ion BatteryBased on Data Preprocessing and VMD-LSTM- GPR. Transactions of China Electrotechnical Society, 0, (): 9017-17.

链接本文:

https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.230210 https://dgjsxb.ces-transaction.com/CN/Y0/V/I/9017

[1] 周才杰,汪玉洁,李凯铨,等. 基于灰色关联度分析-长短期记忆神经网络的锂离子电池健康状态估计[J]. 电工技术学报,2022,37(23): 6065-6073.
Zhou Caijie, Wang Yujie, Li Kaiquan, et al.State of health estimation for lithium-ion battery based on gray correlation analysis and long short-term memory neural network[J]. Transactions of China Electrotechnical Society,2022, 37(23): 6065-6073.
[2] 黄凯,丁恒,郭永芳,等. 基于数据预处理和长短期记忆神经网络的锂离子电池寿命预测[J].电工技术学报, 2022, 37(15): 3753-3766.
Huang Kai, Ding Heng, Guo Yongfang, et al.Prediction of remaining useful life of lithium-ion battery based on adaptive data preprocessing and long short-term memory network[J]. Transactions of China Electrotechnical Society, 2022, 37(15): 3753-3766.
[3] 王萍,弓清瑞,张吉昂,等. 一种基于数据驱动与经验模型组合的锂电池在线健康状态预测方法[J]. 电工技术学报, 2021, 36(24): 5201-5212.
Wang Ping, Gong Qingrui, Zhang Jiang, et al.An online state of health prediction method for lithium batteries based on combination of data-driven and empirical model[J]. Transactions of China Electrotechnical Society, 2021, 36(24): 5201-5212.
[4] Liu Jian, Chen Ziqiang.Remaining useful life prediction of lithium-ion batteries based on health indicator and gaussian process regression model[J]. IEEE Access, 2019, 7:39474-39484.
[5] Ding Guorong, Wang Wenbo, Zhu Ting.Remaining useful life prediction for lithium-ion batteries based on CS-VMD and GRU[J]. IEEE Access, 2022, 10:89402-89413.
[6] 王义军, 左雪. 锂离子电池荷电状态估算方法及其应用场景综述[J]. 电力系统自动化, 2022,46(14):193-207.
Wang Yijun, Zuo Xue.Review on estimation methods for state of charge of lithium-ion battery and their application scenarios[J].Automation of Electric Power Systems, 2022,46(14):193-207.
[7] 武龙星,庞辉,晋佳敏,等. 基于电化学模型的锂离子电池荷电状态估计方法综述[J]. 电工技术学报, 2022, 37(7): 1703-1725.
Wu Longxing, Pang Hui,Jin Jiamin, et al.A review of SOC estimation methods for lithium-ion batteries based on electrochemical model[J].Transactions of China Electrotechnical Society, 2022, 37(7): 1703-1725.
[8] 颜湘武,邓浩然,郭琪,等. 基于自适应无迹卡尔曼滤波的动力电池健康状态检测及梯次利用研究[J]. 电工技术学报,2019,34(18):3937-3948.
Yan Xiangwu, Deng Haoran, Guo Qi, et al.Study on the state of health detection of power batteries based on adaptive unscented Kalman filters and the battery echelon utilization[J]. Transactions of China Electrotechnical Society, 2019,34(18):3937-3948.
[9] Bian Zengyuan, Ma Yan.An improved particle filter method to estimate state of health of lithium-ion battery[J]. IFAC-PapersOnLine, 2021, 54(10):344-349.
[10] Duan Bin, Zhang Qi, Geng Fei, et al.Remaining useful life prediction of lithium‐ion battery based on extended Kalman particle filter[J]. International Journal of Energy Research, 2019,44(3): 1724-1734.
[11] 贺宁,钱成,李若夏. 自适应模型与改进粒子滤波的电池RUL预测[J]. 哈尔滨工业大学学报, 2022, 54(9): 111-121.
He Ning, Qian Cheng, Li Ruoxia.RUL prediction for lithium-ion batteries via adaptive modeling and improved particle filter[J]. Journal of Harbin Institute of Technology, 2022, 54(9): 111-121.
[12] 徐佳宁,倪裕隆,朱春波. 基于改进支持向量回归的锂电池剩余寿命预测[J]. 电工技术学报, 2021, 36(17): 3693-3704.
Xu Jianing, Ni Yulong, Zhu Chunbo.Remaining useful life prediction for lithium-ion batteries based on improved support vector regression[J].Transactions of China Electrotechnical Society, 2021, 36(17): 3693-3704.
[13] Liu Datong, Zhou Jianbao, Pan Dawei, et al.Lithium-ion battery remaining useful life estimation with an optimized relevance vector machine algorithm with incremental learning[J]. Measurement, 2015, 63:143-151.
[14] Zhu Jun, Tan Tianxiong, Wu Lifeng, et al.RUL prediction of lithium-ion battery based on improved DGWO-ELM method in a random discharge rates environment[J]. IEEE Access, 2019, 7:125176-125187.
[15] Phattara K, Yodo N.A data-driven predictive prognostic model for lithium-ion batteries based on a deep learning algorithm[J]. Energies, 2019, 12(4): 660.
[16] 王萍,范凌峰,程泽. 基于健康特征参数的锂离子电池SOH和RUL联合估计方法[J]. 中国电机工程学报, 2022, 42(4): 1523-1534.
Wang Ping, Fan Lingfeng, Cheng Zhe.A joint state of health and remaining useful life estimation approach for lithium-ion batteries based on health factor parameter[J].
Proceedings of the CSEE, 2022,42(04): 1523-1534.
[17] Cheng Gong, Wang Xinzhi, He Yurong.Remaining useful life and state of health prediction for lithium batteries based on empirical mode decomposition and a long and short memory neural network[J]. Energy, 2021,232: 121022.
[18] Qu Jingtao, Liu Feng, Ma Yuxing, et al.A neural-network-based method for RUL prediction and SOH monitoring of lithium-ion battery[J]. IEEE Access, 2019,7: 87178-87191.
[19] 杨彦茹,温杰,史元浩,等. 基于CEEMDAN和SVR的锂离子电池剩余使用寿命预测[J]. 电子测量与仪器学报,2020,34(12):197-205.
Yang Yanru, Wen Jie, Shi Yuanhao, et al.Remaining useful life prediction for lithium-ion battery based on CEEMDAN and SVR[J].Journal of Electronic Measurement and Instrument,2020,34(12):197-205.
[20] Sun Chuang, Qu An, Zhang Jun, et al.Remaining useful life prediction for lithium-ion batteries based on improved variational mode decomposition and machine learning algorithm[J]. Energies, 2023,16(1): 313.
[21] Saha B, KG. Battery data set.NASA Ames Prognostics Data Repository [DB/OL]. NASA Ames Research Center, Moffett Field, CA2017, https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository.
[22] Hell S M, Kim C D.Development of a data-driven method for online battery remaining-useful-life prediction[J]. Batteries, 2022, 8(10): 192.
[23] Jia Shun, Ma Bo, Guo Wei, et al.A sample entropy based prognostics method for lithium-ion batteries using relevance vector machine[J]. Journal of Manufacturing Systems, 2021, 61:773-781.
[24] 王竹晴,郭阳明,徐聪. 基于SAE-VMD的锂离子电池健康因子提取方法[J]. 西北工业大学学报, 2020, 38(4): 814-821.
Wang Zhuqing, Guo Yangming, Xu Cong.An HIextraction framework for lithium-ion battery prognostics based on SAE-VMD[J]. Journal of Northwestern Polytechnical University, 2020, 38(4): 814-821.
[25] Jia Jianfang, Liang Jianyu, Shi Yuanhao, et al.SOH and RUL Prediction of lithium-ion batteries based on gaussian process regression with indirect health indicators[J]. Energies, 2020, 13(2): 375.
[26] Zhang Yagang,Li Ruixuan, Zhang Jinghui.Optimization scheme of wind energy prediction based on artificial intelligence[J]. Environmental Science and Pollution Research, 2021,28(29):39966-39981.
[27] 王瀛洲,倪裕隆,郑宇清,等. 基于ALO-SVR的锂离子电池剩余使用寿命预测[J]. 中国电机工程学报, 2021, 41(4): 1445-1457, 1550.
Wang Yingzhou, Ni Yulong, Zheng Yuqing, et al.Remaining useful life prediction of lithium-ion batteries based on support vector regression optimized and ant lion optimizations[J]. Proceedings of the CSEE, 2021, 41(4): 1445-1457,1550.