基于机器学习的钠离子电池剩余使用寿命预测

史永胜; 翟欣然; 胡玙珺

doi:10.6054/j.jscnun.2023031

基于机器学习的钠离子电池剩余使用寿命预测

doi: 10.6054/j.jscnun.2023031

陕西科技大学电气与控制工程学院，西安 710021

基金项目:

国家自然科学基金项目 22279076

陕西省科技厅工业科技攻关计划项目 2019GY-175

详细信息

通讯作者:
翟欣然，Email: 210611008@sust.edu.cn

中图分类号: TM912
计量
- 文章访问数: 117
- HTML全文浏览量: 13
- PDF下载量: 54
- 被引次数: 0
出版历程
- 收稿日期: 2023-05-15
- 刊出日期: 2023-06-25

Machine Learning-based Prediction of the Remaining Life of Sodium-ion Batteries

School of Electrical and Control Engineering, Shaanxi University of Science and Technology, Xi'an 710021, China

摘要

摘要: 钠离子电池容量退化具有非线性和复杂性，准确预测剩余使用寿命(RUL)十分困难，因此构建SSA优化算法结合梯度提升回归树(GBRT)的RUL预测模型。对实验数据进行滤波平滑处理，绘制容量增量(IC)曲线，从IC曲线中提取与容量衰减相关性高的健康指标(HI)，即IC峰值、峰值对应的电压值、峰值面积以及峰值斜率。利用PCA算法对数据进行降维处理以减少数据间的冗余性，将处理后的数据输入GBRT模型，并采用SSA算法寻找最优超参数提高预测精度。利用多组老化实验数据进行验证，RMSE、MAPE和MAE分别在15.2、7%、11.2以下，结果表明该模型有较高的预测精度及稳健性且优于其他主流算法。
- 钠离子电池 /
- 剩余使用寿命 /
- 增量容量曲线 /
- 梯度提升回归树
Abstract: Sodium-ion battery capacity degradation exhibits nonlinearity and complexity, making accurate prediction of remaining useful life (RUL) extremely difficult. Therefore, this paper proposes an RUL prediction model combining Singular Spectrum Analysis (SSA) optimization algorithm with Gradient Boosting Regression Tree (GBRT). Firstly, experimental data is filtered and smoothed, and Incremental Capacity (IC) curves are plotted. The health indicators (HI) with high correlation to capacity decay are extracted from the IC curves. Secondly, PCA algorithm is used to reduce data redundancy, and the processed data is inputted into the GBRT model. SSA algorithm is used to search for the optimal hyperparameters to improve prediction accuracy. Finally, multiple sets of aging experimental data are used for validation, with RMSE, MAPE, and MAE all below 15.2, 7%, and 11.2, respectively. The results show that the proposed model has high prediction accuracy and robustness and outperforms other mainstream algorithms.
- sodium-ion batteries /
- remaining useful life /
- incremental capacity curve /
- Gradient boosting regression tree

HTML全文

图 1 充放电策略

Figure 1. The charging and discharging strategy

下载: 全尺寸图片幻灯片

图 2 钠离子电池的容量衰减曲线

Figure 2. The capacity decay curves of sodium ion batteries

下载: 全尺寸图片幻灯片

图 3 未滤波处理的IC曲线

Figure 3. The unfiltered IC curves

下载: 全尺寸图片幻灯片

图 4 电压平台与IC曲线

Figure 4. The voltage plateau and the IC curves

下载: 全尺寸图片幻灯片

图 5 放电IC曲线

Figure 5. The IC curves for discharge

下载: 全尺寸图片幻灯片

图 6 钠离子电池RUL预测流程

Figure 6. The RUL prediction process for sodium ion batteries

下载: 全尺寸图片幻灯片

图 7 SSA迭代曲线

Figure 7. The fitness curves

下载: 全尺寸图片幻灯片

图 8 RUL预测结果对比以及误差分布

Figure 8. The comparison of RUL prediction results and error distribution

下载: 全尺寸图片幻灯片

图 9 评价指标对比

Figure 9. The comparison of evaluation indicators

下载: 全尺寸图片幻灯片

图 10 评价指标平均值

Figure 10. The average of evaluation indicators

下载: 全尺寸图片幻灯片

表 1 HI与RUL的相关系数

Table 1. The correlation coefficient between health indicators and RUL

充放电倍率	HI1	HI2	HI3	HI4
2.0C	0.937	0.831	0.994	0.757
1.5C	0.924	0.845	0.989	0.701
1.0C	0.941	0.826	0.968	0.797

下载: 导出CSV

表 2 SSA参数设置

Table 2. Parameter settings for SSA

参数名称	数值
变量维度	3
麻雀的总数	30
最大迭代次数	10
预警值	0.8
发现者比例	0.2

下载: 导出CSV

表 3 GBRT参数设置

Table 3. Parameter setting for GBDT

倍率	决策树数量	决策树的最大深度	学习率
2.0C	92	20	0.884 8
1.5C	30	20	0.865 9
1.0C	84	20	0.743 7

下载: 导出CSV

参考文献(19)

[1]	陈鹏飞, 冯杰仪, 吴镝. 高能量密度全固态锂金属电池Li_6.4La₃Zr_1.4Ta_0.6O₁₂基锂硼负极的制备及性能[J]. 华南师范大学学报(自然科学版), 2022, 54(3): 28-33. doi: 10.6054/j.jscnun.2022040 CHEN P F, FENG J Y, WU D. The preparation and performance of Li_6.4La₃Zr_1.4Ta_0.6O₁₂-based lithium boron anode for the high energy density all-solid-state lithium metal battery[J]. Journal of South China Normal University (Natural Science Edition), 2022, 54(3): 28-33. doi: 10.6054/j.jscnun.2022040
[2]	USISKIN R, LU Y, POPOVIC J, et al. Fundamentals, status and promise of sodium-based batteries[J]. Nature Reviews Materials, 2021, 6(11): 1020-1035. doi: 10.1038/s41578-021-00324-w
[3]	菅夏琰, 金俊腾, 王瑶, 等. 钠离子电池层状氧化物正极材料研究进展[J]. 工程科学学报, 2022, 44(4): 601-611. https://www.cnki.com.cn/Article/CJFDTOTAL-BJKD202204013.htm JIAN X Y, JIN J T, WANG Y, et al. Recent progress on layered oxide cathode materials for sodium-ion batteries[J]. Chinese Journal of Engineering, 2022, 44(4): 601-611. https://www.cnki.com.cn/Article/CJFDTOTAL-BJKD202204013.htm
[4]	陈晓秋, 汝强, 王朕, 等. 高容量钠离子电池SnSbCo/rGO负极材料[J]. 华南师范大学学报(自然科学版), 2018, 50(2): 34-37. http://journal-n.scnu.edu.cn/article/id/3954 CHEN X Q, RU Q, WANG Z, et al. SnSbCo/rGO anodes of high capacity sodium ion batteries[J]. Journal of South China Normal University(Natural Science Edition), 2018, 50(2): 34-37. http://journal-n.scnu.edu.cn/article/id/3954
[5]	肖迁, 穆云飞, 焦志鹏, 等. 基于改进LightGBM的电动汽车电池剩余使用寿命在线预测[J]. 电工技术学报, 2022, 37(17): 4517-4527. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202217021.htm XIAO Q, MU Y F, JIAO Z P, et al. Improved LightGBM based remaining useful life prediction of lithium-ion battery under driving conditions[J]. Transactions of China Electrotechnical Society, 2022, 37(15): 3753-3766. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202217021.htm
[6]	ZHU R, CHEN Y, PENG W, et al. Bayesian deep-learning for RUL prediction: an active learning perspective[J]. Reliability Engineering & System Safety, 2022, 228: 108758/1-15.
[7]	AHWIADI M, WANG W. An enhanced particle filter technology for battery system state estimation and RUL prediction[J]. Measurement, 2022, 191: 110817/1-9. doi: 10.1016/j.measurement.2022.110817
[8]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202215004.htm HUANG K, DING H, GUO Y F, 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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202215004.htm
[9]	LI W, FAN Y, RINGBECK F, et al. Unlocking electrochemical model-based online power prediction for lithium-ion batteries via Gaussian process regression[J]. Applied Energy, 2022, 306: 118114/1-16. doi: 10.1016/j.apenergy.2021.118114
[10]	LUO K, CHEN X, ZHENG H, et al. A review of deep learning approach to predicting the state of health and state of charge of lithium-ion batteries[J]. Journal of Energy Chemistry, 2022, 74: 159-173. doi: 10.1016/j.jechem.2022.06.049
[11]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202117015.htm XU J N, NI Y L, ZHU C B. 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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202117015.htm
[12]	CHEHADE A A, HUSSEIN A A. A collaborative Gaussian process regression model for transfer learning of capacity trends between Li-ion battery cells[J]. IEEE Transactions on Vehicular Technology, 2020, 69(9): 9542-9552.
[13]	QIN P, ZHAO L, LIU Z. State of health prediction for lithium-ion battery using a gradient boosting-based data-driven method[J]. Journal of Energy Storage, 2022, 47: 103644/1-23.
[14]	刘素贞, 袁路航, 张闯, 等. 基于超声时域特征及随机森林的磷酸铁锂电池荷电状态估计[J]. 电工技术学报, 2022, 37(22): 5872-5885. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202222021.htm LIU S Z, YUAN L H, ZHANG C, et al. State of charge estimation of LiFeO₄ batteries based on time domain features of ultrasonic waves and random forest[J]. Transactions of China Electrotechnical Society, 2022, 37(22): 5872-5885. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202222021.htm
[15]	ROMAN D, SAXENA S, ROBU V, et al. Machine learning pipeline for battery state-of-health estimation[J]. Nature Machine Intelligence, 2021, 3(5): 447-456.
[16]	ZHU J, DEWI DARMA M S, KNAPP M, et al. Investigation of lithium-ion battery degradation mechanisms by combining differential voltage analysis and alternating current impedance[J]. Journal of Power Sources, 2020, 448: 227575/1-12.
[17]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202223014.htm ZHOU C J, WANG Y J, LI K Q, 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. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202223014.htm
[18]	LI X, WANG Z, ZHANG L, et al. State-of-health estimation for Li-ion batteries by combing the incremental capacity analysis method with grey relational analysis[J]. Journal of Power Sources, 2019, 410/411: 106-114.
[19]	LIN C P, CABRERA J, YANG F, et al. Battery state of health modeling and remaining useful life prediction through time series model[J]. Applied Energy, 2020, 275: 115338/1-12.