基于SDAE-Transformer-ECA网络的锂电池剩余使用寿命预测

doi:10.19799/j.cnki.2095-4239.2023.0369

摘要/Abstract

摘要：

锂离子电池剩余使用寿命(remaining useful life，RUL)的准确预测对于提高电池使用寿命、降低异常事故的概率，起着至关重要的作用。本文结合堆叠噪声自编码器(stacked denoising auto encoder，SDAE)和变压器(transformer)的优势，提出了一种结合高效通道注意力(efficient channel attention，ECA)的SDAE-Transformer-ECA的锂离子电池RUL预测网络。首先，针对电池在使用过程中存在的容量再生现象和数据集采集误差等噪声污染，利用SDAE对输入数据进行重构去噪、提取特征。然后，通过Transformer网络对重构数据进行序列信息的捕获。最后，结合ECA网络对捕获信息进行跨通道整合和交互，从而实现锂离子电池的RUL的预测。本文先基于美国马里兰大学先进生命周期工程中心(Center for Advanced Life Cycle Engineering，CALCE)提供的电池容量数据集进行实验验证，实验证明本文模型的各项误差都较低，具有较高的准确性，且与次优算法Bi-LSTM相比，平均RE相对降低了62.67%，平均MAE相对降低了40.68%，平均RMSE相对降低了34.33%。再使用美国航空航天局(National Aeronautics and Space Administration，NASA)提供的B0007号电池容量数据集进行泛化性验证，实验得到的RE、MAE和RMSE结果分别是1.98%、3.12%和4.16%，与RNN、LSTM、GRU和Bi-LSTM等现有算法相比，本文模型预测准确性更高，证明了该模型的泛化性。

关键词: 锂离子电池, SDAE, Transformer, 注意力机制, 剩余使用寿命预测

Abstract:

The accurate prediction of the remaining useful life (RUL) of lithium-ion batteries plays a crucial role in improving the battery life and reducing the probability of accidents. This study combines the advantages of a stacked denoising autoencoder (SDAE) and a transformer to propose a lithium-ion battery RUL prediction network that combines the SDAE, transformer, and efficient channel attention (ECA). Considering the noise pollution brought about by the capacity regeneration phenomenon and the dataset acquisition error during battery usage, the SDAE is used to reconstruct and denoise the input data and extract the features. The sequence information of the reconstructed data is then captured through the transformer network. Finally, the cross-channel integration and interaction of the captured information are performed in combination with the ECA network to realize the RUL prediction of the lithium-ion batteries. This study uses the battery capacity dataset provided by the Center for Advanced Life Cycle Engineering at the University of Maryland. The experimental results show that the proposed algorithm has low error and high accuracy. Compared with that for the suboptimal bi-directional long short-term memory (Bi-LSTM) algorithm, the average RE, mean absolute error (MAE), and root-mean-squared error (RMSE) for the proposed algorithm are relatively reduced by 62.67%, 40.68%, and 34.33%, respectively. Using the B0007 battery capacity dataset provided by the National Aeronautics and Space Administration for generalization verification, the experimental results of the RE, MAE, and RMSE were found to be 1.98%, 3.12%, and 4.16%, respectively. With that being said, the prediction accuracy of the proposed algorithm is higher than that of existing algorithms, such as recurrent neural networks, LSTM, gated recurrent units, and Bi-LSTM. Thus, the generalization of the model is proven.

Key words: lithium-ion battery, SDAE, transformer, attention mechanism, remaining useful life prediction

中图分类号:

TM 912

宋兴海, 张小乾, 梁惠施, 史梓男, 李棉刚, 周奎, 贡晓旭. 基于SDAE-Transformer-ECA网络的锂电池剩余使用寿命预测[J]. 储能科学与技术, 2023, 12(10): 3181-3190.

Xinghai SONG, Xiaoqian ZHANG, Huishi LIANG, Zinan SHI, Miangang LI, Kui ZHOU, Xiaoxu GONG. Predicting the remaining service life of lithium batteries based on the SDAE-transformer-ECA network[J]. Energy Storage Science and Technology, 2023, 12(10): 3181-3190.

图/表 9

图1

图2

图3

表1

表2

图4

图5

图6

图7

参考文献 29

1	GAO Y Z, ZHANG X, GUO B J, et al. Health-aware multiobjective optimal charging strategy with coupled electrochemical-thermal-aging model for lithium-ion battery[J]. IEEE Transactions on Industrial Informatics, 2020, 16(5): 3417-3429.
2	SU C, CHEN H J, WEN Z J. Prediction of remaining useful life for lithium-ion battery with multiple health indicators[J]. Eksploatacjai Niezawodność‍-Maintenance and Reliability, 2021, 23(1): 176-183.
3	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 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.
4	段双明, 杨耀微. 基于分数阶模型的锂电池SOC估计[J]. 电源技术, 2022, 46(8): 862-866.
	DUAN S M, YANG Y W. State of charge estimation of lithium batteries based on fractional model[J]. Chinese Journal of Power Sources, 2022, 46(8): 862-866.
5	叶乙福. 基于数据驱动的锂离子电池剩余使用寿命预测方法研究[D]. 湖州: 湖州师范学院, 2022.
	YE Y F. Research on prediction method of remaining useful life of lithium-ion battery based on data drive[D]. Huzhou: Huzhou University, 2022.
6	李彦梅, 刘惠汉, 张朝龙, 等. 基于双高斯模型的锂电池剩余使用寿命预测方法[J]. 电气工程学报, 2022, 17(4): 32-40.
	LI Y M, LIU H H, ZHANG C L, et al. Lithium-ion battery RUL prediction method based on double Gaussian model[J]. Journal of Electrical Engineering, 2022, 17(4): 32-40.
7	GUHA A, PATRA A. Online estimation of the electrochemical impedance spectrum and remaining useful life of lithium-ion batteries[J]. IEEE Transactions on Instrumentation and Measurement, 2018, 67(8): 1836-1849.
8	宋胜, 李云伍, 赵颖, 等. 锂离子电池片段数据的荷电状态估计研究[J]. 电源技术, 2022, 46(7): 734-738.
	SONG S, LI Y W, ZHAO Y, et al. Research on SOC estimation based on fragment data of lithium-ion battery[J]. Chinese Journal of Power Sources, 2022, 46(7): 734-738.
9	YANG J S, FANG W G, CHEN J Y, et al. A lithium-ion battery remaining useful life prediction method based on unscented particle filter and optimal combination strategy[J]. Journal of Energy Storage, 2022, 55: 105648.
10	LIU K L, SHANG Y L, OUYANG Q, et al. A data-driven approach with uncertainty quantification for predicting future capacities and remaining useful life of lithium-ion battery[J]. IEEE Transactions on Industrial Electronics, 2020, 68(4): 3170-3180.
11	LI X Y, SHU X, SHEN J W, et al. An on-board remaining useful life estimation algorithm for lithium-ion batteries of electric vehicles[J]. Energies, 2017, 10(5): 691.
12	何冰琛, 杨薛明, 王劲松, 等. 基于PCA-GPR的锂离子电池剩余使用寿命预测[J]. 太阳能学报, 2022, 43(5): 484-491.
	HE B C, YANG X M, WANG J S, et al. Prediction of remaining useful life of lithium-ion batteries based on pca-gpr[J]. Acta Energiae Solaris Sinica, 2022, 43(5): 484-491.
13	刘健, 陈自强, 黄德扬, 等. 基于等压差充电时间的锂离子电池寿命预测[J]. 上海交通大学学报, 2019, 53(9): 1058-1065.
	LIU J, CHEN Z Q, HUANG D Y, et al. Remaining useful life prediction for lithium-ion batteries based on time interval of equal charging voltage difference[J]. Journal of Shanghai Jiao Tong University, 2019, 53(9): 1058-1065.
14	何畏, 罗潇, 曾珍, 等. 利用QPSO改进相关向量机的电池寿命预测[J]. 电子测量与仪器学报, 2020, 34(6): 18-24.
	HE W, LUO X, ZENG Z, et al. Battery life prediction based on QPSO improved relevant vector machine[J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(6): 18-24.
15	徐帅, 刘雨辰, 周飞. 基于RNN的锂离子电池SOC估算研究进展[J]. 电源技术, 2021, 45(2): 263-269.
	XU S, LIU Y C, ZHOU F. Research progress of SOC estimation of lithium ion batteries based on RNN[J]. Chinese Journal of Power Sources, 2021, 45(2): 263-269.
16	ZHANG Y Z, XIONG R, HE H W, et al. Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Transactions on Vehicular Technology, 2018, 67(7): 5695-5705.
17	高峰, 贾建芳, 元淑芳, 等. 基于GRU-UKF的锂离子电池SOC估计方法研究[J]. 电子测量与仪器学报, 2022, 36(11): 160-169.
	GAO F, JIA J F, YUAN S F, et al. Research on SOC estimation method of lithium-ion battery based on GRU-UKF[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(11): 160-169.
18	闫志远, 孙桓五, 刘世闯, 等. 一种基于GRU的氢燃料重卡汽车工况下锂离子电池温度预测模型[J/OL]. 中国电机工程学报: 1-12[2023-04-26]. https: //doi.org/10.13334/j.0258-8013.pcsee.221789.
	YAN Z Y, SUN H W, LIU S C, et al. Temperature prediction model of lithium-ion battery under working conditions of hydrogen fuel heavy trucks based on GRU[J/OL]. Proceedings of the CSEE: 1-12[2023-04-26]. https: //doi.org/10.13334/j.0258-8013.pcsee.221789.
19	YAYAN U, ARSLAN A T, YUCEL H. A novel method for SoH prediction of batteries based on stacked LSTM with quick charge data[J]. Applied Artificial Intelligence, 2021, 35(6): 421-439.
20	邢子轩, 张凡, 武明虎, 等. 基于WD-GRU的锂离子电池剩余寿命预测[J]. 电源技术, 2022, 46(8): 867-871.
	XING Z X, ZHANG F, WU M H, et al. Remaining life prediction of lithium ion batteries based on WD-GRU[J]. Chinese Journal of Power Sources, 2022, 46(8): 867-871.
21	张朝龙, 何怡刚, 袁莉芬. 基于EEMD和MKRVM的锂电池剩余寿命预测方法[J]. 电力系统及其自动化学报, 2018, 30(7): 38-44.
	ZHANG C L, HE Y G, YUAN L F. Prediction approach for remaining useful life of lithium-ion battery based on EEMD and MKRVM[J]. Proceedings of the CSU-EPSA, 2018, 30(7): 38-44.
22	赵沁峰, 蔡艳平, 王新军. 锂电池在不同放电区间下的剩余寿命预测[J]. 中国测试, 2023, 49(3): 159-165, 180.
	ZHAO Q F, CAI Y P, WANG X J. Remaining useful life prediction of lithium battery under different discharge intervals[J]. China Measurement & Test, 2023, 49(3): 159-165, 180.
23	刘芊彤, 邢远秀. 基于VMD-PSO-GRU模型的锂离子电池剩余寿命预测[J]. 储能科学与技术, 2023, 12(1): 236-246.
	LIU Q T, XING Y X. Remaining life prediction of lithium-ion battery based on VMD-PSO-GRU model[J]. Energy Storage Science and Technology, 2023, 12(1): 236-246.
24	WANG F K, AMOGNE Z E, CHOU J H, et al. Online remaining useful life prediction of lithium-ion batteries using bidirectional long short-term memory with attention mechanism[J]. Energy, 2022, 254: 124344.
25	LI L, LI Y J, MAO R Z, et al. Remaining useful life prediction for lithium-ion batteries with a hybrid model based on TCN-GRU-DNN and dual attention mechanism[J]. IEEE Transactions on Transportation Electrification, 2023, doi: 10.1109/TTE.2023.3247614.
26	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all You need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. New York: ACM, 2017: 6000-6010.
27	VINCENT P, LAROCHELLE H, LAJOIE I, et al. Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion[J]. Journal of Machine Learning Research, 2010, 11: 3371-3408.
28	WANG Q L, WU B G, ZHU P F, et al. ECA-net: Efficient channel attention for deep convolutional neural networks[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2020: 11531-11539.
29	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, 2018: 7132-7141.

参数名	参数设置
采样窗口s	64
模型学习速率lr	0.005
Transformer网络深度d	1
Transformer网络的隐藏层大小h	32
学习正则化α	10^-6
优化函数	Adam
损失函数loss	MSE

算法	RE/%	MAE/%	RMSE/%
RNN	6.51	9.10	11.39
LSTM	3.94	3.77	4.78
GRU	5.54	4.28	5.11
Bi-LSTM	3.08	4.13	5.01
Transformer	79.92	60.88	70.80
SDAE-Transformer	6.33	3.13	4.27
SDAE-Transformer-ECA	1.15	2.45	3.29

[1]	李悦, 王博, 吴楠. 石墨烯/Si/SiO _x 纳米复合材料的制备及储锂性能研究[J]. 储能科学与技术, 2023, 12(9): 2752-2759.
[2]	曾少鸿, 吴伟雄, 刘吉臻, 汪双凤, 叶石丰, 冯振宇. 锂离子电池浸没式冷却技术研究综述[J]. 储能科学与技术, 2023, 12(9): 2888-2903.
[3]	李纪伟, 刘睿涵, 吕桃林, 潘隆, 马常军, 李清波, 赵芝芸, 杨文, 解晶莹. 基于局部离群点检测和标准差方法的锂离子电池组早期故障诊断[J]. 储能科学与技术, 2023, 12(9): 2917-2926.
[4]	黄晓伟, 李少鹏, 张校刚. 负极补锂锂化裕度对电芯性能的影响及机理研究[J]. 储能科学与技术, 2023, 12(9): 2727-2734.
[5]	陈欣, 李云伍, 梁新成, 李法霖, 张志冬. 基于模态分解的Transformer-GRU联合电池健康状态估计[J]. 储能科学与技术, 2023, 12(9): 2927-2936.
[6]	江婉薇, 梁呈景, 钱历, 刘梅城, 朱孟想, 马骏. 锡基三维石墨烯泡沫调控及其锂电池负极性能[J]. 储能科学与技术, 2023, 12(9): 2746-2751.
[7]	高欣, 王若谷, 高文菁, 邓泽军, 梁睿祺, 杨騉. 基于运行数据的储能电站电池组一致性评估方法[J]. 储能科学与技术, 2023, 12(9): 2937-2945.
[8]	周向阳, 胡颖杰, 梁家浩, 周其杰, 文康, 陈松, 杨娟, 唐晶晶. 天然鳞片石墨球化尾料的高性能负极材料制备及储锂特性研究[J]. 储能科学与技术, 2023, 12(9): 2767-2777.
[9]	官亦标, 沈进冉, 刘家亮, 渠展展, 高飞, 刘施阳, 郭翠静, 周淑琴, 付珊珊. 以安全高质量应用为导向的储能锂离子电池综合性能评价标准[J]. 储能科学与技术, 2023, 12(9): 2946-2953.
[10]	唐程波, 锁要红, 何昭坤. 基于正弦函数的液冷板上流体流向对锂离子电池散热性能的影响[J]. 储能科学与技术, 2023, 12(8): 2547-2555.
[11]	左安昊, 方儒卿, 李哲. 锂离子电池单颗粒动力学表征方法综述[J]. 储能科学与技术, 2023, 12(8): 2457-2481.
[12]	陈满, 程志翔, 赵春朋, 彭鹏, 雷旗开, 金凯强, 王青松. 锂离子电池储能集装箱爆炸危害数值模拟[J]. 储能科学与技术, 2023, 12(8): 2594-2605.
[13]	黄永浩, 臧国景, 朱霨亚, 廖友好, 李伟善. LiF添加剂改善含锂陶瓷隔膜与4.35 V LiNi_0.8Co_0.1Mn_0.1O₂ 正极的界面稳定性[J]. 储能科学与技术, 2023, 12(8): 2361-2369.
[14]	杨佳兴, 张恒运, 徐屹东. 基于电化学-热耦合模型的锂离子电池组件产热分析[J]. 储能科学与技术, 2023, 12(8): 2615-2625.
[15]	张吉禄, 董育辰, 宋强, 袁思鸣, 郭孝东. 多晶及单晶高镍三元材料LiNi_0.9Co_0.05Mn_0.05O₂ 的可控制备及其电化学储锂特性[J]. 储能科学与技术, 2023, 12(8): 2382-2389.