基于Informer的电池荷电状态估算及其稀疏优化方法

doi:10.12263/DZXB.20211640

电子学报 ›› 2023, Vol. 51 ›› Issue (1): 50-56.DOI: 10.12263/DZXB.20211640

基于Informer的电池荷电状态估算及其稀疏优化方法

何滢婕, 刘月峰, 边浩东, 郭威, 张小燕

内蒙古科技大学信息工程学院，内蒙古包头 014010

收稿日期:2021-12-09 修回日期:2022-08-15 出版日期:2023-01-25
- 通讯作者:
- 刘月峰
- 作者简介:
- 何滢婕女，1998年出生，山西太原人.现为内蒙古科技大学信息工程学院计算机系硕士研究生.从事电动汽车电池状态估计及深度学习方面的研究.E-mail: 1353517169@qq.com
  刘月峰（通讯作者）男，1977年出生，博士.现为内蒙古科技大学副教授.主要研究方向为机器学习、知识图谱、图像处理、大数据分析与应用.E-mail: liuyuefeng01035@163.com
- 基金资助:
- 内蒙古纪检监察大数据实验室开放课题基金 (IMDBD20200220)

State-of-Charge Estimation of Lithium-ion Battery Based on Informer and Its Sparse Optimization Method

HE Ying-jie, LIU Yue-feng, BIAN Hao-dong, GUO Wei, ZHANG Xiao-yan

School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou, Inner Mongolia 014010, China

Received:2021-12-09 Revised:2022-08-15 Online:2023-01-25 Published:2023-02-23
- Corresponding author:
- LIU Yue-feng
- Supported by:
- Inner Mongolia Discipline Inspection and Supervision Big Data Laboratory Open Project Fund (IMDBD20200220)

摘要/Abstract

摘要：

准确估计电池荷电状态（State Of Charge，SOC）是延长电动汽车电池使用寿命，确保电动汽车行驶安全的重要基础.传统的深度学习估计方法存在并行化计算效率不高、训练时间长的问题.为此，利用基于自注意力机制的Informer模型来估计电池SOC.其降低了传统自注意力机制的时间复杂度、提高了硬件使用率、降低了训练时长，与其他深度学习方法相比估计更准确.然而Informer模型仍然存在体量大及参数冗余的问题，故提出稀疏优化方法.利用基于彩票假设的幅值迭代剪枝方法对Informer进行稀疏化处理，突出主导注意力特征，实现了在降低参数冗余的同时提升模型估计精度.在室温下，提出的稀疏化Informer模型估计电池SOC的平均绝对误差和均方根误差分别达到0.285 8%和0.383 0%，相比于Informer模型在平均绝对误差指标上估计精度提升了25%.并验证了其具备估计不同类型锂电池SOC的泛化能力.与循环神经网络、卷积神经网络这类传统的深度学习模型相比，本模型进行电池SOC估计时训练速度更快，估计准确性和稳定性更高.

关键词: 荷电状态, 锂离子电池, 深度学习, 编解码结构, 自注意力机制

Abstract:

Accurate estimation of the state of charge (SOC) is an important basis for extending the battery life and ensuring the safety of electric vehicles. Traditional deep learning estimation methods suffer from inefficient parallelization and lengthy training time. To this end, the Informer model based on the self-attention mechanism is used to estimate the battery SOC, which reduces the time complexity of the traditional self-attention and training time, increases the hardware usage. It is more accurate than other deep learning methods. However, the Informer model still has the problems of large and redundant parameters, therefore a sparse optimization method is proposed. The iterative magnitude pruning method based on the lottery ticket hypothesis is used to sparse the Informer, highlighting the dominant attention feature and improving model estimation accuracy while reducing parameter redundancy. The proposed sparse-Informer's root mean square error and mean absolute error are 0.383 0% and 0.285 8% at room temperature, respectively. The mean absolute error is decreased by 25% compared to Informer model. Additionally, the generalization ability to estimate the SOC across other lithium-ion battery types is confirmed. When performing battery SOC estimation, this model outperforms other established deep learning models in terms of training speed, estimation accuracy and stability.

Key words: state of charge, lithium-ion battery, deep learning, encoder and decoder, self-attention mechanism

中图分类号:

TM912

何滢婕, 刘月峰, 边浩东, 等. 基于Informer的电池荷电状态估算及其稀疏优化方法[J]. 电子学报, 2023, 51(1): 50-56.

Ying-jie HE, Yue-feng LIU, Hao-dong BIAN, et al. State-of-Charge Estimation of Lithium-ion Battery Based on Informer and Its Sparse Optimization Method[J]. Acta Electronica Sinica, 2023, 51(1): 50-56.

图/表 10

参考文献 16

1	付诗意, 吕桃林, 闵凡奇, 等. 电动汽车用锂离子电池SOC估算方法综述[J]. 储能科学与技术, 2021, 10(3): 1127-1136.
	FU Shi-yi, LV Tao-lin, MIN Fan-qi, et al. Review of estimation methods on SOC of lithium-ion batteries in electric vehicles[J].Energy Storage Science and Technology, 2021, 10(3): 1127-1136. (in Chinese)
2	WANG Yu-jie, TIAN Jia-qiang, SUN Zhen-dong, et al. A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems[J]. Renewable and Sustainable Energy Reviews, 2020, 131: 110015. DOI: 10.1016/j.rser.2020.110015 .
3	张照娓, 郭天滋, 高明裕, 等. 电动汽车锂离子电池荷电状态估算方法研究综述[J]. 电子与信息学报, 2021, 43(7): 1803-1815.
	Zhang Zhao-wei, GUO Tian-zi, GAO Ming-yu, et al. Review of socestimation methods for electric vehicle li-ion batteries[J]. Journal of Electronics & Information Technology, 2021, 43(7): 1803-1815. (in Chinese)
4	HANNAN M A, LIPU M S H, HUSSAIN A, et al. A review of lithium-ion battery state of charge estimation and management system in electric vehicle applications: Challenges and recommendations[J]. Renewable and Sustainable Energy Reviews, 2017, 78: 834-854.
5	Hu Xiao-song, Feng Fei, Liu Kai-long, et al. State estimation for advanced battery management: Key challenges and future trends[J]. Renewable and Sustainable Energy Reviews, 2019, 114: 109334.
6	HOSSAIN LIPU M S, HANNAN M A, KARIM T F, et al. Intelligent algorithms and control strategies for battery management system in electric vehicles: Progress, challenges and future outlook[J]. Journal of Cleaner Production, 2021, 292: 126044. DOI: 10.1016/j.jclepro. 2021. 126044 .
7	CHEMALI E, KOLLMEYER P J, PREINDL M, et al. Long short-term memory networks for accurate state-of-charge estimation of li-ion batteries[J]. IEEE Transactions on Industrial Electronics, 2018, 65(8): 6730-6739.
8	LI Chao-ran, XIAO Fei, FAN Ya-xiang. An approach to state of charge estimation of lithium-ion batteries based on recurrent neural networks with gated recurrent unit[J]. Energies, 2019, 12(9): 1592.
9	HANNAN M A, HOW D N T, LIPU M S H, et al. SOC estimation of li-ion batteries with learning rate-optimized deep fully convolutional network[J]. IEEE Transactions on Power Electronics, 2021, 36(7): 7349-7353.
10	HUANG Zhe-lin, YANG Fang-fang, XU Fan, et al. Convolutional gated recurrent unit-recurrent neural network for state-of-charge estimation of lithium-ion batteries[J]. IEEE Access, 2019, 7: 93139-93149.
11	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: Curran Associates Inc, 2017: 6000-6010.
12	ZHOU Hao-yi, ZHANG Shang-hang, PENG Jie-qi, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//Proceedings of the 35th Association for the Advancement of Artificial Intelligence (AAAI). New York: AAAI, 2021: 11106-11115.
13	GEHRING J, AULI M, GRANGIER D, et al. Convolutional sequence to sequence learning[C]//Proceedings of the 34th International Conference on Machine Learning - Volume 70. Sydney: JMLR.org, 2017: 1243-1252.
14	FRANKLE J, CARBIN M. The lottery ticket hypothesis: Finding sparse, trainable neural networks[C]//International Conference on Learning Representations(ICLR). New Orleans: OpenReview.net, 2019.
15	VidalC, Kollmeyer P, Chemali E, et al. Li-ion battery state of charge estimation using long short-term memory recurrent neural network with transfer learning[C]//2019 IEEE Transportation Electrification Conference and Expo (ITEC). New York: IEEE, 2019: 1-6.
16	HANNAN M A, HOW D N T, LIPU M S H, et al. Deep learning approach towards accurate state of charge estimation for lithium-ion batteries using self-supervised transformer model[J]. Sci Rep, 2021, 11(1): 19541.

指标	工况
指标	UDDS	LA92	US06
RMSE/%	0.375 0	0.394 4	0.582 7
MAE/%	0.297 8	0.315 9	0.454 7

指标	工况
指标	UDDS	LA92	US06
RMSE/%	0.375 0	0.394 4	0.582 7
MAE/%	0.297 8	0.315 9	0.454 7

温度/℃	工况
	LA92			US06			UDDS
	MAE/%	RMSE/%	MAX/%	MAE/%	RMSE/%	MAX/%	MAE/%	RMSE/%	MAX/%
0	0.399 2	0.473 4	1.471 2	0.532 3	0.614 7	1.820 1	0.364 7	0.418 2	0.936 2
10	0.415 7	0.484 9	1.365 5	0.539 1	0.627 1	2.021 7	0.389 0	0.454 2	1.336 7
25	0.435 1	0.506 5	1.337 9	0.528 8	0.636 3	2.063 0	0.391 0	0.446 9	1.040 7
40	0.454 1	0.541 0	1.309 5	0.581 1	0.680 6	2.127 2	0.397 4	0.477 5	1.367 3
-10	0.375 5	0.438 4	1.137 9	0.457 2	0.538 2	1.692 6	0.340 4	0.389 3	0.795 2
-20	0.351 1	0.415 8	1.283 7	0.389 5	0.466 1	1.723 5	0.351 5	0.401 4	0.826 2

温度/℃	工况
	LA92			US06			UDDS
	MAE/%	RMSE/%	MAX/%	MAE/%	RMSE/%	MAX/%	MAE/%	RMSE/%	MAX/%
0	0.399 2	0.473 4	1.471 2	0.532 3	0.614 7	1.820 1	0.364 7	0.418 2	0.936 2
10	0.415 7	0.484 9	1.365 5	0.539 1	0.627 1	2.021 7	0.389 0	0.454 2	1.336 7
25	0.435 1	0.506 5	1.337 9	0.528 8	0.636 3	2.063 0	0.391 0	0.446 9	1.040 7
40	0.454 1	0.541 0	1.309 5	0.581 1	0.680 6	2.127 2	0.397 4	0.477 5	1.367 3
-10	0.375 5	0.438 4	1.137 9	0.457 2	0.538 2	1.692 6	0.340 4	0.389 3	0.795 2
-20	0.351 1	0.415 8	1.283 7	0.389 5	0.466 1	1.723 5	0.351 5	0.401 4	0.826 2

剪枝率	剪枝迭代次数 (最优/全部)	室温/多温度		参数数量(×10⁶)	减去参数量
剪枝率	剪枝迭代次数 (最优/全部)	RMSE/%	MAE/%	参数数量(×10⁶)	减去参数量
不剪枝	—	0.445 6/0.500 5	0.388 3/0.427 3	14.092 633	0
10%	9/12	0.383 0/0.641 5	0.285 8/0.491 7	5.479 422	61.12%
20%	4/6	0.401 8/0.369 3	0.313 7/0.292 3	5.838 433	58.91%
40%	1/3	0.519 3/0.417 2	0.382 8/0.361 3	8.460 227	40%
60%	1/2	0.577 4/0.380 1	0.437 4/0.278 2	5.648 327	59.92%

基于Informer的电池荷电状态估算及其稀疏优化方法

State-of-Charge Estimation of Lithium-ion Battery Based on Informer and Its Sparse Optimization Method

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 16

相关文章 15

编辑推荐

Metrics

训练数据数量	室温/变化温度
训练数据数量	RMSE/%	MAE/%
20%	0.492 0/0.366 8	0.350 8/0.293 2
40%	0.463 8/0.432 6	0.350 3/0.355 7
60%	0.260 2/0.354 3	0.218 2/0.303 3
80%	0.327 3/0.366 2	0.276 0/0.291 1
100%	0.395 5/0.355 0	0.344 2/0.295 3

方法	类型	RMSE/%	MAE/%
方法	类型	室温/多温度	室温/多温度
稀疏化Informer	Transformer	0.401 8/0.380 1	0.313 7/0.278 2
Informer	Transformer	0.519 7/0.500 5	0.391 5/0.427 3
S-Transformer	Transformer	0.905 6/1.191 4	0.445 9/0.650 2
GRU	Recurrent	1.068 6/1.385 6	0.487 7/0.584 7
LSTM	Recurrent	1.138 1/1.449 8	0.534 1/0.730 0
Resnet	Convolutional	1.334 9/1.363 6	0.785 9/0.777 1
FCN	Convolutional	1.555 5/1.780 8	0.964 2/1.081 0
GRU-FCN	Hybird	1.447 7/1.621 5	0.822 8/0.926 9
LSTM-FCN	Hybird	1.777 1/1.924 8	1.195 4/1.155 2

[1]	李钦, 刘伟, 牛朝阳, 宝音图, 惠周勃. 低信噪比下基于分裂EfficientNet网络的雷达信号调制方式识别[J]. 电子学报, 2023, 51(3): 675-686.
[2]	范兵兵, 何庭建, 张聪炫, 陈震, 黎明. 联合遮挡约束与残差补偿的特征金字塔光流计算方法[J]. 电子学报, 2023, 51(3): 648-657.
[3]	张聿远, 闫文君, 张立民. 基于多模态特征融合网络的空时分组码识别算法[J]. 电子学报, 2023, 51(2): 489-498.
[4]	许新征, 李杉. 基于特征膨胀卷积模块的轻量化技术研究[J]. 电子学报, 2023, 51(2): 355-364.
[5]	张永梅, 孙捷. 基于动静态特征双输入神经网络的咳嗽声诊断COVID-19算法[J]. 电子学报, 2023, 51(1): 202-212.
[6]	袁海英, 成君鹏, 曾智勇, 武延瑞. Mobile_BLNet：基于Big-Little Net的轻量级卷积神经网络优化设计[J]. 电子学报, 2023, 51(1): 180-191.
[7]	王神龙, 雍宇, 吴晨睿. 基于伪孪生神经网络的低纹理工业零件6D位姿估计[J]. 电子学报, 2023, 51(1): 192-201.
[8]	李滔, 董秀成, 林宏伟. 基于深监督跨尺度注意力网络的深度图像超分辨率重建[J]. 电子学报, 2023, 51(1): 128-138.
[9]	郭晓轩, 冯其波, 冀振燕, 郑发家, 杨燕燕. 多线激光光条图像缺陷分割模型研究[J]. 电子学报, 2023, 51(1): 172-179.
[10]	贾童瑶, 卓力, 李嘉锋, 张菁. 基于深度学习的单幅图像去雾研究进展[J]. 电子学报, 2023, 51(1): 231-245.
[11]	吴靖, 叶晓晶, 黄峰, 陈丽琼, 王志锋, 刘文犀. 基于深度学习的单帧图像超分辨率重建综述[J]. 电子学报, 2022, 50(9): 2265-2294.
[12]	李雪莹, 王田路, 梁鹏, 王翀. 基于系统模型的用户评论中非功能需求的自动分类[J]. 电子学报, 2022, 50(9): 2079-2089.
[13]	琚长瑞, 秦晓燕, 袁广林, 李豪, 朱虹. 尺度敏感损失与特征融合的快速小目标检测方法[J]. 电子学报, 2022, 50(9): 2119-2126.
[14]	张志昌, 于沛霖, 庞雅丽, 朱林, 曾扬扬. SMGN：用于对话状态跟踪的状态记忆图网络[J]. 电子学报, 2022, 50(8): 1851-1858.
[15]	张亚洲, 俞洋, 朱少林, 陈锐, 戎璐, 梁辉. 一种量子概率启发的对话讽刺识别网络模型[J]. 电子学报, 2022, 50(8): 1885-1893.