电源学报

基于U-Net结构的卷积神经网络锂离子电池SOC估算

PDF下载

靳明旭 ¹^,²^,³ , 陈嘉楠 ¹^,²^,³ , 陈宇昊 ¹^,²^,³ , 徐晗 ¹^,²^,³ , 范鑫源 ⁴^,⁵

电源学报 | 电池与储能 2025,23(7): 304-312

收起

电源学报 |电池与储能 2025 , 23 (7) : 304 -312

基于U-Net结构的卷积神经网络锂离子电池SOC估算

全屏

靳明旭¹^,²^,³, 陈嘉楠¹^,²^,³, 陈宇昊¹^,²^,³, 徐晗¹^,²^,³, 范鑫源⁴^,⁵

作者信息

^1.中国铁道科学研究院集团有限公司机车车辆研究所，北京　100081

^2.动车组和机车牵引与控制国家重点实验室，北京　100081

^3.北京纵横机电科技有限公司，北京　100089

^4.北京交通大学电气工程学院，北京　100044

^5.中车工业研究院有限公司，北京　100070

通讯作者:

范鑫源（1996—　），男，通信作者，博士，讲师。研究方向：电池组仿真建模。E-mail：18117004@bjtu.edu.cn。

作者简介:

靳明旭（1973—　），男，本科，副研究员。研究方向：轨道交通牵引及制动系统。E-mail：jinmingxu@zemt.cn。

陈嘉楠（1992—　），男，硕士，助理研究员。研究方向：轨道车辆牵引电传动系统。E-mail：chenjianan@zemt.cn。

陈宇昊（1993—　），男，硕士，助理研究员。研究方向：轨道车辆牵引电传动系统。E-mail：chenyuhao@zemt.cn。

徐晗（1982—　），男，硕士，高级工程师。研究方向：轨道车辆牵引电传动系统。E-mail：xuhan@zemt.cn。

Convolutional Neural Network Based on U-Net Structure for Lithium-ion Battery SOC Estimation

Mingxu JIN¹^,²^,³, Jianan CHEN¹^,²^,³, Yuhao CHEN¹^,²^,³, Han XU¹^,²^,³, Xinyuan FAN⁴^,⁵

Affiliations

^1.Locomotive & Car Research Institute, China Academy of Railway Sciences Co., Ltd., Beijing 100081, China

^2.State Key Laboratory for Traction and Control System of EMU and Locomotive, Beijing 100081, China

^3.Beijing Zongheng Electro-Mechanical Technology Co., Ltd., Beijing 100089, China

^4.School of Electrical Engineering, Beijing Jiaotong University, Beijing 100044, China

^5.CRRC Industrial Research Institute Co., Ltd., Beijing 100070, China

出版时间: 2025-11-30 doi: 10.13234/j.issn.2095-2805.2025.7.304

文章导航

摘要

收起

荷电状态SOC（state-of-charge）是电动汽车安全运行和能量管理的关键。传统基于卡尔曼滤波与循环神经网络的SOC估算算法需要一定量的数据来保证估算结果的收敛性，且无法准确估算起始点附近的SOC值。基于此，提出1种使用U-Net结构的卷积神经网络SOC估算方法，该方法可以处理可变长度的输入数据，输出等长的SOC估算结果，同时也能在起始点处准确估算SOC。此外还提出了1种全变分损失函数，在不增加模型复杂度的情况下，仅通过优化损失函数来提高估算的稳定性，并显著降低了最大误差。采用5种恒温条件下的动态驾驶循环数据对该模型进行训练，在恒温和变温条件下的SOC估算结果均具有较高的精度。在恒温条件下，SOC估算的平均绝对误差MAE（mean absolute error）在1.1%以内，均方根误差RMSE（root mean square error）在1.4%以内。在变温条件下，SOC估算的MAE在1.5%以内，RMSE在1.8%以内。

关键词

荷电状态 / U-Net结构 / 卷积神经网络 / 荷电状态估算

Abstract

收起

The state-of-charge (SOC) is key to the safe operation and energy management of electric vehicles. The traditional SOC estimation algorithm based on Kalman filter and recurrent neural network requires a period of data to ensure the convergence of estimation results and cannot accurately estimate the SOC value near the starting point. A convolutional neural network SOC estimation method using U-Net structure is proposed, which can deal with variable-length input data and output the equal-length SOC estimation results. Meanwhile, it can also accurately estimate SOC at the starting point. In addition, a full-variance loss function is put forward, which can improve the estimation stability only by optimizing the loss function without increasing the model complexity and significantly reduce the maximum error. This model is trained with dynamic driving cycle data under five constant-temperature conditions, and the SOC estimation results are highly accurate under both the constant-temperature and variable-temperature conditions. Under the constant-temperature condition, the mean absolute error (MAE) and root mean square error (RMSE) of SOC estimation were within 1.1% and 1.4%, respectively. Under the variable-temperature condition, the MAE and RMSE were within 1.5% and 1.8%, respectively.

Key words

State-of-charge (SOC) / U-Net structure / convolutional neural network / SOC estimation

引用本文

靳明旭, 陈嘉楠, 陈宇昊, 徐晗, 范鑫源. 基于U-Net结构的卷积神经网络锂离子电池SOC估算. 电源学报, 2025 , 23 (7) : 304 -312 . DOI: 10.13234/j.issn.2095-2805.2025.7.304

Mingxu JIN, Jianan CHEN, Yuhao CHEN, Han XU, Xinyuan FAN. Convolutional Neural Network Based on U-Net Structure for Lithium-ion Battery SOC Estimation[J]. Journal of Power Supply, 2025 , 23 (7) : 304 -312 . DOI: 10.13234/j.issn.2095-2805.2025.7.304

基金

收起

中国铁道科学研究院集团有限公司科研项目(2020YJ202)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

林加顺, 周娟, 吴乃豪, 等. 一种高精度磷酸铁锂电池OCV-SOC曲线获取方法[J/OL]. 电源学报, 2023: 1-13. (2023-05-17) [2023-08-01]. https://link.cnki.net/urlid/12.1420.TM.20230517.1045.002. Lin Jiashun, Zhou Juan, Wu Naihao, et al. A high precision OCV-SOC curve acquisition method for LiFeO₄ battery[J/OL]. Journal of Power Supply, 2023: 1-13. (2023-05-17) [2023-08-01]. https://link.cnki.net/urlid/12.1420.TM.20230517.1045.002. (in Chinese)

[2]

傅军栋, 陈浩杰, 孙翔, 等. 基于EKF-Markov的UPS荷电状态预测与健康管理系统[J]. 电源学报, 2024, 22(6): 225-233. Fu Jundong, Chen Haojie, Sun Xiang, et al. SOC prediction and health management system of UPS based on EKF-Markov[J]. Journal of Power Supply, 2024, 22(6): 225-233. (in Chinese)

[3]

许傲然, 戴菁, 谷彩莲, 等. 基于WOA优化扩展卡尔曼算法的锂离子电池SOC估算研究[J]. 电源学报, 2025, 23(2): 232-239. Xu Aoran, Dai Jing, Gu Cailian, et al. Research on SOC estimation of lithium-ion battery based on WOA optimized extended Kalman algorithm[J]. Journal of Power Supply, 2025, 23(2): 232-239. (in Chinese)

[4]

Kim M J, Chae S H, Moon Y K. Adaptive battery state-of-charge estimation method for electric vehicle battery management system[C]//2020 International SoC Design Conference (ISOCC). Yeosu, Korea, 2021: 288-289.

[5]

王雪, 游国栋, 房成信, 等. 基于IMOCS-BP神经网络的锂离子电池SOH估计[J]. 电源学报, 2024, 22(1): 94-100. Wang Xue, You Guodong, Fang Chengxin, et al. SOH estimation of lithium-ion battery based on IMOCS-BP neural network[J]. Journal of Power Supply, 2024, 22(1): 94-100. (in Chinese)

[6]

Li Chuanjiang, Chen Zewang, Cui Jiang, et al. The lithium-ion battery state-of-charge estimation using random forest regression[C]//2014 Prognostics and System Health Management Conference (PHM-2014 Hunan). Zhangjiajie, China, 2014: 336-339.

[7]

Song Yuchen, Liu Datong, Liao Haitao, et al. A hybrid statistical data-driven method for on-line joint state estimation of lithium-ion batteries[J]. Applied Energy, 2020, 261: 114408.

[8]

How D N T, Hannan M A, Lipu M S H, et al. State-of-charge estimation of Li-ion battery in electric vehicles: A deep neural network approach[J]. IEEE Transactions on Industry Applications, 2020, 56(5): 5565-5574.

[9]

白俊琦, 贤燕华. 基于BP-多时间尺度UKF优化算法的SOC估计[J]. 电源学报, 2025, 23(3): 363-372. Bai Junqi, Xian Yanhua. State-of-charge estimation based on BP-multi-timescale UKF optimization algorithm[J]. Journal of Power Supply, 2025, 23(3): 363-372. (in Chinese)

[10]

Dong Chao, Wang Guanlan. Estimation of power battery SOC based on improved BP neural network[C]//2014 IEEE International Conference on Mechatronics and Automation. Tianjin, China, 2014: 2022-2027.

[11]

Xu Long, Wang Junping, Chen Quanshi. Kalman filtering state of charge estimation for battery management system based on a stochastic fuzzy neural network battery model[J]. Energy Conversion and Management, 2012, 53(1): 33-39.

[12]

Zhang Chuanwei, Chen Shangrui, Gao Huaibin, et al. State of charge estimation of power battery using improved back propagation neural network[J]. Batteries, 2018, 4(4): 69.

[13]

Charkhgard M, Farrokhi M. State-of-charge estimation for lithium-ion batteries using neural networks and EKF[J]. IEEE Transactions on Industrial Electronics, 2010, 57(12): 4178-4187.

[14]

Chen Xiaopeng, Shen Weixiang, Dai Mingxiang, et al. Robust adaptive sliding-mode observer using RBF neural network for lithium-ion battery state of charge estimation in electric vehicles[J]. IEEE Transactions on Vehicular Technology, 2016, 65(4): 1936-1947.

[15]

Wang Zhihuo, Fotouhi A, Auger D J. State of charge estimation in lithium-sulfur cells using LSTM recurrent neural networks[C]//2020 European Control Conference (ECC). St. Petersburg, Russia, 2020: 1079-1085.

[16]

Yang Fangfang, Song Xiangbao, Xu Fan, et al. State-of-charge estimation of lithium-ion batteries via long short-term memory network[J]. IEEE Access, 2019, 7: 53792-53799.

[17]

Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Cham: Springer, 2015: 234-241.

[18]

Sainath T N, Vinyals O, Senior A, et al. Convolutional, long short-term memory, fully connected deep neural networks[C]//2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). South Brisbane, Queensland, Australia, 2015: 4580-4584.

[19]

Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence. 2016: 640-651.

[20]

Kollmeyer P. Panasonic 18650PF Li-ion battery data[EB/OL]. (2018-06-21) [2023-08-01]. https://data.mendeley.com/datasets/wykht8y7tg/1.

2025年第23卷第7期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.13234/j.issn.2095-2805.2025.7.304

接收时间：2023-08-18
首发时间：2026-10-26
出版时间：2025-11-30

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2023-08-18
修回日期：2023-11-06
录用日期：2023-12-04

基金

Scientific Research Project of China Academy of Railway Sciences Co., Ltd.(2020YJ202)

中国铁道科学研究院集团有限公司科研项目(2020YJ202)

作者信息

^1.中国铁道科学研究院集团有限公司机车车辆研究所，北京　100081

^2.动车组和机车牵引与控制国家重点实验室，北京　100081

^3.北京纵横机电科技有限公司，北京　100089

^4.北京交通大学电气工程学院，北京　100044

^5.中车工业研究院有限公司，北京　100070

通讯作者:

范鑫源（1996—　），男，通信作者，博士，讲师。研究方向：电池组仿真建模。E-mail：18117004@bjtu.edu.cn。

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/dyxb/CN/10.13234/j.issn.2095-2805.2025.7.304

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT