基于Soft-DTW算法与多源特征融合的实车动力电池SOH估算

doi:10.19799/j.cnki.2095-4239.2025.0046

摘要/Abstract

摘要：

电动汽车动力电池的健康状态是保障其高效运行与延长使用寿命的关键因素。然而，由于电动汽车在实际使用中充电模式复杂多变，单次充放电循环的采样间隔大且伴随部分特征信息缺失，导致动力电池SOH的精准评估面临极大挑战。为此，本研究针对实车运行数据提出了一种基于软-动态时间规整(soft-dynamic time warping, Soft-DTW)算法与多源特征融合的SOH估算模型。首先，通过Soft-DTW算法对每周充电片段的容量增量(incremental capacity, IC)曲线进行动态参数融合，随后提取每周总体IC融合特征，并构建融合IC曲线特征与统计学特征的多源特征集。基于此，提出基于双向门控循环单元-极端梯度提升(bidirectional gated recurrent unit-extreme gradient boosting, BiGRU-XGBoost)实车SOH估算模型，该模型在由20辆电动汽车组成的实车数据集上进行了K折交叉验证，结果表明提出的SOH估算方法均方根误差在1.21%以内，均绝对误差低于0.9%。同时与GRU-XGBoost、LSTM进行多模型对比实验，均方根误差降低了36.1%和47.6%，验证了BiGRU-XGBoost模型更强的稳健性和泛化能力。

关键词: 电动汽车, 健康状态估计, 容量增量

Abstract:

The State of Health (SOH) of electric vehicle (EV) power batteries is a critical factor in ensuring efficient operation and extending service life. However, accurately assessing SOH is challenging owing to the complex and variable charging patterns in real-world EV usage, large sampling intervals in individual charge-discharge cycles, and missing feature data. To address these challenges, this study proposes a multisource feature fusion method for SOH estimation method using real-world vehicle operation data. The proposed method utilizes the soft dynamic time warping (Soft-DTW) algorithm to dynamically fuse parameters from weekly charging segment incremental capacity (IC) curves, generating an overall weekly IC fusion feature. By integrating these fused IC curve features with statistical features, a multisource feature set is constructed. Furthermore, a real-world SOH estimation model based on the Bidirectional Gated Recurrent Unit-eXtreme Gradient Boosting (BiGRU-XGBoost) is proposed. The model was tested using a real-world dataset comprising 20 EVs. K-fold cross-validation results demonstrates that the proposed SOH estimation method achieves a root mean square error (RMSE) within 1.21% and a mean absolute error (MAE) below 0.9%. Comparative experiments with GRU-XGBoost and long short-term memory (LSTM) models further validate the superiority of the BiGRU-XGBoost model, showing 36.1% and 47.6% reductions in RMSE. These findings highlight the enhanced robustness and generalization capabilities of the BiGRU-XGBoost model.

Key words: electric vehicle, state of health estimation, incremental capacity

中图分类号:

TM 911

丁萍, 李涛涛, 郑林锋, 吴伟雄. 基于Soft-DTW算法与多源特征融合的实车动力电池SOH估算[J]. 储能科学与技术, 2025, 14(5): 2081-2097.

Ping DING, Taotao LI, Linfeng ZHENG, Weixiong WU. SOH estimation of real-world power batteries based on Soft-DTW algorithm and multisource reature fusion[J]. Energy Storage Science and Technology, 2025, 14(5): 2081-2097.

图/表 19

表1

图1

图2

图3

表2

图4

图5

图6

图7

图8

图9

表3

参数与特征相关性"

参数值	峰值高度与SOH的相关系数	峰值电压与SOH的相关系数
$γ$ =0.25	0.613	-0.633
$γ$ =0.5	0.646	-0.596
$γ$ =0.75	0.724	-0.679
$γ$ =1	0.767	-0.742
$γ$ =1.25	0.654	-0.590
$γ$ =1.5	0.658	-0.587
$γ$ =1.75	0.617	-0.554
$γ$ =2	0.579	-0.532

表3

图10

图11

图12

表4

图13

图14

图15

参考文献 30

1	PAN S, FULTON L M, ROY A, et al. Shared use of electric autonomous vehicles: Air quality and health impacts of future mobility in the United States[J]. Renewable and Sustainable Energy Reviews, 2021, 149: 111380. DOI: 10.1016/j.rser.2021. 111380.
2	DAS K, KUMAR R, KRISHNA A. Analyzing electric vehicle battery health performance using supervised machine learning[J]. Renewable and Sustainable Energy Reviews, 2024, 189: 113967. DOI: 10.1016/j.rser.2023.113967.
3	陈媛, 章思源, 蔡宇晶, 等. 融合多项式特征扩展与CNN-Transformer模型的锂电池SOH估计[J]. 储能科学与技术, 2024, 13(9): 2995-3005. DOI: 10.19799/j.cnki.2095-4239.2024.0465.
	CHEN Y, ZHANG S Y, CAI Y J, et al. State-of-health estimation of lithium batteries based on polynomial feature extension of the CNN-transformer model[J]. Energy Storage Science and Technology, 2024, 13(9): 2995-3005. DOI: 10.19799/j.cnki.2095-4239.2024. 0465.
4	YANG B, QIAN Y C, LI Q, et al. Critical summary and perspectives on state-of-health of lithium-ion battery[J]. Renewable and Sustainable Energy Reviews, 2024, 190: 114077. DOI: 10.1016/j.rser.2023. 114077.
5	DEMIRCI O, TASKIN S, SCHALTZ E, et al. Review of battery state estimation methods for electric vehicles-Part II: SOH estimation[J]. Journal of Energy Storage, 2024, 96: 112703. DOI: 10.1016/j.est.2024.112703.
6	BASIA A, SIMEU-ABAZI Z, GASCARD E, et al. Review on State of Health estimation methodologies for lithium-ion batteries in the context of circular economy[J]. CIRP Journal of Manufacturing Science and Technology, 2021, 32: 517-528. DOI: 10.1016/j.cirpj. 2021.02.004.
7	MARCICKI J, CANOVA M, CONLISK A T, et al. Design and parametrization analysis of a reduced-order electrochemical model of graphite/LiFePO₄ cells for SOC/SOH estimation[J]. Journal of Power Sources, 2013, 237: 310-324. DOI: 10.1016/j.jpowsour.2012.12.120.
8	CACCIATO M, NOBILE G, SCARCELLA G, et al. Real-time model-based estimation of SOC and SOH for energy storage systems[J]. IEEE Transactions on Power Electronics, 2017, 32(1): 794-803. DOI: 10.1109/TPEL.2016.2535321.
9	GUO R H, SHEN W X. A review of equivalent circuit model based online state of power estimation for lithium-ion batteries in electric vehicles[J]. Vehicles, 2022, 4(1): 1-29. DOI: 10.3390/vehicles 4010001.
10	刘定宏, 董文楷, 李召阳, 等. 基于RUN-GRU-attention模型的实车动力电池健康状态估计方法[J]. 储能科学与技术, 2024, 13(9): 3042-3058. DOI: 10.19799/j.cnki.2095-4239.2024.0576.
	LIU D H, DONG W K, LI Z Y, et al. Estimation of real-vehicle battery state of health using the RUN-GRU-attention model[J]. Energy Storage Science and Technology, 2024, 13(9): 3042-3058. DOI: 10.19799/j.cnki.2095-4239.2024.0576.
11	BIAN X L, WEI Z B, HE J T, et al. A novel model-based voltage construction method for robust state-of-health estimation of lithium-ion batteries[J]. IEEE Transactions on Industrial Electronics, 2021, 68(12): 12173-12184. DOI: 10.1109/TIE.2020.3044779.
12	WU J, MENG J H, LIN M Q, et al. Lithium-ion battery state of health estimation using a hybrid model with electrochemical impedance spectroscopy[J]. Reliability Engineering & System Safety, 2024, 252: 110450. DOI: 10.1016/j.ress.2024.110450.
13	WENG C H, CUI Y J, SUN J, et al. On-board state of health monitoring of lithium-ion batteries using incremental capacity analysis with support vector regression[J]. Journal of Power Sources, 2013, 235: 36-44. DOI: 10.1016/j.jpowsour.2013.02.012.
14	OSPINA AGUDELO B, ZAMBONI W, MONMASSON E. Application domain extension of incremental capacity-based battery SoH indicators[J]. Energy, 2021, 234: 121224. DOI: 10.1016/j.energy. 2021.121224.
15	LI X Y, YUAN C G, WANG Z P. State of health estimation for Li-ion battery via partial incremental capacity analysis based on support vector regression[J]. Energy, 2020, 203: 117852. DOI: 10.1016/j.energy.2020.117852.
16	WANG G F, CUI N X, LI C L, et al. A state-of-health estimation method based on incremental capacity analysis for Li-ion battery considering charging/discharging rate[J]. Journal of Energy Storage, 2023, 73: 109010. DOI: 10.1016/j.est.2023.109010.
17	XU Z F, CHEN Z W, YANG L, et al. State of health estimation for lithium-ion batteries based on incremental capacity analysis and Transformer modeling[J]. Applied Soft Computing, 2024, 165: 112072. DOI: 10.1016/j.asoc.2024.112072.
18	WENG C H, FENG X N, SUN J, et al. State-of-health monitoring of lithium-ion battery modules and packs via incremental capacity peak tracking[J]. Applied Energy, 2016, 180: 360-368. DOI: 10.1016/j.apenergy.2016.07.126.
19	CHE Y H, DENG Z W, TANG X L, et al. Lifetime and aging degradation prognostics for lithium-ion battery packs based on a cell to pack method[J]. Chinese Journal of Mechanical Engineering, 2022, 35(1): 4. DOI: 10.1186/s10033-021-00668-y.
20	SHE C Q, ZHANG L, WANG Z P, et al. Battery state-of-health estimation based on incremental capacity analysis method: Synthesizing from cell-level test to real-world application[J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2023, 11(1): 214-223. DOI: 10.1109/JESTPE.2021.3112754.
21	BILFINGER P, ROSNER P, SCHREIBER M, et al. Battery pack diagnostics for electric vehicles: Transfer of differential voltage and incremental capacity analysis from cell to vehicle level[J]. eTransportation, 2024, 22: 100356. DOI: 10.1016/j.etran.2024. 100356.
22	CHANG C, WU Y T, JIANG J C, et al. Prognostics of the state of health for lithium-ion battery packs in energy storage applications[J]. Energy, 2022, 239: 122189. DOI: 10.1016/j.energy.2021.122189.
23	SHE C Q, WANG Z P, SUN F C, et al. Battery aging assessment for real-world electric buses based on incremental capacity analysis and radial basis function neural network[J]. IEEE Transactions on Industrial Informatics, 2020, 16(5): 3345-3354. DOI: 10.1109/TII. 2019.2951843.
24	HONG J C, LI K R, LIANG F W, et al. A novel state of health prediction method for battery system in real-world vehicles based on gated recurrent unit neural networks[J]. Energy, 2024, 289: 129918. DOI: 10.1016/j.energy.2023.129918.
25	LIN C P, XU J, JIANG D L, et al. Multi-model ensemble learning for battery state-of-health estimation: Recent advances and perspectives[J]. Journal of Energy Chemistry, 2025, 100: 739-759. DOI: 10.1016/j.jechem.2024.09.021.
26	DENG Z W, XU L, LIU H A, et al. Prognostics of battery capacity based on charging data and data-driven methods for on-road vehicles[J]. Applied Energy, 2023, 339: 120954. DOI: 10.1016/j.apenergy.2023.120954.
27	郭煜, 王亦伟, 钟隽, 等. 基于增量容量曲线的锂离子电池微内短路故障诊断方法[J]. 储能科学与技术, 2023, 12(8): 2536-2546. DOI: 10.19799/j.cnki.2095-4239.2023.0186.
	GUO Y, WANG Y W, ZHONG J, et al. Fault diagnosis method for microinternal short circuits in lithium-ion batteries based on incremental capacity curve[J]. Energy Storage Science and Technology, 2023, 12(8): 2536-2546. DOI: 10.19799/j.cnki.2095-4239.2023.0186.
28	DAI Y M, YU W J, LENG M M. A hybrid ensemble optimized BiGRU method for short-term photovoltaic generation forecasting[J]. Energy, 2024, 299: 131458. DOI: 10.1016/j.energy.2024.131458.
29	陆继忠, 彭思敏, 李晓宇. 基于多特征量分析和LSTM-XGBoost模型的锂离子电池SOH估计方法[J]. 储能科学与技术, 2024, 13(9): 2972-2982. DOI: 10.19799/j.cnki.2095-4239.2024.0289.
	LU J Z, PENG S M, LI X Y. State-of-health estimation of lithium-ion batteries based on multifeature analysis and LSTM-XGBoost model[J]. Energy Storage Science and Technology, 2024, 13(9): 2972-2982. DOI: 10.19799/j.cnki.2095-4239.2024.0289.
30	XIONG G J, ZHANG J, FU X F, et al. Seasonal short-term photovoltaic power prediction based on GSK-BiGRU-XGboost considering correlation of meteorological factors[J]. Journal of Big Data, 2024, 11(1): 164. DOI: 10.1186/s40537-024-01037-x.

记录时间	电池组SOC	电池组电压/V	电池组电流/A	单体电池最高电压/V	单体电池最低电压/V	单体电池最高温度/℃	单体电池最高温度/℃	能量/kW	可用容量/Ah
2019-07-26 20:02:35	27.2	328.2	-52.2	3.656	3.64	41	38	12.4	37.28
2019-07-26 20:02:43	27.6	328.5	-52.2	3.663	3.645	41	38	12.44	37.39
2019-07-26 20:02:51	27.6	328.6	-52.2	3.665	3.647	41	38	12.47	37.51
2019-07-26 20:02:59	27.6	328.6	-52.2	3.666	3.649	41	38	12.52	37.64
2019-07-26 20:03:07	27.6	328.8	-52.2	3.666	3.649	41	38	12.56	37.76
…	…	…	…	…	…	…	…	…	…
2021-11-15 16:46:00	94.8	371.9	-41.7	4.146	4.112	34	33	41.14	123.69
2021-11-15 16:46:08	94.8	371.9	-41.9	4.147	4.114	34	33	41.17	123.78
2021-11-15 16:46:16	94.8	372.1	-41.8	4.147	4.114	34	33	41.2	123.89
2021-11-15 16:46:24	94.8	372.1	-41.8	4.148	4.114	34	33	41.24	124

筛选条件	EV1	EV2	EV3	EV4	EV5	EV6	EV7	EV8	EV9	EV10
原始	1613	1637	1535	1607	1603	1564	1577	1564	1503	1376
S1	1560	1575	1491	1527	1518	1495	1508	1495	1447	1314
[S1，S2]	1485	1491	1422	1449	1437	1403	1429	1413	1378	1249
保留度/%	92.06	91.08	92.63	90.17	89.64	89.70	90.62	94.01	91.68	90.77
筛选条件	EV11	EV12	EV13	EV14	EV15	EV16	EV17	EV18	EV19	EV20
原始	1544	1570	1526	1537	1309	1594	1453	1632	1530	1554
S1	1483	1504	1433	1467	1253	1550	1393	1540	1458	1489
[S1，S2]	1403	1420	1342	1383	1195	1481	1334	1460	1372	1425
保留度/%	90.86	90.45	87.94	89.98	91.29	92.91	81.74	89.46	89.67	91.70

不同K值	数据集	训练集		测试集
不同K值	数据集	MAE/%	RMSE/%	MAE/%	RMSE/%
K=2	max	0.63	0.75	1.37	1.78
	min	0.61	0.78	1.16	1.54
	average	0.62	0.77	1.27	1.67

K=4	max	0.47	0.6	1.18	1.56
	min	0.46	0.59	1.00	1.27
	average	0.47	0.6	1.12	1.50

K=6	max	0.45	0.57	1.09	1.37
	min	0.43	0.58	0.88	1.26
	average	0.44	0.57	1.01	1.35

K=8	max	0.44	0.56	1.12	1.45
	min	0.41	0.55	0.85	1.08
	average	0.42	0.56	0.95	1.21

[1]	吴金涛, 周永军, 杨旭, 邓程欣, 韩一铭. 细水雾对电动汽车火灾抑制效果的数值模拟[J]. 储能科学与技术, 2025, 14(5): 2098-2105.
[2]	彭磊, 倪照鹏, 于越, 孙福鹏, 夏修龙, 张鹏, 孙思博. 过充导致三元锂电池电动汽车火灾的试验研究[J]. 储能科学与技术, 2025, 14(4): 1484-1495.
[3]	蔡志端, 张吴哲, 吴成傲, 童嘉阳. 强干扰下基于VMD三次分解的锂电池健康状态估计方法[J]. 储能科学与技术, 2025, 14(4): 1631-1644.
[4]	张朝龙, 陈阳, 刘梦玲, 张俣峰, 华国庆, 阴盼昐. 一种基于ICA-T特征和CNN-LA-BiLSTM的锂离子电池健康状态估计方法[J]. 储能科学与技术, 2025, 14(3): 1258-1269.
[5]	王海瑞, 徐长宇, 朱贵富, 侯晓建. 一种并行多尺度特征融合模型开展的基于弛豫电压的锂电池SOH估计研究[J]. 储能科学与技术, 2025, 14(2): 799-811.
[6]	陈星光, 沈逸凡, 邵裕新, 郑岳久, 孙涛, 来鑫, 沈凯, 韩雪冰. 面向实车应用的磷酸铁锂电池容量辨识及特异性优化方法研究[J]. 储能科学与技术, 2024, 13(9): 2963-2971.
[7]	刘定宏, 董文楷, 李召阳, 张红烛, 齐昕. 基于RUN-GRU-attention模型的实车动力电池健康状态估计方法[J]. 储能科学与技术, 2024, 13(9): 3042-3058.
[8]	刘松燕, 王卫良, 彭世亮, 吕俊复. 兼顾高/低温环境性能的动力电池热管理系统设计[J]. 储能科学与技术, 2024, 13(7): 2181-2191.
[9]	申小雨, 尹丛勃. 基于卷积Fastformer的锂离子电池健康状态估计[J]. 储能科学与技术, 2024, 13(3): 990-999.
[10]	唐兆祥, 许万涛, 邓昊, 卢文杰. 基于机会约束规划的含电动汽车市域铁路牵引供电系统优化运行[J]. 储能科学与技术, 2024, 13(2): 526-535.
[11]	朱杰. 新能源低碳背景下电动汽车电热相变储能系统的储热性能分析[J]. 储能科学与技术, 2024, 13(12): 4406-4408.
[12]	牛萍健, 郝维健, 苏智阳, 师盛坤, 柳邵辉. GB/T 31467—2023《电动汽车用锂离子动力电池包和系统电性能试验方法》标准解读与分析[J]. 储能科学与技术, 2024, 13(10): 3672-3679.
[13]	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.
[14]	洪睿洁, 顾丹珍, 莫阮清, 蔡思楠, 张超林. 基于用户偏好的电动汽车储能V2G策略优化[J]. 储能科学与技术, 2023, 12(8): 2659-2667.
[15]	张响, 段俊东, 康博阳. 考虑电动汽车灵活储能的微电网双重激励优化调度[J]. 储能科学与技术, 2023, 12(8): 2556-2564.