基于Douglas-Peucker融合闵式距离的锂电池健康因子提取及SOH预测

doi:10.19799/j.cnki.2095-4239.2022.0188

摘要/Abstract

摘要：

针对锂离子电池的健康因子提取困难而导致电池健康状况(state of health，SOH)预测精度低的问题，提出一种基于Douglas-Peucker融合闵式距离的锂电池健康因子特征提取算法，并利用该算法对恒流恒压充电恒功率放电策略下的电池数据进行特征提取，进而实现对锂电池的SOH预测。首先对测量的实验数据建立特征工程，利用闵式距离建立评价指标，实现基于Douglas-Peucker算法的电池健康因子提取，进而得到34维健康因子。然后，针对所提取的健康因子，利用差分变异头脑风暴(difference-mutation brainstorm optimization，DBSO)算法进行寻优，剔除不相关和冗余的特征，避免模型过拟合，提高模型性能。最后，利用支持向量机(support vector machines，SVM)及其优化模型对所提取的健康因子进行电池SOH预测。实验结果表明，所建立的特征工程提取的健康因子在SVM各模型中拟合优度均超过0.96，其中DBSO-SVM模型的预测精度最高，预测效果最好，平均绝对值误差(mean square error，MSE)值低于3。结合不同充放电策略，将所提出的特征提取算法在NASA数据上验证。结果表明，在SVM模型上，电池B0005、B0006、B0007的拟合优度达到0.99，均方根误差(root mean square error，RMSE)值均低于6%。对比多种优化算法，DBSO-SVM模型的性能最好。

关键词: Douglas-Peucker算法, SOH, 闵式距离, DBSO算法, SVM模型

Abstract:

To solve the problem of the low prediction accuracy of state of health (SOH) caused by the difficulty in extracting health factors of lithium-ion batteries, a feature-extraction algorithm of lithium-ion battery health factors based on the Douglas-Peucker fusion Min distance is proposed. The algorithm is used to extract the battery data feature under constant current and voltage charging, as well as constant power discharge strategies, to realize the SOH prediction of lithium batteries. First, the characteristic engineering is established for the measured experimental data, and the evaluation index is established using the Min distance. The battery health factor is extracted based on the Douglas-Peucker algorithm, and then the 34-dimensional health factor is obtained. For the extracted health factors, DBSO (difference-mutation brainstorm optimization) algorithm is used for optimization to eliminate irrelevant and redundant features, avoid overfitting of the model and improve model performance. Finally, the SVM (support vector machines) and its optimization model were used to predict the SOH of the battery. The experimental results show that the goodness of fit of the health factors extracted using feature engineering is greater than 0.96 in each SVM model. The DBSO-SVM model has the highest prediction accuracy and the best prediction effect, and the MSE (mean square error) value is less than 3. The proposed feature-extraction algorithm is verified on NASA data using different charging and discharging strategies; the results show that on the SVM model, the goodness of fit of B0005, B0006, and B0007 reaches 0.99, and the RMSE (root mean square error) values are lower than 6%. Compared with the various optimization algorithms, the DBSO-SVM model demonstrated the best performance.

Key words: douglas-peucker algorithm, SOH, minkowski distance, DBSO algorithm, SVM model

中图分类号:

TM 910.7

陈万利, 张梅, 冯涛. 基于Douglas-Peucker融合闵式距离的锂电池健康因子提取及SOH预测[J]. 储能科学与技术, 2022, 11(10): 3306-3315.

Wanli CHEN, Mei ZHANG, Tao FENG. The health factor extraction and SOH prediction of lithium battery based on douglas-peucker fusion minkowski distance[J]. Energy Storage Science and Technology, 2022, 11(10): 3306-3315.

图/表 15

图1

图2

表1

图3

图4

图5

图6

图7

图8

图9

图10

图11

表2

图12

表3

参考文献 23

1	史永胜, 李锦, 任嘉睿, 等. 基于WOA-XGBoost的锂离子电池剩余使用寿命预测[J/OL]. 储能科学与技术.(2022-04-02)[2022-05-06]. https://esst.cip.com.cn/CN/10.19799/j.cnki.2095-4239.2022.0126.
	SHI Y S, LI J, REN J R, et al. Prediction of residual service life of lithium-ion battery based on WOA-XGBoost[J]. Energy Storage Science and Technology. (2022-04-02) [2022-05-06]. https://esst.cip.com.cn/CN/10.19799/j.cnki.2095-4239.2022.0126.
2	GE M F, LIU Y B, JIANG X X, et al. A review on state of health estimations and remaining useful life prognostics of lithium-ion batteries[J]. Measurement, 2021, 174: doi: 10.1016/j.measurement. 2021.109057.
3	梁新成, 张勉, 黄国钧. 基于BMS的锂离子电池建模方法综述[J]. 储能科学与技术, 2020, 9(6): 1933-1939.
	LIANG X C, ZHANG M, HUANG G J. Review on lithium-ion battery modeling methods based on BMS[J]. Energy Storage Science and Technology, 2020, 9(6): 1933-1939.
4	易灵芝, 张宗光, 范朝冬, 等. 基于EEMD-GSGRU的锂电池寿命预测[J]. 储能科学与技术, 2020, 9(5): 1566-1573.
	YI L Z, ZHANG Z G, FAN C D, et al. Life prediction of lithium battery based on EEMD-GSGRU[J]. Energy Storage Science and Technology, 2020, 9(5): 1566-1573.
5	王萍, 范凌峰, 程泽. 基于健康特征参数的锂离子电池SOH和RUL联合估计方法[J]. 中国电机工程学报, 2022, 42(4): 1523-1534.
	WANG P, FAN L F, CHENG Z. A joint state of health and remaining useful life estimation approach for lithium-ion batteries based on health factor parameter[J]. Proceedings of the CSEE, 2022, 42(4): 1523-1534.
6	FENG X N, WENG C H, HE X M, et al. Online state-of-health estimation for Li-ion battery using partial charging segment based on support vector machine[J]. IEEE Transactions on Vehicular Technology, 2019, 68(9): 8583-8592.
7	HU X S, CHE Y H, LIN X K, et al. Health prognosis for electric vehicle battery packs: A data-driven approach[J]. IEEE/ASME Transactions on Mechatronics, 2020, 25(6): 2622-2632.
8	简献忠, 韦进, 王如志. 基于RPMDE-MKSVM的锂离子电池剩余使用寿命预测[J]. 控制工程, 2021, 28(4): 665-671.
	JIAN X Z, WEI J, WANG R Z. Remaining useful life prediction of lithium-ion battery based on RPMDE-MKSVM[J]. Control Engineering of China, 2021, 28(4): 665-671.
9	王瀛洲, 倪裕隆, 郑宇清, 等. 基于ALO-SVR的锂离子电池剩余使用寿命预测[J]. 中国电机工程学报, 2021, 41(4): 1445-1457, 1550.
	WANG Y Z, NI Y L, ZHENG Y Q, et al. Remaining useful life prediction of lithium-ion batteries based on support vector regression optimized and ant lion optimizations[J]. Proceedings of the CSEE, 2021, 41(4): 1445-1457, 1550.
10	陈峥, 顾青峰, 沈世全, 等. 基于健康特征提取和PSO-RBF神经网络的锂离子电池健康状态预测[J]. 昆明理工大学学报(自然科学版), 2020, 45(6): 92-103.
	CHEN Z, GU Q F, SHEN S Q, et al. State of health prediction for lithium-ion batteries based on health feature extraction and PSO-RBF neural network[J]. Journal of Kunming University of Science and Technology (Natural Science), 2020, 45(6): 92-103.
11	李练兵, 李思佳, 李洁, 等. 基于差分电压和Elman神经网络的锂离子电池RUL预测方法[J]. 储能科学与技术, 2021, 10(6): 2373-2384.
	LI L B, LI S J, LI J, et al. RUL prediction of lithium-ion battery based on differential voltage and Elman neural network[J]. Energy Storage Science and Technology, 2021, 10(6): 2373-2384.
12	黄鹏, 聂枝根, 陈峥, 等. 基于优化Elman神经网络的锂电池容量预测[J]. 储能科学与技术, 2022, 11(7): 2282-2294.
	HUANG P, NIE Z G, CHEN Z, et al. Capacity prediction of lithium battery based on optimized Elman neural network[J]. Energy Storage Science and Technology, 2022, 11(7): 2282-2294.
13	杨彦茹, 温杰, 史元浩, 等. 基于CEEMDAN和SVR的锂离子电池剩余使用寿命预测[J]. 电子测量与仪器学报, 2020, 34(12): 197-205.
	YANG Y R, WEN J, SHI Y H, et al. Remaining useful life prediction for lithium-ion battery based on CEEMDAN and SVR[J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(12): 197-205.
14	王英楷, 张红, 王星辉. 基于1DCNN-LSTM的锂离子电池SOH预测[J]. 储能科学与技术, 2022, 11(1): 240-245.
	WANG Y K, ZHANG H, WANG X H. Hybrid 1DCNN-LSTM model for predicting lithium ion battery state of health[J]. Energy Storage Science and Technology, 2022, 11(1): 240-245.
15	张婷婷, 于明, 李宾, 等. 基于Wavelet降噪和支持向量机的锂离子电池容量预测研究[J]. 电工技术学报, 2020, 35(14): 3126-3136.
	ZHANG T T, YU M, LI B, et al. Capacity prediction of lithium-ion batteries based on wavelet noise reduction and support vector machine[J]. Transactions of China Electrotechnical Society, 2020, 35(14): 3126-3136.
16	张树凯, 刘正江, 张显库, 等. 基于Douglas-Peucker算法的船舶AIS航迹数据压缩[J]. 哈尔滨工程大学学报, 2015, 36(5): 595-599.
	ZHANG S K, LIU Z J, ZHANG X K, et al. A method for AIS track data compression based on Douglas-Peucker algorithm[J]. Journal of Harbin Engineering University, 2015, 36(5): 595-599.
17	杨玉婷, 史玉回, 夏顺仁. 基于讨论机制的头脑风暴优化算法[J]. 浙江大学学报(工学版), 2013, 47(10): 1705-1711, 1746.
	YANG Y T, SHI Y H, XIA S R. Discussion mechanism based brain storm optimization algorithm[J]. Journal of Zhejiang University (Engineering Science), 2013, 47(10): 1705-1711, 1746.
18	马威强, 高永琪, 赵苗. 基于全局最优和差分变异的头脑风暴优化算法[J]. 系统工程与电子技术, 2022, 44(1): 270-278.
	MA W Q, GAO Y Q, ZHAO M. Global-best difference-mutation brain storm optimization algorithm[J]. Systems Engineering and Electronics, 2022, 44(1): 270-278.
19	LIANG Y, HU S S, GUO W S, et al. Abrasive tool wear prediction based on an improved hybrid difference grey wolf algorithm for optimizing SVM[J]. Measurement, 2022, 187: doi: 10.1016/j.measurement. 2021.110247.
20	郑方丹, 姜久春, 陈坤龙, 等. 基于数据统计特性的GS-SVM电池峰值功率预测模型[J]. 电力自动化设备, 2017, 37(9): 56-61.
	ZHENG F D, JIANG J C, CHEN K L, et al. Peak power prediction model for batteries based on data statistical characteristic and GS-SVM[J]. Electric Power Automation Equipment, 2017, 37(9): 56-61.
21	GOEBEL K, SAHA B, SAXENA A, et al. Prognostics in battery health management[J]. IEEE Instrumentation & Measurement Magazine, 2008, 11(4): 33-40.
22	庞晓琼, 王竹晴, 曾建潮, 等. 基于PCA-NARX的锂离子电池剩余使用寿命预测[J]. 北京理工大学学报, 2019, 39(4): 406-412.
	PANG X Q, WANG Z Q, ZENG J C, et al. Prediction for the remaining useful life of lithium-ion battery based on PCA-NARX[J]. Transactions of Beijing Institute of Technology, 2019, 39(4): 406-412.
23	梁海峰, 袁芃, 高亚静. 基于CNN-Bi-LSTM网络的锂离子电池剩余使用寿命预测[J]. 电力自动化设备, 2021, 41(10): 213-219.
	LIANG H F, YUAN P, GAO Y J. Remaining useful life prediction of lithium-ion battery based on CNN-Bi-LSTM network[J]. Electric Power Automation Equipment, 2021, 41(10): 213-219.

项目	规格	项目		规格
外壳材质	镀镍钢	充电方法 (CC/CV) 操作温度	标准	0.5 C_5 A×7.5 h
标称容量	1300 mAh		快速	1 C_5 A×2.5 h
额定容量	3.7 V		充电	0～45 ℃ 32～113 °F
最大充电电压	4.2 V		充电	0～45 ℃ 32～113 °F
放电截止电压	2.7 V		放电	-15～60 ℃ 5～140 °F
最大充电电流	1 C₅A		放电	-15～60 ℃ 5～140 °F
最大放电电流	3 C₅A		贮存	-20～60 ℃ -4～113 °F
内阻(Max at 1000 Hz)	≤25 mΩ		贮存	-20～60 ℃ -4～113 °F

模型	R²	MSE	RMSE	SSE
DBSO-SVM	0.981	2.858	1.691	277
BSO-SVM	0.979	4.378	2.092	425
ABC-SVM	0.978	8.501	2.916	825
SVM	0.963	76.15	8.73	7386

数据编号	方法	实际寿命	预测寿命	AE	R²	RMSE	SSE
B0005	DBSO-SVM	129	129	0	0.99	0.0012	0.000123
	BSO-SVM		129	0	0.99	0.0066	0.003584
	ABC-SVM		137	8	0.99	0.0163	0.022011
	SVM		141	12	0.99	0.0292	0.070760
B0006	DBSO-SVM	113	113	0	0.99	0.0036	0.001049
	BSO-SVM		113	0	0.99	0.0073	0.004466
	ABC-SVM		116	3	0.99	0.0199	0.032895
	SVM		127	14	0.99	0.0553	0.254118
B0007	DBSO-SVM	126	126	0	0.99	0.0021	0.000383
	BSO-SVM		127	1	0.99	0.0051	0.002195
	ABC-SVM		130	4	0.99	0.0115	0.010924
	SVM		132	6	0.99	0.0208	0.035861
B0018	DBSO-SVM	97	97	0	0.94	0.0108	0.006590
	BSO-SVM		98	1	0.93	0.0114	0.007220
	ABC-SVM		100	3	0.92	0.0183	0.018775
	SVM		120	23	0.81	0.0368	0.075779

[1]	刘伟霞, 田勋, 肖家勇, 常伟, 李源, 毛樑. 基于混合模型及LSTM的锂电池SOH与剩余寿命预测[J]. 储能科学与技术, 2021, 10(2): 689-694.
[2]	徐超, 李立伟, 杨玉新, 王凯. 基于改进粒子滤波的锂电池SOH预测[J]. 储能科学与技术, 2020, 9(6): 1954-1960.