State of Health Estimation of Li-ion Batteries Based on GWO-LSSVM
-
摘要: 针对目前应用于电池健康状态(SOH)估算的算法需用大量数据样本来进行训练且估算效果不佳的问题, 提出了一种基于灰狼优化(GWO)算法的最小二乘支持向量机(LSSVM)算法来估算电池SOH, 依据灰色关联度分析法筛选出恒流充电时间作为适合估算电池SOH的输入特征参数。以18650钴酸锂电池充放电循环试验为例, 采用所建立的算法模型在不同比例的训练集样本下对不同容量规格的电池进行SOH估算, 并将估算结果与基于网格搜索法的LSSVM算法、基于粒子群优化算法的LSSVM算法的估算结果进行对比, 结果表明, 基于GWO算法的LSSVM算法模型适用于小样本数据且估算误差较小, 用于电池SOH估算的效果更好。Abstract: The algorithms currently applied to state of health(SOH) estimation require numerous data samples for training and the estimation effect is not good. To address this issue, this study proposed a least-squares support vector machine(LSSVM) algorithm based on the grey wolf optimization(GWO) algorithm to estimate the SOH using the grey relational analysis method to choose constant current charging time as the input characteristic. Considering the 18650 lithium cobalt oxide battery charge/discharge cycle test as an example, the established algorithm model was used to estimate the SOH of batteries with different capacity specifications under different proportions of training set samples. The estimated results were compared with those obtained by the LSSVM algorithm based on the grid search method and the LSSVM algorithm based on the particle swarm optimization algorithm. The experimental results showed that the LSSVM algorithm model based on the GWO algorithm is suitable for small-sample data and is characterized by small estimation errors; therefore, it is more effective for battery SOH.
-
图 7 第1组电池的估算结果与真实值对比
Figure 7. Comparison between estimated results and actual values of batteries from group 1
图 8 第2组电池的估算结果与真实值对比
Figure 8. Comparison between estimated results and actual values of batteries from group 2
图 9 不同算法在训练样本为30%时的估算结果对比
Figure 9. Comparison of estimated results by different algorithms when the training sample proportion is 30%
图 10 不同算法在训练样本为60%时的估算结果对比
Figure 10. Comparison of estimated results by different algorithms when the training sample proportion is 60%
图 11 不同算法在训练样本为90%时的估算结果对比
Figure 11. Comparison of estimated results of different algo- rithms when the training sample proportion is 90%
表 1 各个参数与容量的关联度均值
Table 1. Mean value of relational grade between capacity and each factor
参数 关联度 恒流充电时间 0.939 6 充电时间 0.902 5 充电30 min时的端电压 0.883 6 等压段充电时间 0.819 7 恒压充电时间 0.803 0 
下载: 导出CSV
表 2 第1组电池的超参数寻优结果
Table 2. Optimization results of hyper-parameter of batteries from group 1
超参数 训练集占比 30% 60% 90% c $ 7.95 \times {10^3} $ $ 4.97 \times {10^4} $ $ 1.42 \times {10^4} $ σ $ 5.59 \times {10^7} $ $ {10^8} $ $ 9.99 \times {10^7} $
下载: 导出CSV
表 3 第1组电池的估算精度对比
Table 3. Comparison of estimation accuracy for batteries from group 1
指标 训练集占比 30% 60% 90% RMSE 0.0197 0.0180 0.0178 MAE 0.0142 0.0137 0.0136
下载: 导出CSV
表 4 第2组电池的超参数寻优结果
Table 4. Optimization results of hyperparameter of batteries from group 2
超参数 估算 估算1 估算2 估算3 c $ 1.02 \times {10^7} $ $ 8.23 \times {10^6} $ $ 1.04 \times {10^6} $ σ $ {10^8} $ $ {10^8} $ $ {10^8} $
下载: 导出CSV
表 5 第2组电池的估算精度对比
Table 5. Comparison of estimation accuracy for batteries from group 2
指标 估算 估算1 估算2 估算3 RMSE 0.0200 0.0123 0.004 MAE 0.0177 0.0105 0.0029
下载: 导出CSV
表 6 2组电池在不同算法不同训练集样本下的估算性能指标对比
Table 6. Comparison of estimation performance specifications of different algorithms for the two groups of batteries with different proportions of training set samples
电池 优化
算法训练集样本比例 RMSE MAE 1 PSO-LSSVM 30% 0.0440 0.0345 60% 0.0263 0.0212 90% 0.0203 0.0157 GWO-LSSVM 30% 0.0197 0.0142 60% 0.0180 0.0137 90% 0.0178 0.0136 2 PSO-LSSVM 30% 0.0273 0.0246 60% 0.0126 0.0105 90% 0.0040 0.0029 GWO-LSSVM 30% 0.0200 0.0177 60% 0.0123 0.0105 90% 0.0040 0.0029
下载: 导出CSV
-
[1] 张金龙, 佟微, 孙叶宁, 等. 锂电池健康状态估算方法综述[J]. 电源学报, 2017, 15(2): 128-134.Zhang Jin-long, Tong Wei, Sun Ye-ning, et al. Summarize of Lithium Battery Status of Health Estimation Method[J]. Journal of Power Supply, 2017, 15(2): 128-134. [2] Khare N, Singh P, Vassiliou J K. A Novel Magnetic Field Probing Technique for Determining State of Health of Sealed Lead-Acid Batteries[J]. Journal of Power Sources, 2012, 218: 462-473. doi: 10.1016/j.jpowsour.2012.06.085 [3] Li Y, Abdel-Monem M, Gopalakrishnan R, et al. A Quick On-Line State of Health Estimation Method for Li-Ion Battery with Incremental Capacity Curves Processed by Gaussian Filter[J]. Journal of Power Sources, 2018, 373: 40-53. doi: 10.1016/j.jpowsour.2017.10.092 [4] Ning G, White R E, Popov B N. A Generalized Cycle Life Model of Rechargeable Li-Ion Batteries[J]. Electrochimica Acta, 2006, 51(10): 2012-2022. doi: 10.1016/j.electacta.2005.06.033 [5] Remmlinger J, Buchholz M, Meiler M, et al. State-of- Health Monitoring of Lithium-Ion Batteries in Electric Vehicles by on-Board Internal Resistance Estimation[J]. Journal of Power Sources, 2011, 196(12): 5357-5363. doi: 10.1016/j.jpowsour.2010.08.035 [6] 何发尧, 胡欲立, 郭广华, 等. 基于人工神经网络估算锂离子电池的SOH[J]. 电源技术, 2017, 41(5): 708-710. doi: 10.3969/j.issn.1002-087X.2017.05.013He Fa-yao, Hu Yu-li, Guo Guang-hua, et al. State of Health Estimation for Lithium-ion Batteries Based on ANN[J]. Chinese Journal of Power Sources, 2017, 41(5): 708-710. doi: 10.3969/j.issn.1002-087X.2017.05.013 [7] Xie J, Li W, Hu Y. Aviation Lead-Acid Battery State-of-Health Assessment Using PSO-SVM Technique[C]//Proceedings of 2014 IEEE 5th International Conference on Software Engineering and Service Science. Beijing: IEEE, 2014. [8] Widodo A, Shim M C, Caesarendra W, et al. Intelligent Prognostics for Battery Health Monitoring Based on Sample Entropy[J]. Expert Systems with Applications, 2011, 38(9): 11763-11769. doi: 10.1016/j.eswa.2011.03.063 [9] 刘思峰, 蔡华, 杨英杰. 灰色关联分析模型研究进展[J]. 系统工程理论与实践, 2013, 33(8): 2041-2046. doi: 10.3969/j.issn.1000-6788.2013.08.018Liu Si-feng, Cai Hua, Yang Ying-jie, et al. Advance in Grey Incidence Analysis Modelling[J]. Systems Engineerings-Theory & Practice, 2013, 33(8): 2041-2046. doi: 10.3969/j.issn.1000-6788.2013.08.018 [10] 顾燕萍, 赵文杰, 吴占松. 最小二乘支持向量机的算法研究[J]. 清华大学学报(自然科学版), 2010, 50(7): 1063-1066.Gu Yan-ping, Zhao Wen-jie, Wu Zhan-song. Least Squares Support Vector Machine Algorithm[J]. Journal of Tsinghua University(Science and Technology), 2010, 50(7): 1063-1066. [11] Mirjalili S, Mirjalili S M, Lewis A. Grey Wolf Optimizer[J]. Advances in Engineering Software, 2014, 69: 46-61. doi: 10.1016/j.advengsoft.2013.12.007 -
点击查看大图
计量
- 文章访问数: 214
- HTML全文浏览量: 111
- PDF下载量: 32
- 被引次数: 0
