基于增量学习相关向量机的锂离子电池SOC预测方法
|
摘要
针对锂离子电池荷电状态(SOC)预测精度不高以及在线适应性差的问题,提出一种改进的增量学习相关向量机模型对锂离子电池SOC进行在线预测。选择锂离子电池电压、充放电电流和表面温度作为模型的输入,SOC作为模型的输出,构造模型的训练集。选用快速序列稀疏贝叶斯学习算法进行训练,并结合增量学习法建立增量学习相关向量机模型进行锂离子电池SOC在线预测方法研究。研究发现,通过自动调整核参数的方法,可以保证有较高的预测精度。算法验证实验表明,核参数可以控制算法的预测精度和计算效率,该算法预测精度高、计算速度快、通用性强,可为锂离子电池SOC的预测与应用提供参考。
Aiming at the problems of low prediction accuracy and poor online adaptability for state of charge(SOC) of li-ion battery,an improved incremental relevance vector machine(RVM) model is proposed to predict the SOC of li-ion battery online.The measured voltage,discharge current and surface temperature are selected as the model input,and the SOC is used as the model output,both of them are constructed the training set of the model.A fast-sequence sparse Bayesian learning algorithm is chosen to train RVM,and the incremental learning relevance vector machine model was established by connecting RVM algorithm with incremental learning method to research the prediction for SOC of liion battery online.The study found that this model can guarantee a higher prediction accuracy by adjusting the kernel parameters automatically.The result of the experiment indicates that the kernel parameters can control the prediction accuracy and calculation efficiency of the algorithm,and the IRVM algorithm has the characteristics of high prediction accuracy,fast calculation speed and strong universality,in terms of prodiction and application of the SOC li-ion battery,the algorithm can provide a reference for it.
Aiming at the problems of low prediction accuracy and poor online adaptability for state of charge(SOC) of li-ion battery,an improved incremental relevance vector machine(RVM) model is proposed to predict the SOC of li-ion battery online.The measured voltage,discharge current and surface temperature are selected as the model input,and the SOC is used as the model output,both of them are constructed the training set of the model.A fast-sequence sparse Bayesian learning algorithm is chosen to train RVM,and the incremental learning relevance vector machine model was established by connecting RVM algorithm with incremental learning method to research the prediction for SOC of liion battery online.The study found that this model can guarantee a higher prediction accuracy by adjusting the kernel parameters automatically.The result of the experiment indicates that the kernel parameters can control the prediction accuracy and calculation efficiency of the algorithm,and the IRVM algorithm has the characteristics of high prediction accuracy,fast calculation speed and strong universality,in terms of prodiction and application of the SOC li-ion battery,the algorithm can provide a reference for it.
引文
[1]佟晶晶,温俊强,王丹,等.基于分时电价的电动汽车多目标优化充电策略[J].电力系统保护与控制,2016,44(1):17-23.Tong Jingjing,Wen Junqiang,Wang Dan,et al.Multiobjective optimization charging strategy for plug-in electric vehicles based on time-of-use price[J].Power System Protection and Control,2016,44(1):17-23.
[2]冯飞,宋凯,逯仁贵,等.磷酸铁锂电池组均衡控制策略及荷电状态估计[J].电工技术学报,2015,30(1):22-29.Feng Fei,Song Kai,Lu Rengui,et al.Equalization control strategy and SOC estimation for LiFePO4battery pack[J].Transactions of China Electrotechnical Society,2015,30(1):22-29.
[3]郑旭,黄鸿,郭汾.动力电池SOC估算复杂方法综述[J].电子技术应用,2018,44(3):3-6,10.Zheng Xu,Huang Hong,Guo Fen.Review of complex methods for state of charge estimation of power battery[J].Application of Electronic Technique,2018,44(3):3-6,10.
[4]谢长君,费亚龙,曾春年,等.基于无迹粒子滤波的车载锂离子电池状态估计[J].电工技术学报,2018,33(17):3958-3964.Xie Changjun,Fei Yalong,Zeng Chunnian,et al.State of charge estimation of lithium-ion battery using unscented particle filter in vehicle[J].Transactions of China Electrotechnical Society,2018,33(17):3958-3964.
[5]刘毅,谭国俊,何晓群.优化电池模型的自适应Sigma卡尔曼荷电状态估算[J].电工技术学报,2017,32(2):108-118.Liu Yi,Tan Guojun,He Xiaoqun.Optimized battery model based adaptive sigma Kalman filter for state of charge estimation[J].Transactions of China Electrotechnical Society,2017,32(2):108-118.
[6]潘海鸿,吕治强,李君子,等.基于灰色扩展卡尔曼滤波的锂离子电池荷电状态估算[J].电工技术学报,2017,32(21):1-8.Pan Haihong,LüZhiqiang,Li Junzi,et al.Estimation of lithium-ion battery state of charge based on grey prediction model-extended Kalman filter[J].Transactions of China Electrotechnical Society,2017,32(21):1-8.
[7]He Hongwen,Xiong Rui,Zhang Xiaowei,et al.Stateof-charge estimation of the lithium-ion battery using an adaptive extended Kalman filter based on an improved thevenin model[J].IEEE Transactions on Vehicular Technology,2011,60(4):1461-1469.
[8]张金龙,佟微,漆汉宏,等.平方根采样点卡尔曼滤波在磷酸铁锂电池组荷电状态估算中的应用[J].中国电机工程学报,2016,36(22):6246-6253.Zhang Jinlong,Tong Wei,Qi Hanhong,et al.Application of square root sigma point Kalman filter to SOC estimation of LiFePO4 battery pack[J].Proceedings of the CSEE,2016,36(22):6246-6253.
[9]Plett G L.Extended Kalman filtering for battery management systems of LiPB-based HEV battery packs[J].Journal of Power Sources,2004,134(2):252-292.
[10]雷肖,陈清泉,刘开培,等.电动车蓄电池荷电状态估计的神经网络方法[J].电工技术学报,2007,22(8):155-160.Lei Xiao,Chen Qingquan,Liu Kaipei,et al.Battery state of charge estimation based on neural-network for electric vehicles[J].Transactions of China Electrotechnical Society,2007,22(8):155-160.
[11]Ruping S.Incremental learning with support vector machines[J].Technical Reports,2001,228(4):641-642.
[12]刘大同,周建宝,郭力萌,等.锂离子电池健康评估和寿命预测综述[J].仪器仪表学报,2015,36(1):1-16.Liu Datong,Zhou Jianbao,Guo Limeng,et al.Survey on lithium-ion battery health assessment and cycle life estimation[J].Chinese Journal of Scientific Instrument,2015,36(1):1-16.
[13]高向阳,张骏,宁宁.基于相关向量机的蓄电池荷电状态预测[J].电源技术,2010,34(12):1273-1275.Gao Xiangyang,Zhang Jun,Ning Ning.SOCprediction for battery based on relevance vector machine[J].Chinese Journal of Power Sources,2010,34(12):1273-1275.
[14]Liu Datong,Zhou Jianbao,Pan Dawei,et al.Lithiumion battery remaining useful life estimation with an optimized relevance vector machine algorithm with incremental learning[J].Measurement,2015,63:143-151.
[15]周建宝,王少军,马丽萍,等.可重构卫星锂离子电池剩余寿命预测系统研究[J].仪器仪表学报,2013,34(9):2034-2044.Zhou Jianbao,Wang Shaojun,Ma Liping,et al.Study on the reconfigurable remaining useful life estimation system for satellite lithium-ion battery[J].Chinese Journal of Scientific Instrument,2013,34(9):2034-2044
[16]孙国强,卫志农,翟玮星.基于RVM与ARMA误差校正的短期风速预测[J].电工技术学报,2012,27(8):187-193.Sun Guoqiang,Wei Zhinong,Zhai Weixing.Short term wind speed forecasting based on RVM and ARMA error correcting[J].Transactions of China Electrotechnical Society,2012,27(8):187-193.
[17]Tipping M E.Sparse Bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001,1(3):211-244.
[18]Tipping M E,Faul A C.Fast marginal likelihood maximisation for sparse Bayesian models[C]//Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics,Key West,2003:3-6
[19]冯飞,逯仁贵,朱春波.一种锂离子电池低温SOC估计算法[J].电工技术学报,2014,29(7):53-58.Feng Fei,Lu Rengui,Zhu Chunbo.State of charge estimation of li-ion battery at low temperature[J].Transactions of China Electrotechnical Society,2014,29(7):53-58.
[2]冯飞,宋凯,逯仁贵,等.磷酸铁锂电池组均衡控制策略及荷电状态估计[J].电工技术学报,2015,30(1):22-29.Feng Fei,Song Kai,Lu Rengui,et al.Equalization control strategy and SOC estimation for LiFePO4battery pack[J].Transactions of China Electrotechnical Society,2015,30(1):22-29.
[3]郑旭,黄鸿,郭汾.动力电池SOC估算复杂方法综述[J].电子技术应用,2018,44(3):3-6,10.Zheng Xu,Huang Hong,Guo Fen.Review of complex methods for state of charge estimation of power battery[J].Application of Electronic Technique,2018,44(3):3-6,10.
[4]谢长君,费亚龙,曾春年,等.基于无迹粒子滤波的车载锂离子电池状态估计[J].电工技术学报,2018,33(17):3958-3964.Xie Changjun,Fei Yalong,Zeng Chunnian,et al.State of charge estimation of lithium-ion battery using unscented particle filter in vehicle[J].Transactions of China Electrotechnical Society,2018,33(17):3958-3964.
[5]刘毅,谭国俊,何晓群.优化电池模型的自适应Sigma卡尔曼荷电状态估算[J].电工技术学报,2017,32(2):108-118.Liu Yi,Tan Guojun,He Xiaoqun.Optimized battery model based adaptive sigma Kalman filter for state of charge estimation[J].Transactions of China Electrotechnical Society,2017,32(2):108-118.
[6]潘海鸿,吕治强,李君子,等.基于灰色扩展卡尔曼滤波的锂离子电池荷电状态估算[J].电工技术学报,2017,32(21):1-8.Pan Haihong,LüZhiqiang,Li Junzi,et al.Estimation of lithium-ion battery state of charge based on grey prediction model-extended Kalman filter[J].Transactions of China Electrotechnical Society,2017,32(21):1-8.
[7]He Hongwen,Xiong Rui,Zhang Xiaowei,et al.Stateof-charge estimation of the lithium-ion battery using an adaptive extended Kalman filter based on an improved thevenin model[J].IEEE Transactions on Vehicular Technology,2011,60(4):1461-1469.
[8]张金龙,佟微,漆汉宏,等.平方根采样点卡尔曼滤波在磷酸铁锂电池组荷电状态估算中的应用[J].中国电机工程学报,2016,36(22):6246-6253.Zhang Jinlong,Tong Wei,Qi Hanhong,et al.Application of square root sigma point Kalman filter to SOC estimation of LiFePO4 battery pack[J].Proceedings of the CSEE,2016,36(22):6246-6253.
[9]Plett G L.Extended Kalman filtering for battery management systems of LiPB-based HEV battery packs[J].Journal of Power Sources,2004,134(2):252-292.
[10]雷肖,陈清泉,刘开培,等.电动车蓄电池荷电状态估计的神经网络方法[J].电工技术学报,2007,22(8):155-160.Lei Xiao,Chen Qingquan,Liu Kaipei,et al.Battery state of charge estimation based on neural-network for electric vehicles[J].Transactions of China Electrotechnical Society,2007,22(8):155-160.
[11]Ruping S.Incremental learning with support vector machines[J].Technical Reports,2001,228(4):641-642.
[12]刘大同,周建宝,郭力萌,等.锂离子电池健康评估和寿命预测综述[J].仪器仪表学报,2015,36(1):1-16.Liu Datong,Zhou Jianbao,Guo Limeng,et al.Survey on lithium-ion battery health assessment and cycle life estimation[J].Chinese Journal of Scientific Instrument,2015,36(1):1-16.
[13]高向阳,张骏,宁宁.基于相关向量机的蓄电池荷电状态预测[J].电源技术,2010,34(12):1273-1275.Gao Xiangyang,Zhang Jun,Ning Ning.SOCprediction for battery based on relevance vector machine[J].Chinese Journal of Power Sources,2010,34(12):1273-1275.
[14]Liu Datong,Zhou Jianbao,Pan Dawei,et al.Lithiumion battery remaining useful life estimation with an optimized relevance vector machine algorithm with incremental learning[J].Measurement,2015,63:143-151.
[15]周建宝,王少军,马丽萍,等.可重构卫星锂离子电池剩余寿命预测系统研究[J].仪器仪表学报,2013,34(9):2034-2044.Zhou Jianbao,Wang Shaojun,Ma Liping,et al.Study on the reconfigurable remaining useful life estimation system for satellite lithium-ion battery[J].Chinese Journal of Scientific Instrument,2013,34(9):2034-2044
[16]孙国强,卫志农,翟玮星.基于RVM与ARMA误差校正的短期风速预测[J].电工技术学报,2012,27(8):187-193.Sun Guoqiang,Wei Zhinong,Zhai Weixing.Short term wind speed forecasting based on RVM and ARMA error correcting[J].Transactions of China Electrotechnical Society,2012,27(8):187-193.
[17]Tipping M E.Sparse Bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001,1(3):211-244.
[18]Tipping M E,Faul A C.Fast marginal likelihood maximisation for sparse Bayesian models[C]//Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics,Key West,2003:3-6
[19]冯飞,逯仁贵,朱春波.一种锂离子电池低温SOC估计算法[J].电工技术学报,2014,29(7):53-58.Feng Fei,Lu Rengui,Zhu Chunbo.State of charge estimation of li-ion battery at low temperature[J].Transactions of China Electrotechnical Society,2014,29(7):53-58.