可充电锂电池剩余电量预测方法
|
摘要
为了对锂离子电池剩余电量(SOC)进行准确测量,以2 200 mA?h的聚合物锂电池为研究对象,利用Hyperion平衡充放电设备采集6个不同放电电流下(0.7,1.1,1.7,2.2,2.7和3.3A)的放电电压和放电倍率;采取误差反向传播(BP)和径向基函数(RBF) 2个原理不同的神经网络算法进行SOC预测;把采集的样本数据分为训练组和测试组,采用不同的神经网络算法对训练组进行训练后,选择合适的参数构建神经网络,并用测试组数据进行测试;最终比较2种算法的预测效果和误差。研究结果表明:RBF预测结果的相对误差比BP的低,且预测速度更快,RBF较BP更适合于锂离子电池剩余电量的预测。
In order to accurately measure SOC(state of charge) of lithium ion batteries, the discharge voltage and discharge rate of 6 different discharge currents(0.7, 1.1, 1.7, 2.2, 2.7 and 3.3 A) were collected by using Hyperion balanced charge and discharge equipment with 2 200 mA?h polymer lithium battery as the research object. The neural network algorithm based on backpropagation(BP) and radial basis function(RBF) was used. Collected sample data was divided into training group and test group. After training the training group with different neural network algorithms, appropriate parameters were selected to construct the neural network and the test group data were tested. Finally, predicted results and errors of the two algorithms were compared. The results show that the average relative error of RBF prediction is lower than that of BP, and the prediction speed is faster, therefore, RBF is more suitable for the prediction of residual capacity of lithium-ion batteries than BP.
In order to accurately measure SOC(state of charge) of lithium ion batteries, the discharge voltage and discharge rate of 6 different discharge currents(0.7, 1.1, 1.7, 2.2, 2.7 and 3.3 A) were collected by using Hyperion balanced charge and discharge equipment with 2 200 mA?h polymer lithium battery as the research object. The neural network algorithm based on backpropagation(BP) and radial basis function(RBF) was used. Collected sample data was divided into training group and test group. After training the training group with different neural network algorithms, appropriate parameters were selected to construct the neural network and the test group data were tested. Finally, predicted results and errors of the two algorithms were compared. The results show that the average relative error of RBF prediction is lower than that of BP, and the prediction speed is faster, therefore, RBF is more suitable for the prediction of residual capacity of lithium-ion batteries than BP.
引文
[1]汪世国.电动汽车电池管理系统(BMS)现状分析[J].汽车实用技术,2014(2):65-67.WANG Shiguo.Status analysis of battery management system(BMS)for electric vehicles[J].Automobile Practical Technology,2014(2):65-67.
[2]潘佰韬.锂离子电池管理系统的设计与SOC估算研究[D].天津:河北工业大学车辆工程系,2015:12PAN Baitao.Design and SOC estimation of Li ion battery management system[D].Tianjin:Department of vehicle engineering,Hebei University of Technology,2015:12
[3]AUNG H,LOW K S,SHU T G.State-of-charge estimation of lithium-ion battery using square root spherical unscented Kalman filter(Sqrt-UKFST)in nanosatellite[J].IEEETransactions on Power Electronics,2015,30(9):4774-4783.
[4]尹安东,张万兴,赵韩,等.基于神经网络的磷酸铁锂电池SOC预测研究[J].电子测量与仪器学报,2011,25(5):433-437.YIN Andong,ZHANG Wanxing,ZHAO Han,et al.Study on SOC prediction of lithium iron phosphate battery based on neural networks[J].Journal of Electronic Measurement and Instrumentation,2011,25(5):433-437.
[5]范兴明,曾求勇,张鑫.基于改进安时积分法的电动汽车电池SOC估计与仿真研究[J].电气应用,2015,34(8):111-115.FAN Xingming,ZENG Qiuyong,ZHANG Xin.Improved ah integral method of electric vehicle battery SOC estimation and Simulation of electrical[J].Electrotechnical Application,2015,34(8):111-115.
[6]季迎旭,杜海江,孙航.蓄电池SOC估算方法综述[J].电测与仪表,2014,51(4):18-22.JI Yingxu,DU Haijiang,SUN Hang.Summarization of battery SOC estimation methods[J].Electrical Measurement and Instrumentation,2014,51(4):18-22.
[7]孟仙雅,刘立炳,惠怀兵,等.放电深度对电池使用寿命的影响[J].汽车科技,2016(3):47-51.MENG Xianya,LIU Libing,HUI Huaibing,et al.Influence of discharge depth on battery life[J].Automotive Science and Technology,2016(3):47-51.
[8]邓涛,孙欢.锂离子电池新型SOC安时积分实时估算方法[J].重庆理工大学学报(自然科学),2015,29(1):101-106.DENG Tao,SUN Huan.A new real-time SOC time integration estimation method for Li ion battery[J].Journal of Chongqing University of Technology(Natural Science),2015,29(1):101-106.
[9]余少波,严隽永.神经网络应用[J].海军工程大学学报,1990(2):61-65.YU Shaobo,YAN Juanyong.Neural network application[J].Journal of Naval University of Engineering,1990(2):61-65.
[10]刘永,张立毅.BP和RBF神经网络的实现及其性能比较[J].电子测量技术,2007,30(4):77-80.LIU Yong,ZHANG Liyi.Implementation and performance comparison of BP and RBF neural networks[J].Electronic Measurement Technology,2007,30(4):77-80.
[11]KARAYIANNIS NICOLAOS B.Reformulated radial basis neural networks trained by gradient descent[J].IEEE Transactions on Neural Networks,1999,10(3):10-27.
[12]李俭川,秦国军,温熙森,等.神经网络学习算法的过拟合问题及解决方法[J].振动、测试与诊断,2002,22(4):260-264.LI Jianchuan,QIN Guojun,WEN Xisen,et al.Over fitting problem of neural network learning algorithm and its solution[J].Vibration,Testing and Diagnosis,2002,22(4):260-264.
[13]周俊武,孙传尧,王福利.径向基函数(RBF)网络的研究及实现[J].矿冶,2001,10(4):71-75.ZHOU Junwu,SUN Chuanyao,WANG Fuli.Research and implementation of radial basis function(RBF)network[J].Mine and Metallurgy,2001,10(4):71-75.
[14]颜七笙,游泳,杨志辉.BP网络结构设计与算法探讨[J].科技广场,2005(3):36-38.YAN Qisheng,You Yong,YANG Zhihui.BP network structure design and algorithm discussion[J].Science and Technology Plaza,2005(3):36-38.
[15]PRIMADUSI U,CAHYADI A I,WAHYUNGGORO O.The comparison of RBF NN and BPNN for SOC estimation of LiFePO4battery[C]//AIP Conference Proceedings 1755.Yogyakarta,INDONESIA:AIP Publishing,2016:090010-1-090010-7.
[16]GUO Yifeng,ZHAO Zeshuang,HUANG Limin.SoC estimation of Lithium battery based on improved BP neural network[J].Energy Procedia,2017,105:4153-4158.
[17]陈强,王洪佩.基于BP神经网络的SOC估算策略研究[J].汽车与安全,2011(8):58-60.CHEN Jiang,WANG Hongpei.Research on SOC estimation strategy based on BP neural network[J].Automotive and Safety,2011(8):58-60.
[18]郭利辉,周雅.基于MATLAB神经网络工具箱的BP网络设计[J].信息技术与信息化,2009(3):20-22.GUO Lihui,ZHOU ya.Design of BP network based on MATLAB neural network toolbox[J].Information Technology and Informatization,2009(3):20-22.
[19]HUANG Sirun,MA Yenhuai,HENG Shiyun,et al.The SOCestimation of LCO battery based on BP neural network[M]//Energy,Transportation and Global Warming.New York,USA:Springer International Publishing,2016:543-552.
[20]高宁,张建中.MATLAB在RBF神经网络模型中的应用[J].农业网络信息,2009(2):111-112,117.GAO Ning,ZHANG Jianzhong.Application of MATLAB in RBF neural network model[J].Agricultural Network Information,2009(2):111-112,117.
[21]SHARMA V.Analysis of spread spectrum in MATLAB[J].International Journal of Scientific and Engineering Research,2014,5:1899-1902.
[22]史艳霞,乔佳.基于RBF网络的锂电池SOC估算研究[J].自动化与仪表,2015,30(9):89-92.SHI Yanxia,QIAO Jia.Research on SOC estimation of lithium battery based on RBF network[J].Automation and Instrumentation,2015,30(9):89-92.
[2]潘佰韬.锂离子电池管理系统的设计与SOC估算研究[D].天津:河北工业大学车辆工程系,2015:12PAN Baitao.Design and SOC estimation of Li ion battery management system[D].Tianjin:Department of vehicle engineering,Hebei University of Technology,2015:12
[3]AUNG H,LOW K S,SHU T G.State-of-charge estimation of lithium-ion battery using square root spherical unscented Kalman filter(Sqrt-UKFST)in nanosatellite[J].IEEETransactions on Power Electronics,2015,30(9):4774-4783.
[4]尹安东,张万兴,赵韩,等.基于神经网络的磷酸铁锂电池SOC预测研究[J].电子测量与仪器学报,2011,25(5):433-437.YIN Andong,ZHANG Wanxing,ZHAO Han,et al.Study on SOC prediction of lithium iron phosphate battery based on neural networks[J].Journal of Electronic Measurement and Instrumentation,2011,25(5):433-437.
[5]范兴明,曾求勇,张鑫.基于改进安时积分法的电动汽车电池SOC估计与仿真研究[J].电气应用,2015,34(8):111-115.FAN Xingming,ZENG Qiuyong,ZHANG Xin.Improved ah integral method of electric vehicle battery SOC estimation and Simulation of electrical[J].Electrotechnical Application,2015,34(8):111-115.
[6]季迎旭,杜海江,孙航.蓄电池SOC估算方法综述[J].电测与仪表,2014,51(4):18-22.JI Yingxu,DU Haijiang,SUN Hang.Summarization of battery SOC estimation methods[J].Electrical Measurement and Instrumentation,2014,51(4):18-22.
[7]孟仙雅,刘立炳,惠怀兵,等.放电深度对电池使用寿命的影响[J].汽车科技,2016(3):47-51.MENG Xianya,LIU Libing,HUI Huaibing,et al.Influence of discharge depth on battery life[J].Automotive Science and Technology,2016(3):47-51.
[8]邓涛,孙欢.锂离子电池新型SOC安时积分实时估算方法[J].重庆理工大学学报(自然科学),2015,29(1):101-106.DENG Tao,SUN Huan.A new real-time SOC time integration estimation method for Li ion battery[J].Journal of Chongqing University of Technology(Natural Science),2015,29(1):101-106.
[9]余少波,严隽永.神经网络应用[J].海军工程大学学报,1990(2):61-65.YU Shaobo,YAN Juanyong.Neural network application[J].Journal of Naval University of Engineering,1990(2):61-65.
[10]刘永,张立毅.BP和RBF神经网络的实现及其性能比较[J].电子测量技术,2007,30(4):77-80.LIU Yong,ZHANG Liyi.Implementation and performance comparison of BP and RBF neural networks[J].Electronic Measurement Technology,2007,30(4):77-80.
[11]KARAYIANNIS NICOLAOS B.Reformulated radial basis neural networks trained by gradient descent[J].IEEE Transactions on Neural Networks,1999,10(3):10-27.
[12]李俭川,秦国军,温熙森,等.神经网络学习算法的过拟合问题及解决方法[J].振动、测试与诊断,2002,22(4):260-264.LI Jianchuan,QIN Guojun,WEN Xisen,et al.Over fitting problem of neural network learning algorithm and its solution[J].Vibration,Testing and Diagnosis,2002,22(4):260-264.
[13]周俊武,孙传尧,王福利.径向基函数(RBF)网络的研究及实现[J].矿冶,2001,10(4):71-75.ZHOU Junwu,SUN Chuanyao,WANG Fuli.Research and implementation of radial basis function(RBF)network[J].Mine and Metallurgy,2001,10(4):71-75.
[14]颜七笙,游泳,杨志辉.BP网络结构设计与算法探讨[J].科技广场,2005(3):36-38.YAN Qisheng,You Yong,YANG Zhihui.BP network structure design and algorithm discussion[J].Science and Technology Plaza,2005(3):36-38.
[15]PRIMADUSI U,CAHYADI A I,WAHYUNGGORO O.The comparison of RBF NN and BPNN for SOC estimation of LiFePO4battery[C]//AIP Conference Proceedings 1755.Yogyakarta,INDONESIA:AIP Publishing,2016:090010-1-090010-7.
[16]GUO Yifeng,ZHAO Zeshuang,HUANG Limin.SoC estimation of Lithium battery based on improved BP neural network[J].Energy Procedia,2017,105:4153-4158.
[17]陈强,王洪佩.基于BP神经网络的SOC估算策略研究[J].汽车与安全,2011(8):58-60.CHEN Jiang,WANG Hongpei.Research on SOC estimation strategy based on BP neural network[J].Automotive and Safety,2011(8):58-60.
[18]郭利辉,周雅.基于MATLAB神经网络工具箱的BP网络设计[J].信息技术与信息化,2009(3):20-22.GUO Lihui,ZHOU ya.Design of BP network based on MATLAB neural network toolbox[J].Information Technology and Informatization,2009(3):20-22.
[19]HUANG Sirun,MA Yenhuai,HENG Shiyun,et al.The SOCestimation of LCO battery based on BP neural network[M]//Energy,Transportation and Global Warming.New York,USA:Springer International Publishing,2016:543-552.
[20]高宁,张建中.MATLAB在RBF神经网络模型中的应用[J].农业网络信息,2009(2):111-112,117.GAO Ning,ZHANG Jianzhong.Application of MATLAB in RBF neural network model[J].Agricultural Network Information,2009(2):111-112,117.
[21]SHARMA V.Analysis of spread spectrum in MATLAB[J].International Journal of Scientific and Engineering Research,2014,5:1899-1902.
[22]史艳霞,乔佳.基于RBF网络的锂电池SOC估算研究[J].自动化与仪表,2015,30(9):89-92.SHI Yanxia,QIAO Jia.Research on SOC estimation of lithium battery based on RBF network[J].Automation and Instrumentation,2015,30(9):89-92.