多核相关向量机优化模型的锂电池剩余寿命预测方法
|
摘要
基于相关向量机的剩余寿命预测方法,核函数是影响相关向量机模型预测性能的重要因素.目前的相关向量机预测模型以单核为主,且核函数的选择存在较大主观性,导致所构建的预测模型性能有限.本文提出一种融合多个核函数构建相关向量机预测模型的方法,通过果蝇算法优化多个核函数优化组合的线性方程系数,提高了模型的预测性能,并将该方法应用于预测锂离子电池的循环剩余寿命.分别采用美国NASA和马里兰大学的电池退化数据集,对本文的方法进行了实验验证.实验结果表明:多核相关向量机预测方法的平均绝对误差和均方根误差都小于最优的单核相关向量机预测方法.
For the remaining useful life(RUL) prediction method based on relevance vector machine(RVM),kernel function is the important item of RVM model for the final prediction result.The current RVM prediction models are dominated by single kernel,and the selection of RVM kernel is a little bit subjective.So,the prediction performance of the constructed RVM model is limited.To address this problem,a multi kernel RVM model is proposed for the RUL estimation,using the fruit fly optimization algorithm(FOA) to find the best corresponding coefficients of multi kernel in the linear combination of multi kernel functions,and to improve the prediction performance of RVM model applied in the RUL estimation of lithium-ion battery.The battery test data sets of the national aeronautics and space administration(NASA) and the center of advanced life cycle engineering(CACLE) in the university of Maryland are used respectively.Experiments have been carried out to test the performance of the proposed method.The results show that the mean absolute error(MAE) and root mean square error(RMSE) of multi kernel RVM method are both less than the single kernel RVM algorithm.
For the remaining useful life(RUL) prediction method based on relevance vector machine(RVM),kernel function is the important item of RVM model for the final prediction result.The current RVM prediction models are dominated by single kernel,and the selection of RVM kernel is a little bit subjective.So,the prediction performance of the constructed RVM model is limited.To address this problem,a multi kernel RVM model is proposed for the RUL estimation,using the fruit fly optimization algorithm(FOA) to find the best corresponding coefficients of multi kernel in the linear combination of multi kernel functions,and to improve the prediction performance of RVM model applied in the RUL estimation of lithium-ion battery.The battery test data sets of the national aeronautics and space administration(NASA) and the center of advanced life cycle engineering(CACLE) in the university of Maryland are used respectively.Experiments have been carried out to test the performance of the proposed method.The results show that the mean absolute error(MAE) and root mean square error(RMSE) of multi kernel RVM method are both less than the single kernel RVM algorithm.
引文
[1] 胡昌华,王兆强,周志杰,司小胜.一种RVM模糊模型辨识方法及在故障预报中的应用[J].自动化学报,2011,37(04):503-512.HU Chang-hua,WANG Zhao-qiang,ZHOU Zhi-jie,SI Xiao-sheng.A RVM fuzzy model identification method and its application to fault prediction[J].Acta Automatic Sinica,2011,37(04):503-512.(in Chinese)
[2] 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(09):2034-44.ZHOU Jian-bao,WANG Shao-jun,MA Li-ping,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):2034-44.(in Chinese)
[3] WANG D,MIAO Q,PECHT M.Prognostics of lithium-ion batteries based on relevance vectors and a conditional three-parameter capacity degradation model[J].Journal of Power Sources,2013,239:253-64.
[4] HU Y,LUO P CH.Performance data prognostics based on relevance vector machine and particle filter[J].Chemical Engineering Transactions,2013,33:349-354.
[5] 韩敏,许美玲,任伟杰.多元混沌时间序列的相关状态机预测模型研究[J].自动化学报,2014,40(05):822-829.HAN Min,XU Mei-ling,REN Wei-jie.Reasearch on multivariate chaotic time series prediction using mRSM model[J].Acta Automatic Sinica,2014,40(05):822-829.(in Chinese)
[6] 刘月峰,赵光权,彭喜元.锂离子电池循环寿命的融合预测方法[J].仪器仪表学报,2015,36(07):1462-1469.LIU Yue-feng,ZHAO Guang-quan,PENG Xi-yuan.A fusion prediction method of lithiumion battery cycle-life[J].Chinese Journal of Scientific Instrument,2015,36(07):1462-1469.(in Chinese)
[7] 雷亚国,陈吴,李乃鹏,林京.自适应多核组合相关向量机预测方法及其在机械设备剩余寿命预测中的应用[J].机械工程学报,2016,52(01):87-93.LEI Ya-guo,CHEN Wu,LI Nai-peng,LIN Jing.A relevance vector machine prediction method based on adaptive multi-kernel combination and its application to remaining useful life prediction of machinery[J].Chinese Journal of Mechanical Engineering,2016,52(01):87-93.(in Chinese)
[8] 尚海昆,苑津莎,王瑜,张利伟.多核多分类相关向量机在变压器局部放电模式识别中的应用[J].电工技术学报,2014,29(11):221-228.SHANG Hai-kun,YUAN Jin-sha,WANG Yu,ZHANG Li-wei.Partial discharge pattern recognition in power transformer based on multi-kernel multi-class relevance vector machine[J].Transactions of China Electrotechnical Society,2014,29(11):221-228.(in Chinese)
[9] YIN J L,ZHOU X S,MA Y J,WU Y J,XU X N.Power transformer fault diagnosis based on multi-class multi-kernel learning relevance vector machine[A].2015 IEEE International Conference on Mechatronics and Automation (ICMA)[C].Beijing:IEEE,2015.217-221.
[10] 王庆超,付光远,汪洪桥,辜弘扬,王超.基于局部空间变稀疏约束的多核学习方法[J].电子学报,2018,46(04):930-937.WANG Qing-chao,FU Guang-yuan,WANG Hong-qiao,GU Hong-yang,WANG Chao.Local variable sparsity based multiple kernel learning algorithm[J].Acta Electronica Sinica,2018,46(04):930-937.(in Chinese)
[11] 宋婉莹,李明,张鹏,吴艳,贾璐,刘高峰.基于加权合成核与三重Markov场的极化SAR图像分类方法[J].电子学报,2016,44(03):520-526.SONG Wan-ying,LI Ming,ZHANG Peng,WU Yan,JIA Lu,LIU Gao-feng.A classification method of pol SAR image based on weighted composite kernel and triplet markov field[J].Acta Electronica Sinica,2016,44(03):520-526.(in Chinese)
[12] 李琳,应时,赵翀,董波.基于蚁群算法的面向服务软件的部署优化方法[J].电子学报,2016,44(01):123-129.LI Lin,YING Shi,ZHAO Chong,DONG Bo.Deployment optimization of service-oriented software based on ant colony algorithm[J].Acta Electronica Sinica,2016,44(01):123-129.(in Chinese)
[13] 曹玉莲,李文锋,张煜.基于拟熵自适应启动局部搜索策略的混合粒子群算法[J].电子学报,2018,46(01):110-117.CAO Yu-lian,LI Wen-feng,ZHANG Yu.Hybrid particle swarm optimization algorithm with adaptive starting strategy of local search based on quasi-entropy[J].Acta Electronica Sinica,2018,46(01):110-117.(in Chinese)
[14] 孔德阳,彭华,马金全.基于人工鱼群算法的自适应随机共振方法研究[J].电子学报,2017,45(08):1864-1872.KONG De-yang,PENG Hua,MA Jin-quan.Adaptive stochastic resonance method based on artificial-fish swarm optimization[J].Acta Electronica Sinica,2017,45(08):1864-1872.(in Chinese)
[15] PAN WCH.A new Fruit Fly Optimization Algorithm:Taking the financial distress model as an example[J].Knowledge-Based Systems,2012,26:69-74.
[16] 王欣,杜康,秦斌,等.基于果蝇优化算法的LSSVR干燥速率建模[J].控制工程,2012,19(4):630-638.WANG Xin,DU Kang,QIN Bin,et al.Drying rate modeling based on FOALSSVR[J].Control Engineering of China,2012,19(4):630-638.(in Chinese)
[17] 王雪刚,邹早建.基于果蝇优化算法的支持向量机参数优化在船舶操纵预报中的应用[J].上海交通大学学报,2013,47(06):884-888.WANG Xue-gang,ZOU Zao-jian.FOA-based SVM parameter optimization and its application in ship manoeuvring prediction[J].Journal of Shanghai Jiao Tong University,2013,47(06):884-888.(in Chinese)
[18] 周平,白广忱.基于神经网络与果蝇优化算法的涡轮叶片低循环疲劳寿命健壮性设计[J].航空动力学报,2013,28(5):1013-1018.ZHOU Ping,BAI Guang-chen.Robust design of turbine-blade low cycle fatigue life based on neural networks and fruit fly optimization algorithm[J].Journal of Aero-space Power,2013,28(5):1013-1018.(in Chinese)
[19] HONG Z,LI S,GUO CH J,et al.A hybrid annual power load forecasting model based on generalized regression neural network with fruit fly optimization algorithm[J].Knowledge-Based Systems,2013,37:378-387.
[20] MING W L,JING G,DUAN F H,et al.Ship motion prediction using dynamic seasonal RvSVR with phase space reconstruction and the chaos adaptive efficient FOA[J].Neurocomputing,2016,174:661-680.
[21] 田旭,李杰.一种改进的果蝇优化算法及其在气动优化设计中的应用[J].航空学报,2016,37(06):1-11.TIAN Xu,LI Jie.An improved fruit fly optimization algorithm and its application in aerodynamic optimization design[J].Acta Aeronautica et Astronautica Sinica,2016,37(06):1-11.(in Chinese)
[22] Tipping M E.Sparse Bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001,1(3):211-244.
[23] Kim H S,Eykholt R,Salas J D.Nonlinear dynamics,delay times,and embedding windows[J].Physical D,1999,12(2):48-60.
[24] WANG S,ZHAO L,SU X,et al.Prognostics of lithium-ion batteries based on battery performance analysis and flexible support vector regression[J].Energies,2014,7(10):6492-6508.
[2] 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(09):2034-44.ZHOU Jian-bao,WANG Shao-jun,MA Li-ping,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):2034-44.(in Chinese)
[3] WANG D,MIAO Q,PECHT M.Prognostics of lithium-ion batteries based on relevance vectors and a conditional three-parameter capacity degradation model[J].Journal of Power Sources,2013,239:253-64.
[4] HU Y,LUO P CH.Performance data prognostics based on relevance vector machine and particle filter[J].Chemical Engineering Transactions,2013,33:349-354.
[5] 韩敏,许美玲,任伟杰.多元混沌时间序列的相关状态机预测模型研究[J].自动化学报,2014,40(05):822-829.HAN Min,XU Mei-ling,REN Wei-jie.Reasearch on multivariate chaotic time series prediction using mRSM model[J].Acta Automatic Sinica,2014,40(05):822-829.(in Chinese)
[6] 刘月峰,赵光权,彭喜元.锂离子电池循环寿命的融合预测方法[J].仪器仪表学报,2015,36(07):1462-1469.LIU Yue-feng,ZHAO Guang-quan,PENG Xi-yuan.A fusion prediction method of lithiumion battery cycle-life[J].Chinese Journal of Scientific Instrument,2015,36(07):1462-1469.(in Chinese)
[7] 雷亚国,陈吴,李乃鹏,林京.自适应多核组合相关向量机预测方法及其在机械设备剩余寿命预测中的应用[J].机械工程学报,2016,52(01):87-93.LEI Ya-guo,CHEN Wu,LI Nai-peng,LIN Jing.A relevance vector machine prediction method based on adaptive multi-kernel combination and its application to remaining useful life prediction of machinery[J].Chinese Journal of Mechanical Engineering,2016,52(01):87-93.(in Chinese)
[8] 尚海昆,苑津莎,王瑜,张利伟.多核多分类相关向量机在变压器局部放电模式识别中的应用[J].电工技术学报,2014,29(11):221-228.SHANG Hai-kun,YUAN Jin-sha,WANG Yu,ZHANG Li-wei.Partial discharge pattern recognition in power transformer based on multi-kernel multi-class relevance vector machine[J].Transactions of China Electrotechnical Society,2014,29(11):221-228.(in Chinese)
[9] YIN J L,ZHOU X S,MA Y J,WU Y J,XU X N.Power transformer fault diagnosis based on multi-class multi-kernel learning relevance vector machine[A].2015 IEEE International Conference on Mechatronics and Automation (ICMA)[C].Beijing:IEEE,2015.217-221.
[10] 王庆超,付光远,汪洪桥,辜弘扬,王超.基于局部空间变稀疏约束的多核学习方法[J].电子学报,2018,46(04):930-937.WANG Qing-chao,FU Guang-yuan,WANG Hong-qiao,GU Hong-yang,WANG Chao.Local variable sparsity based multiple kernel learning algorithm[J].Acta Electronica Sinica,2018,46(04):930-937.(in Chinese)
[11] 宋婉莹,李明,张鹏,吴艳,贾璐,刘高峰.基于加权合成核与三重Markov场的极化SAR图像分类方法[J].电子学报,2016,44(03):520-526.SONG Wan-ying,LI Ming,ZHANG Peng,WU Yan,JIA Lu,LIU Gao-feng.A classification method of pol SAR image based on weighted composite kernel and triplet markov field[J].Acta Electronica Sinica,2016,44(03):520-526.(in Chinese)
[12] 李琳,应时,赵翀,董波.基于蚁群算法的面向服务软件的部署优化方法[J].电子学报,2016,44(01):123-129.LI Lin,YING Shi,ZHAO Chong,DONG Bo.Deployment optimization of service-oriented software based on ant colony algorithm[J].Acta Electronica Sinica,2016,44(01):123-129.(in Chinese)
[13] 曹玉莲,李文锋,张煜.基于拟熵自适应启动局部搜索策略的混合粒子群算法[J].电子学报,2018,46(01):110-117.CAO Yu-lian,LI Wen-feng,ZHANG Yu.Hybrid particle swarm optimization algorithm with adaptive starting strategy of local search based on quasi-entropy[J].Acta Electronica Sinica,2018,46(01):110-117.(in Chinese)
[14] 孔德阳,彭华,马金全.基于人工鱼群算法的自适应随机共振方法研究[J].电子学报,2017,45(08):1864-1872.KONG De-yang,PENG Hua,MA Jin-quan.Adaptive stochastic resonance method based on artificial-fish swarm optimization[J].Acta Electronica Sinica,2017,45(08):1864-1872.(in Chinese)
[15] PAN WCH.A new Fruit Fly Optimization Algorithm:Taking the financial distress model as an example[J].Knowledge-Based Systems,2012,26:69-74.
[16] 王欣,杜康,秦斌,等.基于果蝇优化算法的LSSVR干燥速率建模[J].控制工程,2012,19(4):630-638.WANG Xin,DU Kang,QIN Bin,et al.Drying rate modeling based on FOALSSVR[J].Control Engineering of China,2012,19(4):630-638.(in Chinese)
[17] 王雪刚,邹早建.基于果蝇优化算法的支持向量机参数优化在船舶操纵预报中的应用[J].上海交通大学学报,2013,47(06):884-888.WANG Xue-gang,ZOU Zao-jian.FOA-based SVM parameter optimization and its application in ship manoeuvring prediction[J].Journal of Shanghai Jiao Tong University,2013,47(06):884-888.(in Chinese)
[18] 周平,白广忱.基于神经网络与果蝇优化算法的涡轮叶片低循环疲劳寿命健壮性设计[J].航空动力学报,2013,28(5):1013-1018.ZHOU Ping,BAI Guang-chen.Robust design of turbine-blade low cycle fatigue life based on neural networks and fruit fly optimization algorithm[J].Journal of Aero-space Power,2013,28(5):1013-1018.(in Chinese)
[19] HONG Z,LI S,GUO CH J,et al.A hybrid annual power load forecasting model based on generalized regression neural network with fruit fly optimization algorithm[J].Knowledge-Based Systems,2013,37:378-387.
[20] MING W L,JING G,DUAN F H,et al.Ship motion prediction using dynamic seasonal RvSVR with phase space reconstruction and the chaos adaptive efficient FOA[J].Neurocomputing,2016,174:661-680.
[21] 田旭,李杰.一种改进的果蝇优化算法及其在气动优化设计中的应用[J].航空学报,2016,37(06):1-11.TIAN Xu,LI Jie.An improved fruit fly optimization algorithm and its application in aerodynamic optimization design[J].Acta Aeronautica et Astronautica Sinica,2016,37(06):1-11.(in Chinese)
[22] Tipping M E.Sparse Bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001,1(3):211-244.
[23] Kim H S,Eykholt R,Salas J D.Nonlinear dynamics,delay times,and embedding windows[J].Physical D,1999,12(2):48-60.
[24] WANG S,ZHAO L,SU X,et al.Prognostics of lithium-ion batteries based on battery performance analysis and flexible support vector regression[J].Energies,2014,7(10):6492-6508.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |