基于多尺度排列熵与双核极限学习机的滚动轴承故障诊断方法
|
摘要
针对极限学习机隐含层节点数需人为设定,分类的准确性与稳定性较差,核极限学习机(K-ELM)对核函数选取要求较高,单一核函数难以对非线性样本充分学习、泛化性仍有不足等缺点,提出一种基于多尺度排列熵(MPE)和非线性加权组合的双核极限学习机(DK-ELM)的滚动轴承故障诊断方法并证明了其可行性与优越性。首先,计算不同故障状态轴承信号的多尺度排列熵,获取一系列无量纲特征;然后,利用双核函数计算其高维特征向量集并输入DK-ELM中建立轴承信号状态分类模型,对不同状态的轴承信号进行分类。实验结果证明,核函数的引入可以有效提高ELM分类性能,DK-ELM的分类模型比支持向量机(SVM)、ELM以及各单核极限学习机具有更高的分类精度,而且对训练样本数量较少的情况有更好的分类效果。
The number of hidden layer nodes of the extreme learning machine needs to be set by artificial,and the accuracy and stability are not good,The kernel extreme learning machine( K-ELM) has high requirements for the selection of kernel functions. The single kernel function cannot fully learn the non-linear feature,and the generalization is still insufficient. A fault diagnosis method of rolling bearing based on multi-scale permutation entropy( MPE) and nonlinear weighted-dual kernel extreme learning machine( DK-ELM) is proposed and its feasibility and superiority are proved. First,the MPE of bearing signal in different fault states is extracted. Next,the high dimension feature vector is expressed by dual kernel and used as the input of the DK-ELM to establish the classification model to determine the state of signal. The result of experimental shows that the addition of kernels can improve the performance of ELM,classification model of DK-ELM has higher classification accuracy than sopport vector machine( SVM),ELM and single-kernel ELM,and has better classification effect on the insufficient number of samples.
The number of hidden layer nodes of the extreme learning machine needs to be set by artificial,and the accuracy and stability are not good,The kernel extreme learning machine( K-ELM) has high requirements for the selection of kernel functions. The single kernel function cannot fully learn the non-linear feature,and the generalization is still insufficient. A fault diagnosis method of rolling bearing based on multi-scale permutation entropy( MPE) and nonlinear weighted-dual kernel extreme learning machine( DK-ELM) is proposed and its feasibility and superiority are proved. First,the MPE of bearing signal in different fault states is extracted. Next,the high dimension feature vector is expressed by dual kernel and used as the input of the DK-ELM to establish the classification model to determine the state of signal. The result of experimental shows that the addition of kernels can improve the performance of ELM,classification model of DK-ELM has higher classification accuracy than sopport vector machine( SVM),ELM and single-kernel ELM,and has better classification effect on the insufficient number of samples.
引文
[1]赵元喜,胥永刚,高立新,等.基于谐波小波包和BP神经网络的滚动轴承声发射故障模式识别技术[J].振动与冲击,2010,29(10):162-165,257.ZHAO Y,XU Y G,GAO L X,et al.Fault pattern recognition technology of rolling bearing acoustic emission based on harmonic wavelet packet and BP neural network[J].Journal of Vibration and Shock,2010,29(10):162-165,257.
[2]钟小倩,马文科,宋萌萌.基于GA和LM组合优化BP神经网络的滚动轴承故障诊断方法[J].组合机床与自动化加工技术,2014(12):91-95.ZHONG X Q,MA W K,SONG M M.Fault diagnosis method of rolling bearing based on GA and LMcombination optimization BP neural network[J].Modular Machine Tool&Automatic Manufacturing Technique,2014(12):91-95.
[3]CAI M.Global optimization algorithm for BP neural networks[J].Engineering Journal of Wuhan University,2013,46(6):794-810.
[4]张达敏,林辉品,林智勇,等.基于神经网络预测控制的节能电梯能量管理[J].仪器仪表学报,2017,38(12):3137-3142.ZHANG D M,LIN H P,LIN Z Y,et al.Energy-saving elevator energy management based on neural network predictive control[J].Journal of Instruments and Instruments,2017,38(12):3137-3142.
[5]付大鹏,翟勇,于青民.基于EMD和支持向量机的滚动轴承故障诊断研究[J].机床与液压,2017,45(11):184-187.FU D P,ZHAI Y,YU Q M.Fault diagnosis of rolling bearing based on EMD and support vector machine[J].Machine Tool&Hydraulics,2017,45(11):184-187.
[6]汤晓全.基于小波包与极限学习机的滚动轴承故障诊断方法研究[D].重庆:重庆大学,2017.TANG X Q.Research on rolling bearing fault diagnosis method based on wavelet packet and extreme learning machine[D].Chongqing:Chongqing University,2017.
[7]秦波,孙国栋,陈帅,等.排列熵与核极限学习机在滚动轴承故障诊断中的应用[J].组合机床与自动化加工技术,2017(2):73-76.QIN B,SUN G D,CHEN S,et al.Application of permutation entropy and nuclear extreme learning machine in rolling bearing fault diagnosis[J].Modular Machine Tool&Automatic Manufacturing Technique,2017(2):73-76.
[8]宋坤骏,丁建明,林建辉.基于改进Fisher准则、VMD、距离相关系数和核极限学习机的轴承故障诊断[J].铁道机车车辆,2018,38(3):22-28.SONG K J,DING J M,LIN J H.Bearing fault diagnosis based on improved Fisher criterion,VMD,distance correlation coefficient and nuclear limit learning machine[J].Railway Locomotive&Car,2018,38(3):22-28.
[9]秦波,王祖达,孙国栋,等.VMD能量熵与核极限学习机在滚动轴承故障诊断中的应用[J].中国测试,2017,43(5):91-95.QIN B,WANG Z D,SUN G D,et al.Application of VMD energy entropy and nuclear extreme learning machine in rolling bearing fault diagnosis[J].China Measurement&Test,2017,43(5):91-95.
[10]刘美容,曾黎,何怡刚,等.基于LMD多尺度熵和极限学习机的模拟电路故障诊断[J].电子测量与仪器学报,2017,31(4):530-536.LIU M R,ZENG L,HE Y G,et al.Analog circuit fault diagnosis based on LMD multi-scale entropy and extreme learning machine[J].Journal of Electronic Measurement and Instruments,2017,31(4):530-536.
[11]徐存知,熊新.基于广义形态差值滤波与极限学习机的滚动轴承故障诊断方法研究[J].化工自动化及仪表,2019,46(1):54-57.XU C ZH,XIONG X.Research on fault diagnosis method of rolling bearing based on generalized morphological difference filter and limit learning machine[J].Chemical automation and instrumentation,2019,46(1):54-57.
[12]LI D,LI X L,LIANG Z H,et al.Multiscale permutation entropy analysis of eeg recordings during sevoflurance anesthesia[J].Journal of Neural Engineering,2010,7(4):046010-1-046010-14.
[13]瞿金秀,石长全,丁锋,等.基于多尺度排列熵和支持向量机的轴承故障诊断[J].煤矿机械,2018,39(9):143-146.QU J X,SHI CH Q,DING F,et al.Bearing fault diagnosis based on multi-scale permutation entropy and support vector machine[J].Coal Mine Machinery,2018,39(9):143-146.
[14]GE J,ZHANG G P,Novel images extraction model using improved delay vector variance feature extraction and multi-kernel neural network for EEG detection and prediction[J],Technology and Health Care,2015,2(S1):51-55.
[15]LIU X,WANG L,HUANG G B,et al.Multiple kernel extreme learning machine[J].Neurocomputing,2013,149(PA):253-264.
[16]CONG L,CHI M P,CHI M V,et al.Efficient shape classification using region descriptors[J].Multimedia Tools and Applications,2017,76(1):83-102.
[2]钟小倩,马文科,宋萌萌.基于GA和LM组合优化BP神经网络的滚动轴承故障诊断方法[J].组合机床与自动化加工技术,2014(12):91-95.ZHONG X Q,MA W K,SONG M M.Fault diagnosis method of rolling bearing based on GA and LMcombination optimization BP neural network[J].Modular Machine Tool&Automatic Manufacturing Technique,2014(12):91-95.
[3]CAI M.Global optimization algorithm for BP neural networks[J].Engineering Journal of Wuhan University,2013,46(6):794-810.
[4]张达敏,林辉品,林智勇,等.基于神经网络预测控制的节能电梯能量管理[J].仪器仪表学报,2017,38(12):3137-3142.ZHANG D M,LIN H P,LIN Z Y,et al.Energy-saving elevator energy management based on neural network predictive control[J].Journal of Instruments and Instruments,2017,38(12):3137-3142.
[5]付大鹏,翟勇,于青民.基于EMD和支持向量机的滚动轴承故障诊断研究[J].机床与液压,2017,45(11):184-187.FU D P,ZHAI Y,YU Q M.Fault diagnosis of rolling bearing based on EMD and support vector machine[J].Machine Tool&Hydraulics,2017,45(11):184-187.
[6]汤晓全.基于小波包与极限学习机的滚动轴承故障诊断方法研究[D].重庆:重庆大学,2017.TANG X Q.Research on rolling bearing fault diagnosis method based on wavelet packet and extreme learning machine[D].Chongqing:Chongqing University,2017.
[7]秦波,孙国栋,陈帅,等.排列熵与核极限学习机在滚动轴承故障诊断中的应用[J].组合机床与自动化加工技术,2017(2):73-76.QIN B,SUN G D,CHEN S,et al.Application of permutation entropy and nuclear extreme learning machine in rolling bearing fault diagnosis[J].Modular Machine Tool&Automatic Manufacturing Technique,2017(2):73-76.
[8]宋坤骏,丁建明,林建辉.基于改进Fisher准则、VMD、距离相关系数和核极限学习机的轴承故障诊断[J].铁道机车车辆,2018,38(3):22-28.SONG K J,DING J M,LIN J H.Bearing fault diagnosis based on improved Fisher criterion,VMD,distance correlation coefficient and nuclear limit learning machine[J].Railway Locomotive&Car,2018,38(3):22-28.
[9]秦波,王祖达,孙国栋,等.VMD能量熵与核极限学习机在滚动轴承故障诊断中的应用[J].中国测试,2017,43(5):91-95.QIN B,WANG Z D,SUN G D,et al.Application of VMD energy entropy and nuclear extreme learning machine in rolling bearing fault diagnosis[J].China Measurement&Test,2017,43(5):91-95.
[10]刘美容,曾黎,何怡刚,等.基于LMD多尺度熵和极限学习机的模拟电路故障诊断[J].电子测量与仪器学报,2017,31(4):530-536.LIU M R,ZENG L,HE Y G,et al.Analog circuit fault diagnosis based on LMD multi-scale entropy and extreme learning machine[J].Journal of Electronic Measurement and Instruments,2017,31(4):530-536.
[11]徐存知,熊新.基于广义形态差值滤波与极限学习机的滚动轴承故障诊断方法研究[J].化工自动化及仪表,2019,46(1):54-57.XU C ZH,XIONG X.Research on fault diagnosis method of rolling bearing based on generalized morphological difference filter and limit learning machine[J].Chemical automation and instrumentation,2019,46(1):54-57.
[12]LI D,LI X L,LIANG Z H,et al.Multiscale permutation entropy analysis of eeg recordings during sevoflurance anesthesia[J].Journal of Neural Engineering,2010,7(4):046010-1-046010-14.
[13]瞿金秀,石长全,丁锋,等.基于多尺度排列熵和支持向量机的轴承故障诊断[J].煤矿机械,2018,39(9):143-146.QU J X,SHI CH Q,DING F,et al.Bearing fault diagnosis based on multi-scale permutation entropy and support vector machine[J].Coal Mine Machinery,2018,39(9):143-146.
[14]GE J,ZHANG G P,Novel images extraction model using improved delay vector variance feature extraction and multi-kernel neural network for EEG detection and prediction[J],Technology and Health Care,2015,2(S1):51-55.
[15]LIU X,WANG L,HUANG G B,et al.Multiple kernel extreme learning machine[J].Neurocomputing,2013,149(PA):253-264.
[16]CONG L,CHI M P,CHI M V,et al.Efficient shape classification using region descriptors[J].Multimedia Tools and Applications,2017,76(1):83-102.