基于ITD能量特征与K-ELM的滚动轴承故障诊断方法研究
|
摘要
针对滚动轴承故障诊断中故障特征难提取与极限学习机稳定性、泛化能力差,致使故障辨识精度差的问题,提出了一种基于ITD(Intrinsic Time-scale Decomposition)能量特征与KELM的滚动轴承故障诊断方法。首先将采集到的滚动轴承故障信号进行ITD分析得到一系列蕴含有信号瞬时频率的固有旋转分量;然后,提取分解后与原信号相关程度较大固有旋转分量的能量特征;最后,建立核极限学习机的滚动轴承故障分类模型,并将所得能量特征向量矩阵作为K-ELM模型的输入进行故障模式辨识。试验结果表明:与基于ITD-SVM,ITD-ELM,EMD-K-ELM的故障诊断方法相比,ITD-K-ELM方法具有更高的分类精度,能更好地应用于滚动轴承的故障诊断。
In order to solve the problems that the characteristics is difficult to extract and extreme learning machine' poor stability and generalization ability,which causes fault identification precision is bad,puts forward a rolling bearing fault diagnosis method based on ITD energy feature and K-ELM.First,it analyzes the collected bearing fault signals using ITD.Then,this method extracts energy characteristics from the proper rotation components that are largest related to the original signal.Finally,the energy eigenvector matrix is inputted into K-ELM to recognize the fault pattern. The test results show this method has good classification accuracy for rolling bearing status than the methods of ITD-SVM,ITD-ELM,EMD-K-ELM.
In order to solve the problems that the characteristics is difficult to extract and extreme learning machine' poor stability and generalization ability,which causes fault identification precision is bad,puts forward a rolling bearing fault diagnosis method based on ITD energy feature and K-ELM.First,it analyzes the collected bearing fault signals using ITD.Then,this method extracts energy characteristics from the proper rotation components that are largest related to the original signal.Finally,the energy eigenvector matrix is inputted into K-ELM to recognize the fault pattern. The test results show this method has good classification accuracy for rolling bearing status than the methods of ITD-SVM,ITD-ELM,EMD-K-ELM.
引文
[1]王天金,冯志鹏,郝如江,等.基于Teager能量算子的滚动轴承故障诊断研究[J].振动与冲击,2012,31(2):l-5.
[2]张弦,王宏力.进化小波消噪方法及其在滚动轴承故障诊断中的应用[J].机械工程学报,2010,46(15):76-81.
[3]Jiang Ling-li,Liu Yi-lun,Li Xue-jun.Gear fault diagnosis based on SVM and multi-sensor information fusion[J].Journal of Central South University(Science and Technology),2010,41(6):2 184-2 188.
[4]徐卓飞,张海燕.基于EMD与KPCA的滚动轴承故障特征提取及诊断方法研究[J].机械科学与技术,2014,33(10):1 518-1 524.
[5]Yu Yang,Dejie Yu,Junsheng Cheng.A fault diagnosis approach for roller bearing based on IMF envelope spectrum and SVM[J].Measurement,2007(40):943-950.
[6]程军圣,李海龙,杨宇.改进ITD和能量矩在齿轮故障诊断中的应用[J].振动、测试与诊断,2013,33(6):954-958.
[7]柏林,陆超,赵鑫.基于ITD与ICA的滚动轴承故障特征提取方法[J].振动与冲击,2015,34(14):153-156.
[8]安学利,蒋东翔.基于ITD和LS-SVM的风力发电机组轴承故障诊断[J].电力自动化设备,2011,31(9):10-13.
[9]尹刚,张英堂,李志宁.运用在线贯序极限学习机的故障诊断方法[J].振动、测试与诊断,2013,33(2):325-329.
[10]秦波,刘永亮,王建国,等.基于在线贯序极限学习机的齿轮故障诊断方法[J].机械设计与制造,2016,(12):101-104.
[11]苑津莎,张利伟,王瑜.基于极限学习机的变压器故障诊断方法研究[J].电测与仪表,2013,50(12):21-26.)
[2]张弦,王宏力.进化小波消噪方法及其在滚动轴承故障诊断中的应用[J].机械工程学报,2010,46(15):76-81.
[3]Jiang Ling-li,Liu Yi-lun,Li Xue-jun.Gear fault diagnosis based on SVM and multi-sensor information fusion[J].Journal of Central South University(Science and Technology),2010,41(6):2 184-2 188.
[4]徐卓飞,张海燕.基于EMD与KPCA的滚动轴承故障特征提取及诊断方法研究[J].机械科学与技术,2014,33(10):1 518-1 524.
[5]Yu Yang,Dejie Yu,Junsheng Cheng.A fault diagnosis approach for roller bearing based on IMF envelope spectrum and SVM[J].Measurement,2007(40):943-950.
[6]程军圣,李海龙,杨宇.改进ITD和能量矩在齿轮故障诊断中的应用[J].振动、测试与诊断,2013,33(6):954-958.
[7]柏林,陆超,赵鑫.基于ITD与ICA的滚动轴承故障特征提取方法[J].振动与冲击,2015,34(14):153-156.
[8]安学利,蒋东翔.基于ITD和LS-SVM的风力发电机组轴承故障诊断[J].电力自动化设备,2011,31(9):10-13.
[9]尹刚,张英堂,李志宁.运用在线贯序极限学习机的故障诊断方法[J].振动、测试与诊断,2013,33(2):325-329.
[10]秦波,刘永亮,王建国,等.基于在线贯序极限学习机的齿轮故障诊断方法[J].机械设计与制造,2016,(12):101-104.
[11]苑津莎,张利伟,王瑜.基于极限学习机的变压器故障诊断方法研究[J].电测与仪表,2013,50(12):21-26.)