作者：秦波, 王祖达, 孙国栋, 王建国

作者单位：内蒙古科技大学机械工程学院, 内蒙古 包头 014010

关键词：滚动轴承;变分模态分解;自动编码器;极限学习机

摘要：

针对滚动轴承信号的不规则特性，致使信号故障特征难提取及难以辨识，为实现滚动轴承故障的智能诊断，提出基于VMD能量熵与核极限学习机（kernel extreme learning machine，K-ELM）的滚动轴承故障诊断方法。首先将测得振动信号进行变分模态分解（variational mode decomposition，VMD），利用能量熵进一步提取各模态特征组成高维特征向量集；然后将构建的特征向量作为K-ELM算法的输入，通过训练建立K-ELM滚动轴承故障分类模型。实验结果表明：VMD能够很好地分解轴承振动信号，且K-ELM滚动轴承故障分类模型比SVM、ELM故障分类模型具有更高的精度、更强的稳定性。

Application of VMD and hierarchical extreme learning machine in rolling bearing fault diagnosis

QIN Bo, WANG Zuda, SUN Guodong, WANG Jianguo

School of Mechanical Engineering, Inner Mongolia University of Science & Technology, Baotou 014010, China

Abstract: According to the irregularity characteristics of the rolling bearing signals causing the bearing condition identified difficultly, the kernel extreme learning machine fault diagnosis model is proposed. Firstly, the measured vibration signals are decomposed into variational mode decomposition, using the energy entropy to extract the features of each model to form a high dimensional feature vector set. Secondly, the combined feature vector is used as the input of the algorithm, and the fault classification model of the rolling bearing of the hierarchical limit learning machine is established. The experimental results show that the K-ELM rolling bearing fault classification model is better than ELM, and the SVM fault classification model has higher accuracy and stronger stability.

Keywords: rolling bearing;variational mode decomposition;automatic encoder;extreme learning machine

2017, 43(5): 91-95  收稿日期: 2016-08-03;收到修改稿日期: 2016-09-19

基金项目: 国家自然科学基金（51565046）；内蒙古自然科学基金（2017MS0509）；内蒙古科技大学创新基金（2015QDL12）

作者简介: 秦波(1980-),男,河南南阳市人,讲师,硕士,研究方向为复杂工业过程建筑、优化及故障诊断。

参考文献

[1] ZHANG X Y, LIANG Y T, ZHOU J Z, et al. A novel bearing fault diagnosis model integrated permutationentropy, ensemble empirical mode decomposition andoptimized SVM[J].Measurement,2015(69):164-179.
[2] 郑近德,程军圣,杨宇,等. 多尺度排列熵及其在滚动轴承故障诊断中的应用[J]. 中国机械工程,2013,24(19):2641-2646.
[3] 黄海波,李人宪,丁渭平,等. 基于EMD-WVD的车辆悬架减振器异响声品质客观评价研究[J]. 振动与冲击,2015, 34(18):154-160.
[4] 王宏超,陈进,董广明,等. 可调品质因子小波变换在转子早期碰摩故障诊断中应用[J]. 振动与冲击,2014,33(10):77-80.
[5] 马文朋,张俊红,马梁,等. 改进的经验模式分解在机械故障诊断中的应用[J]. 振动、测试与诊断,2015,35(4):637-644.
[6] 武英杰,甄成刚,刘长良. 变分模态分解在风电机组故障诊断中的应用[J]. 机械传动,2015(10):129-132.
[7] WANG Y X, MARKERT R, XIANG J W, et al. Research on variational mode decomposition and its application in detecting rub-impact fault ofthe otor system[J]. Mechanical Systems and Signal Processing,2015(60-61):243-251.
[8] 王斌. 基于小波分析和BP神经网络的滚动轴承故障诊断方法研究[D]. 兰州:兰州交通大学,2014.
[9] 向丹,岑健. 基于EMD熵特征融合的滚动轴承故障诊断方法[J]. 航空动力学报,2015,30(5):1149-1155.
[10] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machineaheory and applications[J]. Neurocom-puling,2006(70):489-501.
[11] 陈立军,孙凯,侯媛媛. 基于极限学习机的汽轮机故障诊断[J]. 化工自动化及仪表,2013,40(4):435-438.
[12] 周东凯,李刚,王学馄. 基于PSO-ELM的双目视觉摄像机标定[J]. 广西大学学报(自然科学版),2014,39(6):1285-1290.
[13] 朱志洁,张宏伟. 基于GA-ELM的冲击地压危险性预测研究[J]. 中国安全生产科学技术,2014,10(8):46-51.
[14] DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing,2014,62(3):531-544.
[15] LIU H, SONG B, QIN H. An adaptive-ADMM algorithm with supportand signal value detection for compressed sensing[J]. IEEE Signal Processing Letters,2013, 20(4):315-318.
[16] WIDROW B, GREENBLATT A, KIM Y, et al. The No-Prop algorithm:a new learning algorithm for multilayer neural networks[J]. Neural Networks,2013(37):182-188.