第38卷第3期 2017年3月
自动化仪表
PROCESS AUTOMATION INSTRUMENTATION
基于全矢Hilbert时域边际谱的轴承诊断
王宏涛
（洛阳轴研科技股份有限公司，河南洛阳471039）
Vol. 38 No.3 Mar.2017
搞要：为实现有效的故障诊断，提出一种基于全失希尔伯特（Hilbert)时域边际谱的轴承诊断方法，对使用中的滚动轴承进行故障诊断及预警。通过采集运转中滚动轴承的同源双通道振动信号，采用经验模态分解（EMD）和Hilbert变换对同源双通道的振动信号进行处理，把非平稳报动信号分解为一系列固有模态函数（IMF)和信号的Hilbert谱：再对Hilbert谱进行积分，得到能够体现信号时域特征的时域边际谱；结合全失谱技术对其进行同源双通道的信息融合，最终得到滚动轴承振动信号的全失HHT时域边际谱。该方法可以避免由于传统单通道信息采集可能发生的信息遗漏而导致的对故障诊断的漏判和误判，从而提高诊断的准确性和可靠性。在国家轴承质量监督检验中心滚动轴承全周期疫劳寿命试验机上，对滚动轴承外图故障进行试验。结果表明，与传统的Hilbert边际谱相
比，该方法能够更有效地识别外圈剥落、内固剥落等局部损伤类故障的部位和类型，可以在工程应用上实践、推广。关键词：HHT，EMD；Hitbent变换；滚动承；故障诊断
中图分类号：TH133；TP2
文献标志码：A
D0I; 10. 16086/j. enki. issn1000 0380. 201703016
BearingDiagnosisBasedonHilbertMarginalSpectrum
TimeDomainFullVector
WANG Hongtao
(Laoyang Bearing Science & Technology Co. ,Lad. ,Luoyang 471039, China)
Abstract: A bearing diagnosis method based on the Hilbert time domain marginal spectrum is proposed for fault diagnosis and early warning of the rolling bearing in use, Through collecting the homologous dual channel vibration signals of the running rolling bearing;and processing these signals by using empirical mode decomposition ( EMD) and Hilbert transform, the nonstationary vibration signals are decomposed into a series of intrinsic mode functions ( IMF) , and Hilbert spectrum of the signal; then the Hilbert spectrum is integrated to obtain the time domain marginal spectrum which can reflect the time domain eharacteristics of the signal, The infomation fusion of the homologous dual channel is carried out by using the full vector spectrum technique, and
the full vector Hilbert - Huang transfom ( HHT) time domain m
pectrum of the vibration signal of rolling bearing is
obtained finally. It is possible to avoid the missing and misjudgment of the fault diagnosis due to information missing caused by single channel information acquisition,thus the correctness and reliability of the diagnosis can be enhanced. The failure tests of the outer ring of the rolling bearing carry on the full cycle fatigue life testing machine of the National Bearing Quality Supervision and Inspection Center. The results show that the method is more effective than the traditional Hilbert marginal spectrum in recognizing the position and types of the local damage class failures stch as outer ring peeling and inner ring peeling;the method can be promoted in engineering applications.
Keywords: Hilbert Huang transform ( HHT) ; Empirical mode decomposition ( EMD) ; Hilbert transform; Rolling bearing: Fault diagnosis
0
引言
滚动轴承是各型旋转机械设备的关键组成部件，
被誉为机械设备的关节，是应用广泛的关键零部件。据权威机构统计，滚动轴承故障约占机械设备各类故障总和的30%，内、外圈剥落的局部损伤类故障达到滚动轴承故障总数的90%")。振动信号一直是衡量
轴承综合性的重要指标(2)。由于机械设备的工作转速不稳定、负载不断变化以及各零部件之间的相互冲击、摩擦等因素，导致传感器采集到的滚动轴承的振动信号往往是典型的非平稳(3)、非线性信号。特别是对于轴承的早期故障，其振动信号的故障成分非常微弱，常常被其他零件的工作振动和背景噪声信号没，很难从幅值谱上发现这些故障特征，因而不能实施有效的
修改稿收到日期：2016-06-30
作者简介：王宏涛（1969—），男，学士，工程师，主要从事轴承检测仪器及诊断预警技术的研究。E-mail：ewt2000@163.eom。万方数据