
ISSN10003762 CN41 1148/TH
测量与仪器
轴承2016年7期 Bearing 2016, No.7
基于变分模态分解的风机滚动轴承早期
故障诊断
郑小霞,周国旺",任浩翰,符杨
(1.上海电力学院
自动化工程学院,上海200090;2.上海东海风力发电有限公司,上海200090)
4853
摘要:针对海上风电机组滚动轴承故障多且早期故障特征难以提取的问题,提出了一种基于变分模态分解的滚动轴承故障诊断方法。从额率方面研究了模态分量个数对信号特征信息的影响,提出故障特征信息提取时确定分解个数的一般原则,据此确定滚动轴承早期故障振动信号的分解个数并得到若干模态分量,从中筛选出最
佳模态分量进行包络解调分析,最终通过对比包络谱特征频率对滚动轴承进行早期故障诊断。关键词:滚动轴承;故障诊断;变分模态分解;海上风电;模态确定
中图分类号:TH133.33;TP206*.3
文献标志码:B
文章编号:10003762(2016)07004806
Incipient Fault Diagnosis for Rolling Bearings Used in Wind Turbine
Based on Variational Mode Decomposition Zheng Xiaoxia',Zhou Guowang',Ren Haohan’,FuYang
(1. School of Automation Engineering, Shanghai University of Electrie Power,Shanghai 200090, China;2. Shanghai
Donghai Wind Power Co. , Lad. , Shanghai 200090, China)
Abstract: There are many faults occuring in the rolling bearings for offshore wind turbines and the incipient fault fea-ture is difficult to extract. A fault diagnosis method for rolling bearings is proposed based on variational mode decompo-sition( VMD) for extracting fault feature. The effect of number of mode components on characteristie infomation of sig-nal is studied from the aspects of frequency, and the general principles are proposed to determine decomposition number during extraction of fault feature information. The decomposition number of incipient fault vibration signal of rolling bearings is determined and the several mode components are obtained. The best mode component is selected, and the envelope demodulation analysis is carried out, Finally, the incipient fault diagnosis for rolling bearings is carried out by comparison of characteristic frequency of envelope spectrum.
Key words: rolling bearing; fault diagnosis; variational mode decomposition; offshore wind turbine; mode determina-tion
滚动轴承是风电机组传动系统的关键部件,海上风电机组的工作环境极为恶劣,腐蚀性气体收稿日期:2015-1229修回日期:2016-0222
基金项目:国家自然科学基金项目(51507098);上海绿色能源并网工程技术研究中心项目(13DZ2251900);上海市科委重点科技攻关项目(14DZ1200905);上海市电站自动化技术重点实验室项目(04DZ05901)
作者简介:郑小霞(1978一),女,山东烟台人,博士,副教授,研究方向为风力发电故障诊断与运行维护,E-mail:
zxxkx@126.com。万方数据
对轴承的影响比陆地上更加严重。因此,对海上风电机组滚动轴承进行早期故障诊断是保证风电机组正常运行的重要手段。
振动信号分析是一种常见的机械系统故障诊断技术,其中的小波变换"-4]及经验模态分解(EMD)[6-8)在机械故障信号提取和诊断方面得到了广泛的应用,但这些理论和方法还需要进一步完善,如小波分析中小波基和滤波阀值的选取问题;EMD的端点效应和模态混叠间题等。
变分模态分解(VariationalModeDecomposi-