发电技术 ›› 2018, Vol. 39 ›› Issue (1): 63-69.DOI: 10.12096/j.2096-4528.pgt.2018.011

• 新能源 • 上一篇    下一篇

基于VMD与不同包络阶次构造的风电机组滚动轴承故障诊断

王建国1(),林语桐1,田野2,杜鹏3,张培焱4,辛红伟1,武英杰1   

  1. 1 东北电力大学自动化工程学院, 吉林省 吉林市 132012
    2 吉林中电投新能源有限公司, 吉林省 长春市 130012
    3 长岭中电投第二风力发电有限公司, 吉林省 松原市 131500
    4 国网襄阳市襄州区供电公司, 湖北省 襄阳市 441100
  • 收稿日期:2017-12-11 出版日期:2018-02-28 发布日期:2018-07-27
  • 作者简介:王建国(1963),男,博士,教授,博士生导师,研究方向为旋转机械故障诊断、电站特殊参数测量等, 397625835@qq.com
  • 基金资助:
    吉林省科技发展计划重大科技招标专项(20150203001SF)

Fault Diagnosis of Rolling Bearing of Wind Turbine Based on VMD and Different Envelope Order Structure

Jianguo WANG1(),Yutong LIN1,Ye TIAN2,Peng DU3,Peiyan ZHANG4,Hongwei XIN1,Yingjie WU1   

  1. 1 School of Electrical Engineering and Automation, Northeast Electric Power University, Jilin 132012, Jilin Province, China
    2 Jilin CPI New Energy Co., Ltd., Changchun 130012, Jilin Province, China
    3 Changling Zhongdiantou Second Wind Power Co. Ltd., Songyuan 131500, Jilin Province, China
    4 State Grid Xiangyang City Xiangzhou District Power Supply Company, Xiangyang 441100, Hubei Province, China
  • Received:2017-12-11 Published:2018-02-28 Online:2018-07-27
  • Supported by:
    Science & Technology Development Plan of Jilin Province(20150203001SF)

摘要:

针对风电机组变工况滚动轴承故障诊断问题,将变分模态分解(variational mode decomposition,VMD)与计算阶比跟踪(computed order tracking,COT)、逆包络阶次谱(reversed sequence squared envelope spectrum,RE-SES)2种方法结合进行包络阶次分析;将非平稳的时间域信号转换成平稳的角度域信号,再经包络法找出滚动轴承故障特征信息。仿真与实验结果表明,VMD结合RE-SES包络阶次法可以更加有效地提取出调幅信号中的调制阶次,并且VMD滤波可以使得故障特征阶次更加凸显,易于故障识别,为风电机组变速工况下的滚动轴承故障诊断提供参考。

关键词: 风电机组, 非平稳信号, 计算阶比跟踪(COT), 包络阶次, 变分模态分解(VMD), 逆包络阶次谱(RE-SES), 轴承故障诊断

Abstract:

In order to identify the faults of rolling bearing of wind turbines, the variational modal decomposition (VMD) was combined with two methods, including calculation order tracking (COT) and reversed sequence squared envelope spectrum (RE-SES), to carry out the envelope order analysis. The non-stationary signals in time domain were converted into stationary signals in angle domain by equal angle sampling. And the bearing fault characteristic information was obtained by envelope method. The results show that the modulation order of amplitude modulation signal can be extracted more effectively by VMD combined with RE-SES envelope order method. The VMD filter could make the fault feature order more prominent, and it is easy to identify the fault, and it provide reference for the fault diagnosis of rolling bearing of wind turbine under variable speed condition.

Key words: wind turbine generator, non-stationary signal, computed order tracking (COT), the envelope order, variational mode decomposition (VMD), reversed sequence squared envelope spectrum (RE-SES), bearing fault diagnosis