欢迎访问机床与液压官方网站!

咨询热线:020-32385312 32385313 RSS EMAIL-ALERT
OVMD与三维奇异谱特征融合的往复压缩机气阀故障识别方法
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

桂林市创新平台和人才计划项目(20220123-23);黑龙江省自然科学基金(E2015037)


Fault Identification Method of Gas Valve for Reciprocating Compressor Based on OVMD and 3D Singular Spectrum Feature Fusion
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对往复压缩机气阀断裂型故障危害下故障振动波形的变异特点,为提高常见的气阀阀片失效后期断裂型故障的识别率,提出一种基于最优变分模态分解(OVMD)和三维奇异谱融合的诊断算法。通过VMD参数优化,利用多重分形去趋势波动分析(MFDFA)提取模态分量的三维奇异谱参数分析,结合核主分量分析降维提取不同工况模态分量的特征值,并建立完整的OVMD_MFDFA融合诊断识别方案。模拟试验和算法对比证实,该法能有效提高环状气阀阀片断裂故障诊断效率和准确性。

    Abstract:

    Aiming at the variation characteristics of fault vibration waveform caused by the fracture fault of gas valve of reciprocating compressor,in order to improve the recognition rate of common late fracture type fault in valve plate,a diagnosis algorithm based on optimal variational modal decomposition (OVMD) and 3D singular spectrum fusion was proposed.Through VMD parameter optimization,multifractal detrended fluctuation analysis (MFDFA) was used to extract 3D singular spectrum parameters of modal components to analyze,eigenvalues of modal components under different working conditions were extracted through dimension reduction combined with kernel principal component analysis,and a complete diagnosis and recognition fusion scheme of OVMD_MFDFA was established.Simulation test and algorithm comparison show that this method can be used to effectively improve the efficiency and accuracy of fault diagnosis of annular gas valve plate crack.

    参考文献
    相似文献
    引证文献
引用本文

刘岩,康丽,苏庆勇,王金东. OVMD与三维奇异谱特征融合的往复压缩机气阀故障识别方法[J].机床与液压,2023,51(9):226-232.
LIU Yan, KANG Li, SU Qingyong, WANG Jindong. Fault Identification Method of Gas Valve for Reciprocating Compressor Based on OVMD and 3D Singular Spectrum Feature Fusion[J]. Machine Tool & Hydraulics,2023,51(9):226-232

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2023-05-29
  • 出版日期: 2023-05-15