Scholarlink

<p>Aiming at the problem that vibration signals of the reciprocating compressor valve contain a lot of interference, making it difficult to effectively extract fault features to realize fault diagnosis, a noise reduction method based on red-billed blue magpie optimizer-successive variational mode decomposition (RBMO-SVMD) and a feature extraction algorithm based on composite multiscale fractional Boltzmann-Shannon interaction entropy (CMFBSIE) are proposed in this paper. Based on a composite fitness function, the SVMD parameter is optimized by RBMO to improve the solution accuracy of SVMD. Then, the signal is processed by the optimized SVMD to achieve noise reduction. By combining BSIE with multi-scale and fractional order, CMFBSIE is constructed, which greatly improves the performance of BSIE and effectively characterize the feature information of signals. The simulation and experiment illustrate that fault features of vibration signals could be effectively extracted to realize accurate fault diagnosis, with an average accuracy of 99.56 %.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A fault feature extraction method for a reciprocating compressor based on optimized SVMD and CMFBSIE

  • 期刊论文

  • 发表日期：

    2025年8月22日

查看全文
  • Xinyue Huang,
  • Fengfeng Bie,
  • Qianqian Li,
  • Wensheng Su,
  • Xueping Ding,
  • Yu Xing,
  • Wenqing Huang

摘要

Aiming at the problem that vibration signals of the reciprocating compressor valve contain a lot of interference, making it difficult to effectively extract fault features to realize fault diagnosis, a noise reduction method based on red-billed blue magpie optimizer-successive variational mode decomposition (RBMO-SVMD) and a feature extraction algorithm based on composite multiscale fractional Boltzmann-Shannon interaction entropy (CMFBSIE) are proposed in this paper. Based on a composite fitness function, the SVMD parameter is optimized by RBMO to improve the solution accuracy of SVMD. Then, the signal is processed by the optimized SVMD to achieve noise reduction. By combining BSIE with multi-scale and fractional order, CMFBSIE is constructed, which greatly improves the performance of BSIE and effectively characterize the feature information of signals. The simulation and experiment illustrate that fault features of vibration signals could be effectively extracted to realize accurate fault diagnosis, with an average accuracy of 99.56 %.