Scholarlink

To address the nonlinear and non-stationary characteristics of rolling bearing vibration signals, we propose an integrated approach combining multi-feature extraction with intelligent diagnosis. The method first applies PSO-optimized variational mode decomposition (PSO-VMD) to decompose raw signals, then constructs feature vectors using entropy measures (sample entropy, fuzzy entropy, envelope entropy, energy entropy, and multi-scale permutation entropy) of the resulting intrinsic mode functions (IMFs). Pearson correlation analysis reduces feature dimensionality before input into an improved whale optimization algorithm (IWOA)-enhanced support vector machine (SVM) model. Experimental results demonstrate 99.667% classification accuracy, confirming the method's effectiveness and superiority.

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

Rolling Bearing Fault Diagnosis Based on Multi-feature Extraction and IWOA-SVM

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Hui Zhang,
  • Xinyue Zhou,
  • Chisen Qin,
  • Han Yin,
  • Tianqi Jiang,
  • Licheng Han,
  • Bin Liu

摘要

To address the nonlinear and non-stationary characteristics of rolling bearing vibration signals, we propose an integrated approach combining multi-feature extraction with intelligent diagnosis. The method first applies PSO-optimized variational mode decomposition (PSO-VMD) to decompose raw signals, then constructs feature vectors using entropy measures (sample entropy, fuzzy entropy, envelope entropy, energy entropy, and multi-scale permutation entropy) of the resulting intrinsic mode functions (IMFs). Pearson correlation analysis reduces feature dimensionality before input into an improved whale optimization algorithm (IWOA)-enhanced support vector machine (SVM) model. Experimental results demonstrate 99.667% classification accuracy, confirming the method's effectiveness and superiority.