Scholarlink

In order to achieve high-precision detection of bearings under different working conditions, this paper proposes a bearing fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and probability density analysis. The time-domain and frequency-domain characteristics of different bearing fault signals are analyzed. The kurtosis factor is selected to characterize bearing condition. The original vibration signal is decomposed into multiple intrinsic mode functions (IMF) by the EEMD algorithm, and each IMF component contains local feature information of the original signal at different time scales. The time-domain fault characteristics are extracted by calculating the kurtosis factor of each IMF component and doing probability density analysis. The frequency-domain fault characteristics are extracted by performing Fourier Transform and Hilbert transform on IMF. This paper uses the Case Western Reserve University (CWRU) bearing dataset for simulation analysis, as well as field test data for validation. The experimental results show that the diagnostic method proposed in this paper achieves 99.2% accuracy for field test data collected by the acceleration sensor.

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

Bearing Faults Diagnosis Based on EEMD and Probability Density Analysis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Xiangyang Liu,
  • Zhifeng Li,
  • Shuanglong He,
  • Zhangliang Xu

摘要

In order to achieve high-precision detection of bearings under different working conditions, this paper proposes a bearing fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and probability density analysis. The time-domain and frequency-domain characteristics of different bearing fault signals are analyzed. The kurtosis factor is selected to characterize bearing condition. The original vibration signal is decomposed into multiple intrinsic mode functions (IMF) by the EEMD algorithm, and each IMF component contains local feature information of the original signal at different time scales. The time-domain fault characteristics are extracted by calculating the kurtosis factor of each IMF component and doing probability density analysis. The frequency-domain fault characteristics are extracted by performing Fourier Transform and Hilbert transform on IMF. This paper uses the Case Western Reserve University (CWRU) bearing dataset for simulation analysis, as well as field test data for validation. The experimental results show that the diagnostic method proposed in this paper achieves 99.2% accuracy for field test data collected by the acceleration sensor.