Scholarlink

<p>Direct fast iterative filtering (dFIF) is a viable method for analyzing vibration signals in rotating machinery, but it faces limitations due to arbitrary parameter settings and imprecise component selection. To overcome these issues, this study proposes an innovative adaptive dFIF (AdFIF) method for rotor rub-impact fault diagnosis. The golden eagle optimization algorithm is employed, combined with the Index of orthogonality-error energy ratio (IO-ER) as the fitness function, to determine the optimal parameters of dFIF. Additionally, a high-resolution Hilbert transform-time frequency spectrum (HT-TFS) is developed. The intrinsic mode functions (IMFs) obtained from AdFIF decomposition are demodulated using the Hilbert transform (HT), and the resulting instantaneous amplitude (IA) and instantaneous frequency (IF) are integrated to obtain an accurate HT-TFS. The approach has been applied to study signals from simulations, experimental, and real engineering cases. The results demonstrate that this technique can successfully extract fault features from the data and outperforms existing methods.</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

Application of AdFIF decomposition-based time-frequency analysis in rotor rub-impact

  • 期刊论文

  • 发表日期：

    2026年5月29日

查看全文
  • Leilei Xiang,
  • Haifan Li,
  • Baojia Chen,
  • Shu Chen

摘要

Direct fast iterative filtering (dFIF) is a viable method for analyzing vibration signals in rotating machinery, but it faces limitations due to arbitrary parameter settings and imprecise component selection. To overcome these issues, this study proposes an innovative adaptive dFIF (AdFIF) method for rotor rub-impact fault diagnosis. The golden eagle optimization algorithm is employed, combined with the Index of orthogonality-error energy ratio (IO-ER) as the fitness function, to determine the optimal parameters of dFIF. Additionally, a high-resolution Hilbert transform-time frequency spectrum (HT-TFS) is developed. The intrinsic mode functions (IMFs) obtained from AdFIF decomposition are demodulated using the Hilbert transform (HT), and the resulting instantaneous amplitude (IA) and instantaneous frequency (IF) are integrated to obtain an accurate HT-TFS. The approach has been applied to study signals from simulations, experimental, and real engineering cases. The results demonstrate that this technique can successfully extract fault features from the data and outperforms existing methods.