Scholarlink

<p>Owing to the complexity and diversity of excitation sources, accurately extracting vibration characteristic information from pumping stations is difficult. Therefore, a novel feature information extraction method based on the wavelet threshold and improved variational mode decomposition (WT-IVMD) is proposed in this study. First, the wavelet threshold denoising method is used to preprocess the original signal to suppress white noise interference and reduce the modal aliasing problem. Second, four indices are proposed to optimize the modal number <i>K</i> of the VMD, including the energy difference parameter <i>e</i>, the dispersion entropy <i>D</i><sub>e</sub>, the ratio of the frequency domain maximum value to the submaximum value <i>R</i>, and the dominant frequency interval <i>D</i><sub>f</sub> of adjacent modes, which greatly preserves structurally effective feature information from vibration responses with multisource excitation and severe noise interference. The feasibility and accuracy of the proposed method are verified via simulation signals using the signal-to-noise ratio (SNR) and root mean square error (RSME). The calculated results are compared with those of empirical mode decomposition (EMD), variational mode decomposition (VMD), and wavelet threshold-EMD (WT-EMD). The results show that the SNR is increased by 78.7%, 20.9%, and 17.2%, respectively, and the RMSE is reduced by 25%, 11.1%, and 9.5%, respectively, indicating that the proposed method can accurately identify and effectively separate the frequencies of different excitation sources for pumping stations. Finally, the improved method is applied to the Dayuzhang Pumping Station during a specified operation period, and the dominant frequencies of the pumping station plant structure are 16.68&#xa0;Hz, 29.19&#xa0;Hz, 33.33&#xa0;Hz, 58.37&#xa0;Hz, 66.72&#xa0;Hz, 83.40&#xa0;Hz, and 116.74&#xa0;Hz. The engineering case indicates that the effective vibration characteristic information of a pumping station plant can be identified via the proposed method under multisource excitation, which provides theoretical support for vibration source identification and operation state monitoring of pumping station plant structures.</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 novel vibration feature information identification of pumping station based on WT-IVMD

  • 期刊论文

  • 发表日期：

    2025年5月22日

查看全文
  • Qi Jiang,
  • Qianyi Zhang,
  • Jianwei Zhang,
  • Yu Zhao,
  • Xizhu Liu

摘要

Owing to the complexity and diversity of excitation sources, accurately extracting vibration characteristic information from pumping stations is difficult. Therefore, a novel feature information extraction method based on the wavelet threshold and improved variational mode decomposition (WT-IVMD) is proposed in this study. First, the wavelet threshold denoising method is used to preprocess the original signal to suppress white noise interference and reduce the modal aliasing problem. Second, four indices are proposed to optimize the modal number K of the VMD, including the energy difference parameter e, the dispersion entropy De, the ratio of the frequency domain maximum value to the submaximum value R, and the dominant frequency interval Df of adjacent modes, which greatly preserves structurally effective feature information from vibration responses with multisource excitation and severe noise interference. The feasibility and accuracy of the proposed method are verified via simulation signals using the signal-to-noise ratio (SNR) and root mean square error (RSME). The calculated results are compared with those of empirical mode decomposition (EMD), variational mode decomposition (VMD), and wavelet threshold-EMD (WT-EMD). The results show that the SNR is increased by 78.7%, 20.9%, and 17.2%, respectively, and the RMSE is reduced by 25%, 11.1%, and 9.5%, respectively, indicating that the proposed method can accurately identify and effectively separate the frequencies of different excitation sources for pumping stations. Finally, the improved method is applied to the Dayuzhang Pumping Station during a specified operation period, and the dominant frequencies of the pumping station plant structure are 16.68 Hz, 29.19 Hz, 33.33 Hz, 58.37 Hz, 66.72 Hz, 83.40 Hz, and 116.74 Hz. The engineering case indicates that the effective vibration characteristic information of a pumping station plant can be identified via the proposed method under multisource excitation, which provides theoretical support for vibration source identification and operation state monitoring of pumping station plant structures.