Scholarlink

Purpose <p>To improve the accuracy of modal parameter identification in beam structures, this study presents an enhanced empirical Fourier decomposition (EFD) methodology for signal decomposition and modal parameter recognition under environmental excitations.</p> Methods <p>The proposed identification framework comprises two key innovations: First, we implement multiple signal classification (MUSIC) to determine optimal frequency segmentation boundaries, effectively replacing conventional Fourier spectrum analysis that demonstrates vulnerability to noise contamination. Second, we develop an improved EFD variant (designated as EFDM) to decompose acquired vibration responses into distinct modal components.</p> Results <p>Subsequent processing involves extracting free vibration responses from these modal components through synergistic application of Hilbert transform (HT) and random decrement technique (RDT), enabling precise identification of structural modal parameters. Comprehensive numerical simulations and experimental validations substantiate the method's enhanced identification accuracy and operational robustness, particularly demonstrating superior noise immunity compared to conventional approaches.</p> Conclusion <p>The results confirm the methodology's effectiveness in practical structural health monitoring applications, offering improved reliability in ambient vibration analysis scenarios.</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

Modal Parameter Identification of Beam Structures Based on Improved Empirical Fourier Decomposition

  • 期刊论文

  • 发表日期：

    2025年5月14日

查看全文
  • Zhuyao Du,
  • Hu Sun,
  • Chen Rong,
  • Wang Yan,
  • Sang Dan,
  • He Jia

摘要

Purpose

To improve the accuracy of modal parameter identification in beam structures, this study presents an enhanced empirical Fourier decomposition (EFD) methodology for signal decomposition and modal parameter recognition under environmental excitations.

Methods

The proposed identification framework comprises two key innovations: First, we implement multiple signal classification (MUSIC) to determine optimal frequency segmentation boundaries, effectively replacing conventional Fourier spectrum analysis that demonstrates vulnerability to noise contamination. Second, we develop an improved EFD variant (designated as EFDM) to decompose acquired vibration responses into distinct modal components.

Results

Subsequent processing involves extracting free vibration responses from these modal components through synergistic application of Hilbert transform (HT) and random decrement technique (RDT), enabling precise identification of structural modal parameters. Comprehensive numerical simulations and experimental validations substantiate the method's enhanced identification accuracy and operational robustness, particularly demonstrating superior noise immunity compared to conventional approaches.

Conclusion

The results confirm the methodology's effectiveness in practical structural health monitoring applications, offering improved reliability in ambient vibration analysis scenarios.