Scholarlink

<p>Accurate identification and assessment of bridge damage are vital for ensuring structural integrity and longevity. Conventional methods employing principal component analysis often depend on empirical decomposition of dynamic vibration components, which can introduce uncertainties. To overcome these limitations, this study proposes a bridge damage identification method integrating principal component analysis with dynamic time warping. Studies have shown that the first principal component closely correlates with lower-order bridge modes, effectively capturing structural dynamic behavior. Simulations demonstrate precise single damage localization, achieving errors as low as 2.0% at mid-span region. Experimental validation on a bridge model confirms robust reliability despite noise and boundary effects, with errors increasing to about 5.4%. However, multi-damage scenarios present challenges: simulations show localization errors rising to 9–12%, and experimental results become unreliable due to overlapping dynamic responses and noise, hindering accurate damage severity assessment. By leveraging minimal sensors, this approach provides an efficient, cost-effective monitoring solution. Future research should focus on refining multi-damage algorithms and improving noise filtering to enhance applicability in complex real-world environments.</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

Bridge damage detection based on principal component analysis and dynamic time warping

  • 期刊论文

  • 发表日期：

    2025年7月2日

查看全文
  • Xingtian Li,
  • Jinsong Zhu,
  • Xiaolei Jiao,
  • Shuai Zhou

摘要

Accurate identification and assessment of bridge damage are vital for ensuring structural integrity and longevity. Conventional methods employing principal component analysis often depend on empirical decomposition of dynamic vibration components, which can introduce uncertainties. To overcome these limitations, this study proposes a bridge damage identification method integrating principal component analysis with dynamic time warping. Studies have shown that the first principal component closely correlates with lower-order bridge modes, effectively capturing structural dynamic behavior. Simulations demonstrate precise single damage localization, achieving errors as low as 2.0% at mid-span region. Experimental validation on a bridge model confirms robust reliability despite noise and boundary effects, with errors increasing to about 5.4%. However, multi-damage scenarios present challenges: simulations show localization errors rising to 9–12%, and experimental results become unreliable due to overlapping dynamic responses and noise, hindering accurate damage severity assessment. By leveraging minimal sensors, this approach provides an efficient, cost-effective monitoring solution. Future research should focus on refining multi-damage algorithms and improving noise filtering to enhance applicability in complex real-world environments.