Scholarlink

<p>This article presents an analysis of the frequency response function to identify the locations and extent of structural parameter changes. The adverse effects of incomplete measurements, measurement errors, and uncertainties, such as modelling errors, are reduced to achieve more precise results. This is accomplished by applying a Bayesian approach and repeating multiple data tests in a model updating process. The Bayesian method can utilize the conditions provided by multiple structure measurements and finite element model updating to enhance damage detection. We studied the improvement in a damage detection method’s results through two experimental and numerical examples to prove this assertion. The experiment involved repeatedly testing a free-free beam in each scenario to gather the necessary data for the Bayesian method. However, the Monte Carlo repetitions create the data sets for the numerical example, a bowstring truss structure. The results showed the effectiveness of the proposed approach to lower the estimation errors in locating and quantifying different levels of changes in the 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

Damage Detection in the Structures Using Incomplete Noisy FRF Data by Bayesian Estimation and Model Updating

  • 期刊论文

  • 发表日期：

    2025年2月24日

查看全文
  • Babak Taghavi,
  • Fariba Shadan,
  • Masoud Mahmoodabadi,
  • Ahmad Fakharian

摘要

This article presents an analysis of the frequency response function to identify the locations and extent of structural parameter changes. The adverse effects of incomplete measurements, measurement errors, and uncertainties, such as modelling errors, are reduced to achieve more precise results. This is accomplished by applying a Bayesian approach and repeating multiple data tests in a model updating process. The Bayesian method can utilize the conditions provided by multiple structure measurements and finite element model updating to enhance damage detection. We studied the improvement in a damage detection method’s results through two experimental and numerical examples to prove this assertion. The experiment involved repeatedly testing a free-free beam in each scenario to gather the necessary data for the Bayesian method. However, the Monte Carlo repetitions create the data sets for the numerical example, a bowstring truss structure. The results showed the effectiveness of the proposed approach to lower the estimation errors in locating and quantifying different levels of changes in the structures.