Loading...

Using Vibration Behavior Sensitivity of Beams to Damage for Detection of Damage Location and Extent

Banihashem, Alireza | 2014

570 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 46527 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Joghataie, Abdolreza
  7. Abstract:
  8. This research is on damage detection in beams using changes in their vibration behavior with respect to damage. The aim of this research is to investigate the sensitivity of strain mode shapes to damage and the feasibility of their using in damage detection. Comparison with displacement mode shapes is another main subject of this research. For this aim, at first an experimental study has been conducted at the structures lab of Sharif University of Technology, to investigate the sensitivity of modal parameters of an Aluminum beam to progressive damage. Then a beam has been modeled numerically sing shell elements. Also a new frequency response function (FRF), here called the strain frequency response function (SFRF), has been defined and its corresponding modal parameters have been extracted. Next, noise has been added to the force and also the measured response time histories in order to make them more realistic similar to the real and experimental conditions. The natural frequencies and displacement mode shapes have been extracted from the displacement FRFs while the natural frequencies and strain mode shapes have been extracted from SFRFs. This has been done both for the numerical and experimental models and the results have been compared. The analysis of data and results show that there is practically a strong similarity between the experimental and numerical results. Also comparison of the sensitivity of both displacement and strain mode shapes confirms that the strain mode shapes are more sensitive to damage. In the next stage, the displacement and strain mode shapes have been used in detection of a progressive damage in the beam. This required solving an optimization problem which was accomplished using genetic algorithm (GA). Both displacement and strain data have been separately utilized by the GA program to detect the damage(s). The result show that except for one case, in the other cases, utilizing strain data can lead to more precise results, both in the detection of damage location and extent
  9. Keywords:
  10. Damage Detection ; Modal Analysis ; Frequency Response ; Model Updating ; Sensitivity Analysis ; Strain Mode Shape

 Digital Object List

 Bookmark

No TOC