An Hierarchical Approach for the Structural Damage Identification
Last modified: 2010-04-04
Abstract
The use of global techniques for damage identification is based on the fact that the presence of a fault will cause changes in the structural properties of the system, which in turn lead to a change in its dynamic response. The changes in the structural properties could be assessed by measures in the time, frequency or modal domains. To solve this inverse problem - Structural Health Monitoring - is possible to use, for example, methods from the Calculus of Variations, with the Conjugate Gradient Method (CGM - Alifanov approach) in the form of adjoint equation.
The central objective of this research is to significantly increase the speed of the method developed by Chiwiacowsky et al, 2005 which has been used for the damage identification. The new strategy for attacking the problem is based on the application of an hierarchical search to evaluating the different areas where damage may be present, starting with a coarse discretization and promoting gradual refinement only in the areas where the possibility of damage was not discarded.
Preliminary results has shown that, in case of random (non-correlated) distribution of damage, the hierarchical approach can be until twenty-three times faster than the traditional search.