STRUCTURAL HEALTH MONITORING USINGWAVELET TRANSFORMNUMERICALLY AND EXPERIMENTALLY

Abstract

Monitoring structure condition and detecting structural damage at the earliest possible stage havebeen a focus research recently. In this paper, a procedure for damage detection using waveletanalysis is presented. Damage identification in beams includes two steps. The first step is to detectthe damage location, and the second step is to calculate the damage severity. A numerical simulationby Matlab is created to calculate the beam mode shapes. The damage location could be obtained byanalyzing the fundamental mode shape by wavelet analysis. Effects of noise on detecting the damageare analyzed. Experimental tests are done to validate this technique in detecting damage location anddamage severity. Two wavelet transforms (continuous, discrete) are used to analyze both thecalculated and measured fundamental mode shape. Detailed parametric studies were conducted bychanging damage location, intensity and wavelet families. These studies showed that the proposedtechnique could accurately predict the damage location and its severity.