No abstract

Damage detection in structures using vibration analysis has been a subject of intensive investigation for the last two decades. In this paper, a method for damage detection using wavelet analysis is presented. MATLAB program is prepared to simulate a beam, calculate its mode shapes and analyze the first mode shape by discrete wavelet analysis. The first mode shapes of damaged and undamaged beams are transformed to the wavelet domain. The difference of the detail coefficients of damaged and undamaged mode shapes displays accurately the damage location. A detailed parametric study was conducted by changing the damage location, intensity as well as the boundary conditions and wavelet families. These studies have indicated that discrete wavelet transform accurately predicts the damage location (single or multi damages) from analyzing the first mode shape.

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