Structural Health Monitoring And Damage Detection Based On Wavelet Analysis

Long-term use and environmental factor will cause various defects(damages) in many structures in engineering. In order to improve promptly and effectively salty and reliability of the structures and decrease the cost for repair, it is very important to detect exactly when and where those defects arise.Traditional Fourier transformation(FT) can be used to detect the whole characteristics of signal singularity but it is difficult to identify the exact location of those singular points and distributed situation. Wavelet transformation (WT) call magnify the signals in time and frequency domains. Based on a constructed wavelet basic function, is introduced in detection of time, location and severity of structural damages. The contents of the paper are arranged as follows.According to the prosperities of dynamic parameters for damaged structures, a new exponential-type wavelet base is constructed. The wavelet base presents property of vanishing moment for even terms in it. Using the Lipschitz index, the principal of detection of the signal is demonstrated by analyzing the effect of the signal. A numerical simulation shows that the wavelet base is valid for detection of the damage, and Can be adopted in damage alarming.From the view of pattern classification, extraction of damaged structural signal features is carried out using wavelet packet transform(WPT). which possesses property of high resolution. Using nodes energy for wavelet packet decomposition tree, vibration signal in infinite time domain can be proceed using the index from finite nodes energy. On the basis of the theory of curvature mode based damage detection, a parameter reflecting energy variety rate for wavelet packet nodes curvature mode before and after the damage is chosen as an index for damage alarming. Numerical examples for a beam structure are given to verify the validity of use of the index.Finally, some results are summarized in this project with prospects for the further research.