Structural Damage Detection Using Wavelet Support Vector Machine

Based on wavelet packet decomposition and conditions of the support vector kernel function, a nonlinear wavelet basis is introduced to construct the kernel function of support vector machine (SVM). A tighten wavelet support vector machine (WSVM), which has strong generalization ability is also obtained. In this study, a novel damage classification method based on wavelet support vector machine is developed for structural health monitoring. The response signals of a structure under an impact load are normalized and then decomposed into wavelet packet components. Energies of these wavelet packet components are then calculated as the feature vectors. The feature vectors are used for training and classification of the inputs of the tighten SVM. Then, the structural damage location and extent is determined by prediction. Numerical study is carried out on a simplysupported beam. Accelerations of the structure under impact loads are analysed. Results show that the method can be reliably used for damage monitoring and assessment of the structures.