Damage Identification In Beam-Like Structures Based On Support Vector Machine And Modal Frequencies

L Liu,G Meng

2005-01-01

Abstract:Support Vector Machine (SVM) is a machine learning algorithm based on statistical learning theory, and it has recently been established as a powerful tool for classification and regression problems. A multi-step damage identification approach based on support vector machine (SVM) and modal frequencies is presented to locate and estimate the severity of the damage in beam-like structures. In the first section, the basic theory of the support vector machines is briefly reviewed. Damage features formed by vibration modal frequencies are used as characteristic parameters. Firstly, the possible damage locations are detected using support vector classification according to the probability distribution. Then after reconstructing the training set, the precise damage position is identified using support vector regression. Finally the damage severity is also identified. A series of experiments involving different damage scenarios (including cracks at different location and different sizes in the supported beams), have been conducted to verify this method. The results show that the support vector machine is a powerful and promising method for damage detection.