Application of Support Vector Machine in Structure Damage Identification

Since the security accident of structures occurred continually in recent years, the damage identification is paid close attention by scholars. The support vector machine (SVM) as a new method of statistical theory is applied to identify structural damage in this paper. The relative change quantity of modal flexibility, used as the characteristic index of damage identification, is input in SVM classifier to identify the location and degree of structural damage. A simulative simply supported beam is set up and input with different level noises. The analysis results indicate that this method is feasible to identify the location and degree of structure damage with low noise.