Damage detection based on multi-scale entropy and support vector machine

Abstract

The cracks on the surface of mechanical structure often happen under heavy loads and harsh environments, especially in the aeroengine. So that it is significantly important to identify the cracks and detect the locations. In this paper, a damage detection method based on multiscale entropy (MSE) and support vector machine (SVM) is presented to anaylze the damages on a vibrated steel plate in our experimental platform. And 24 Fibre Bragg Gratings sensors(FBGs) are employed to monitor the dynamic strain on the different points of the steel plate through only 4 channels for data transmission. According to the measurement and analyzed results, this method is effective to determine the area containing the damages on the basis of points detected by the FGBs.