Low energy impact damage identification method of CFRP structure based on wavelet transform and probabilistic neural network

Abstract

Low energy impact damage identification of carbon fiber reinforced plastics (CFRP) structure is of great significance to ensure the safe service of the structure. In this paper, the low energy impact damage identification method of CFRP structure by using fiber Bragg grating (FBG) sensor, continuous wavelet transform and probabilistic neural network (PNN) was studied. Firstly, the CFRP structure with low energy impact damage was actively excited, and the structural dynamic response signals were accurately detected by FBG sensors. Then, the dynamic response signals were subjected to continuous wavelet transform, and their wavelet time-frequency diagrams were extracted as the structural damage feature. On this basis, with damage feature as input and low energy impact damage as output, the damage identification model based on PNN was established. Finally, the experimental system was constructed to verify the feasibility of the method proposed in this paper. The results showed that for 12 groups of test samples, the damage identification model based on PNN can correctly identified 11 groups. This paper provided a feasible method for the low energy impact damage identification of CFRP structure.