Multi-Damage Identification System of CFRP by Using FBG Sensors and Multi-Classification RVM Method

Abstract

Multi-damage identification of carbon fiber reinforced plastics (CFRP) structure is very important to ensure the structural safety. In this paper, the multi-damage identification system of CFRP using fiber Bragg grating (FBG) sensors and multi-classification relevance vector machine (RVM) method was proposed. First, the multi-damage identification method was researched by an active actuation method. In addition, the structural dynamic response signals were detected by the FBG sensors. Then, the multi-damage characteristics extraction was completed by Fourier transform, reliefF, and principal component analysis algorithms. With the multi-damage characteristics as the input and their corresponding multi-damage state as the output, the multi-classification RVM model was trained to identify the multi-damage state. Finally, the multi-damage identification system was established and verified on a CFRP plate with dimensions of 500 mm $\times \,\, 500$ mm $\times \,\, 2$ mm. The results showed that the proposed multi-damage identification method can accurately identify the CFRP structural multi-damage state. This paper provided a reliable method for the CFRP structural multi-damage identification.