Damage shape recognition algorithm of composite woven fabric plate based on guided waves. (1st January 2023)

• Record Type: Journal Article
• Authors: Zhengyan, Yang; Lei, Yang; Jiaqi, Zhang; Shuyi, Ma; Tong, Tian; Deshuang, Deng; Zhanjun, Wu
• Is Part Of: Composite structures. Volume 303(2023)
• Publication Date: 2023-01-01
• Subjects: Damage shape recognition -- Guided wave -- Anisotropic woven composite -- Semi-analytical finite element method -- Wave velocity measurement -- Time of flight; Composite construction -- Periodicals; Composites -- Périodiques; 624.18
• DOI: 10.1016/j.compstruct.2022.116351
• Languages: English
• ISSNs: 0263-8223

Abstract

Abstract: Modern lightweight woven fabric composite structures play an important role in industrial field. The occurrence and accumulation of structure damages can threaten the integrity of in-service structures. With the high damage sensitivity, guided wave-based structural health monitoring has been regarded as the effective tool for damage detection. In this study, a damage shape recognition algorithm has been developed to continuously identify the damage extent in composite woven fabric structures, which is necessary for further analysis of remaining life estimation. The proposed algorithm is established relying on the accurate guided wave propagation information and precise damage scattered wave sources identification, and consists of the following stages: guided wave velocity calculation, guided wave signals collection, damage location identification, extraction of time of flight (TOF), damage boundary estimation, noise points filtering and damage shape and size recognition. The semi-analytical finite element method is presented to model guided wave propagation in multi-ply woven fabric panels. The numerical predictions are experimentally validated using the novel omnidirectional wave velocity measurement of guided wave collected by mobile PZT transducer. The precise TOF extraction and noise point elimination method are first proposed for improving effectiveness of damage identification. The approach was validated by several cases of damage at arbitrary positions. The estimated numerical and experimental results show the potential of the proposed algorithm in identifying various damage features in anisotropic composite plates, such as its location, shape and size, and achieve good agreement with the actual damage.