Defect Characterization Method for Bridge Cables Based on Topology of Dynamical Reconstruction of Magnetostrictive Guided Wave Testing Signals

• Published in: Journal of nondestructive evaluation Vol. 42; no. 2
• Main Authors: Duan, Shuyu, Wu, Xinjun, Wang, Junjie, Zou, Yiqing, Jiang, Lijun, Wei, Yaolin
• Format: Journal Article
• Language: English
• Published: New York Springer US 01-06-2023; Springer Nature B.V
• Subjects: Bridge failure; Cable-stayed bridges; Cables; Characterization and Evaluation of Materials; Classical Mechanics; Control; Defects; Dynamical Systems; Engineering; Goodness of fit; Homology; Load bearing elements; Magnetostriction; Reconstruction; Solid Mechanics; Structural members; Suspension bridges; Topology; Vibration; Wire; Magnetostrictive guided wave testing; Persistent homology; Phase space; Dynamical reconstruction; Bridge cable
• ISSN: 0195-9298; 1573-4862
• DOI: 10.1007/s10921-023-00940-2

Multi-wire cables are widely used in suspension bridges and cable-stayed bridges as primary load bearing structural elements. Broken Wires in cables can lead to catastrophic accidents such as bridge collapse. Magnetostrictive guided wave testing technology has been employed to detect the broken wire defects in multi-wire cables, and the defect size is estimated by analyzing the defect echo signals. However, there are many studies on the guided wave testing for the seven-wire steel strands but fewer for the bridge cables which have a large number of wires. Moreover, the relationship between the guided wave testing signal features and the defect size of multi-wire structures is imprecise, which means the defect size estimated by the features may deviate significantly from the real defect size. In this paper, large-scale topological features are extracted by using persistent homology from the dynamical reconstruction topology of the guided wave testing signals to characterize broken wire defects in the bridge cable. The broken wire experiments were performed on a 61-wire cable. The experimental results show a good linear relationship (the goodness of fit 0.9946) between the large-scale topological features and the number of broken wires in the cable. It indicates that it is feasible to extract topological features from the topological domain of the testing signals to characterize the broken wire defects of bridge cables.