关键词: 桥梁系统, 健康检测, 挠度共振, 线性回归, 异常“团”

Abstract: In order to solve the problem of large-scale bridge fault diagnosis, this paper proposes a new efficient algorithm. It addresses the issue of identifying abnormal cliques on bridges, where each clique is composed of a deflection checkpoint and the dedicated sensor. The detection method consists of three stages. First, inspired by “deflection resonance” between different checkpoints, a test pattern is created based on the theory of regression. Second, a set of tests on cliques are conducted based on the established test pattern, and the test outcomes are collected to form a syndrome. Finally, a detection result is achieved based on the syndrome. The proposed method is then applied to the detection of a real-life bridge, and result shows the detection rate reaches 81.8%.

Key words: bridge system, health detection, deflection resonance, linear regression, abnormal clique