Computer Vision-based Finite Element Model Updating Method Using Measured Static Data: An Experimental Study

Abstract

<p>Accurate FE models play an important role in structure health monitoring (SHM). In the traditional static finite element model updating (FEMU) process, loading tests interrupting the traffic are required for obtaining static data, which is inconvenient. This paper proposes a novel static FEMU method based on computer vision technology and WIM system, avoiding the mentioned defects. Firstly, the static response simulation under traffic load is carried out with the computer vision determining the load location and the BIW system deciding the load value. Secondly, signal processing technology extracts the measured static data from the monitoring data. Thirdly, the PSO method is utilized to perform the FEMU. An experiment is designed on a bridge model with an SHM system, and results verify the convenience and accuracy of the proposed method</p>