Bridge damage detection using precise vision-based displacement influence lines and weigh-in-motion devices: Experimental validation: Fid-Bau Portal

Bridge damage detection using precise vision-based displacement influence lines and weigh-in-motion devices: Experimental validation

Ge, Liangfu / Koo, Ki Young / Wang, Miaomin / Brownjohn, James / Dan, Danhui

Abstract

Highlights • An input–output vision-based measurement system was proposed to obtain bridge Displacement Influence Line (DIL). • Experiments for damage detection using the vision-based measurements of DILs was conducted for different damage scenarios. • Random variations in vehicle speed were considered in the process of DIL measurement and damage detection. • The Chordwise Displacement Influence Line was proposed to compensate adverse friction effects in the boundary supports. • A concept of integrating WIM with vision systems was introduced for DIL-based damage detection on operational bridges.

Abstract This study presents an experimental validation for a high-precision vision-based Displacement Influence Line (DIL) measurement system for a purpose of damage detection on bridges. The vision-based DIL measurement system is a promising tool for structural health monitoring on real operation bridges, which combines two Computer Vision subsystems and weigh-in-motion (WIM) devices. Two vision systems are utilized for tracking vehicle position and measuring structural displacement, while the WIM device obtains vehicle weight information. To demonstrate the feasibility of such a vision-based DIL measurement system, this study developed a vision system using a Go-Pro camera for vehicle positioning and a consumer grade camera for displacement measurement, followed by a series of laboratory experiments on a simply supported bridge using vision-based DILs to assess damage existence and localisation. Five damage scenarios were created by restrengthening the test structure instead of damaging it. Each restrengthened structure was considered intact while the original structure was considered damaged. Vision-based DIL measurements were repeated 12 times for each damage scenario to observe uncertainties in damage localisation as well as DILs. As the measured DILs were found adversely affected by the friction on the boundary supports, the Chordwise Displacement Influence Line (cw-DIL) approach was proposed to compensate for this effect. Damage-induced cw-DILs were shown to be able to assess damage existence and localisation successfully and consistently for all five damage scenarios.