Experimental evaluation of vibration based damage identification techniques on a pedestrian bridge

Abstract

Damage in civil structures such as bridges has generally been identified by visual inspections, however dynamic structural health monitoring has been proposed as an alternative. This paper presents the experimental evaluation of three damage identification techniques on a full-scale footbridge. One of the evaluated techniques is based on damage localization vectors. A second technique is based on changes in the mode shapes curvature. The third technique uses a numerical model and artificial neural networks for locating the damaged section. Five scenarios of controlled damage were induced in the footbridge. Output-only ambient vibration tests were performed at each damage state and the results of the identification techniques were analyzed. The three implemented techniques showed promising results for the numerical simulations, and two of these techniques produced satisfactory results in the experimental evaluation.