Damage detection in steel-concrete composite bridge using vibration characteristics and artificial neural network

Abstract

This paper develops and applies a procedure for detecting damage in a composite slab-on-girder bridge structure comprising of a reinforced concrete slab and three steel I beams, using vibration characteristics and Artificial Neural Network (ANN). ANN is used in conjunction with modal strain energy-based damage index for locating and quantifying damage in the steel beams which are the main load bearing elements of the bridge, while the relative modal flexibility change is used to locate and quantify damage in the bridge deck. Research is carried out using dynamic computer simulations supported by experimental testing. The design and construction of the experimental composite bridge model is based on a 1:10 ratio of a typical multiple girder composite bridge, which is commonly used as a highway bridge. The procedure is applied across a range of damage scenarios and the results confirm its feasibility to detect and quantify damage in composite concrete slab on steel girder bridges.