Damage detection of riveted truss bridge using ANN-aided AMS optimization method

Abstract

Aging transport infrastructure brings increased economic burden and uncertainties regarding the reliability, durability and safe use of structures. Early damage detection to locate incipient damage provides an opportunity for early structural maintenance and can guarantee structural reliability and continuing serviceability. This paper describes the use of the hybrid identification method, which combines a metaheuristic optimization technique aimed multilevel sampling with an artificial neural network-based surrogate model to approximate the inverse relationship between structural response and structural parameters. The method is applied to identify damage in existing riveted truss bridge. The effect of the damage rate and location on the identification speed and the accuracy of the solution is investigated and discussed.Aging transport infrastructure brings increased economic burden and uncertainties regarding the reliability, durability and safe use of structures. Early damage detection to locate incipient damage provides an opportunity for early structural maintenance and can guarantee structural reliability and continuing serviceability. This paper describes the use of the hybrid identification method, which combines a metaheuristic optimization technique aimed multilevel sampling with an artificial neural network-based surrogate model to approximate the inverse relationship between structural response and structural parameters. The method is applied to identify damage in existing riveted truss bridge. The effect of the damage rate and location on the identification speed and the accuracy of the solution is investigated and discussed.