Model updating of an existing bridge with high-dimensional variables using modified particle swarm optimization and ambient excitation data

Abstract

Model updating for bridge engineering structures to obtain precise finite element model is an efficient tool at present. But with the dimension of the potential parameters during optimization increasing, it brings a challenge to present optimization methods. To improve the performance of convergence process and the probability of detecting global extremum, a combination model updating method of modified particle swarm optimization (MPSO) and response surface method was proposed herein. MPSO has less modification to PSO as only modifying the particle position with Gaussian white noise at a probability and the performance was significantly superior than PSO illustrated by testing functions. A beam model updating example based on the proposed method was tested, the convergence ability and accuracy were compared with genetic algorithm which is widely applied in bridge model updating. Finally, the proposed method was successfully applied to model updating of an existing bridge engineering structures with thirteen variables.