Damage detection in an offshore platform using incomplete noisy FRF data by a novel Bayesian model updating method

Abstract

Structural integrity monitoring of jacket structures is an attractive challenge faced by researchers worldwide. Because of numerous uncertainties in marine environments, using statistical methods to reduce the detrimental impacts of uncertainties on model updating and damage detection results are unavoidable. In this study, a new Bayesian model updating framework is proposed using incomplete Frequency Response Function (FRF) data. In this methodology, the incomplete measurements issue is not dealt with the model reduction or data expansion method and the number of data in the objective function is increased using FRF at different excitation frequencies. The experimental verification of a scale 2D fixed platform is implemented to reveal the validity of the proposed methodology. Several numerical damage scenarios are simulated to investigate the effect of noisy data, FE model uncertainties, incomplete measurement, and added mass in the damage detection procedure. According to the results, the introduced method is entirely successful in the model updating and damage detection of the jacket platform. The results also indicate the lower effects of uncertainties and noise levels in damage detection outcomes.