Crack detection in Mindlin-Reissner plates under dynamic loads based on fusion of data and models

Abstract

In this paper, system identification is coupled with optimization-based damage detection to provide accurate localization of cracks in thin plates, under dynamic loading. Detection relies on exploitation of strain measurements from a network of sensors deployed onto the plate structure. The data-driven approach is based on the detection of discrepancies between healthy and damaged modal strain curvatures, while the model-based method exploits an enriched finite element method coupled to an optimization algorithm to minimize discrepancies between the measured and modelled response of the structure. It is demonstrated, through a series of numerical experiments, that the fusion of data-driven and model-based approaches can be beneficial both in terms of accuracy and localization, as well as in terms of computational requirements.