Abstract:Model updating is an effective means of damage identification and surrogate modeling has attracted considerable attention for saving computational cost in finite element (FE) model updating, especially for large-scale structures. In this context, a surrogate model of frequency is normally constructed for damage identification, while the frequency response function (FRF) is rarely used as it usually changes dramatically with updating parameters. This paper presents a new surrogate model based model updating method taking advantage of the measured FRFs. The Frequency Domain Assurance Criterion (FDAC) is used to build the objective function, whose nonlinear response surface is constructed by the Kriging model. Then, the efficient global optimization (EGO) algorithm is introduced to get the model updating results. The proposed method has good accuracy and robustness, which have been verified by a numerical simulation of a cantilever and experimental test data of a laboratory three-story structure.
Abstract. Vibration signal and its derivative have shown some promise in structural damage detection in previous research. However, the theoretical and practical difficulties of multi-damage detection in plate structures based on dynamic responses remain. In this paper, an efficient damage localization index based on frequency response function (FRF) is presented.The imaginary part of FRF (IFRF) is extracted to derive the new localization index due to its relation to modal flexibility. For avoiding the finite element model error, two-dimensional gapped smoothing method (GSM) is employed without the need for baseline data from a presumably undamaged structure. Experimental studies on a steel plate with two localized defects in different boundary conditions are performed. The results are compared with some typical damage indices in the literature, such as mode shapes, uniform load surface and IFRF.In order to mitigate the inherent disadvantages of GSM in anti-noise ability, a simple statistical treatment based on Thompson outlier analysis is finally used for noise suppression. The effect of damage level and boundary condition on the detection results is also investigated.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.