2019 CSCE Annual Conference - Laval (Greater Montreal)

2019 CSCE Annual Conference - Laval (Greater Montreal) Conference


Title
Multiple Damage Localization of Gravity Dam: Strain Energy Based Approach using Random Data


Author(s)
Mr. Saikat Bagchi, Concordia University (Presenter)
Mr. Timir Baran Roy, Concordia University
Dr. Ashutosh Bagchi, Concordia University, Canada
Abstract

Continuous monitoring of large-concrete structures like Dam, using real-time information obtained using the health monitoring system, is necessary for the prognosis of any premature degradation of the structure. In this study, standard section of the Koyna dam, is modeled in 2D using the ABAQUS software. Material properties of the dam material are assumed to be linear and elastic. The first four modes of free vibration are extracted using linear perturbation technique. The model is first validated against the available results for Koyna Dam. As real-time vibration data is sparse for the structures like Dam due to its size and location, Gaussian White Noise is used for next step validation purpose, which in turn provides a robust numerical model applicable for damage detection. Damage is incorporated in the numerical model introducing a suitable change in the modulus of elasticity of the elements at a specific location. Damages at multiple locations are compared individually. Reflecting these damages Displacement Mode Shape (DMS) and Curvature Mode Shape (CMS) of the structure change. Strain Energy based Damage Index Method is another approach which takes into account the change in the flexural rigidity and corresponding change in the stored energy at each element. Modal dynamic parameters like Natural frequencies, Mode Shapes and Element Strain Energy (ESE) are compared between the pristine and a hypothetically damaged state to comprehend the efficiency of the localization method. Studying the modal responses of the numerical model of the gravity dam it is inferred that the change in the natural frequency may not be a clear indicator of damage if the size of the damage is small in comparison to the overall size of the structure. It is observed that although DMS and CMS are able to approximately identify the region of the stiffness change, none of them is precise enough to localize the damaged elements. While on the other hand, the 1st and the 3rd modes of ESE can identify the damaged location with sufficient exactitude. Therefore, it is rational to hold that for the mass concrete structure like gravity dam strain energy is a better parameter for damage localization. Finally, in this study Strain Energy based damage index is estimated to quantify the extent of the damage.