Abstract

Structural Health Monitoring (SHM) and prognosis could play a crucial role in the life cycle of a critical infrastructure such as a large dam. Such monitoring practices serve multiple functions and are essential components in ensuring the safety and effective maintenance of such structures. By providing insights into a structure's behaviour, structural health monitoring assists in calibrating numerical models and complements visual inspections. This process facilitates informed decision-making, aids in planning maintenance and rehabilitation schedules, and ensures the desired safety of the structure throughout its lifespan. This paper presents a case study of a double curvature large thin concrete arch dam that is currently experiencing slow irreversible upstream movement of the central dam crest. The dam is wellinstrumented, with instrumentation falling into categories such as hydrometrological, geotechnical, geodetic, and seismic monitoring. The primary objective of this paper is to validate and establish a level of confidence in numerical modelling by comparing the displacement of an instrumented concrete arch dam with the with that calculated from its finite element method (FEM) model. To achieve this, a sequential thermo-mechanical analysis was performed on a 3D model of the dam foundation reservoir system. Additionally, the study utilizes a well-known statistical model called Hydrostatic-Season-Time (HST) model, based on multi-linear regression analysis, to predict the dam's future displacement using the relevant data from the sensors placed in the dam. The findings indicate a good correlation between the observed dam displacement based on instrumentation data with that calculated from the Finite Element Analysis (FEA) considering the presence of vertical joints. Moreover, the HST model proves to be a suitable choice as it closely matches the observed data and demonstrates good predictive capabilities for future dam behaviour and helps isolate the effects of different parameters on the dam displacement.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call