Assessing failure mechanisms in reinforced concrete frame structures under thermo-mechanical loading using finite element analysis

Abstract

Numerous studies have explored the failure mechanisms of reinforced concrete structures exposed to elevated temperatures. To simulate the action of fire on full-scale reinforced concrete buildings, four factors must be considered: the presence of loading, fire location, intensity, and duration. This is because the material behavior depends on the stress level, intensity, and duration of fire, and the sensitivity of the structural element to the location and application of fire. In this study, finite element analysis (FEA) was used to consider the combined effect of the mechanical loading and high temperature. To simulate the intensity and duration of the temperature rise, transient-state analysis must be performed. Two-dimensional, four-bay, three-story frames were analyzed under different cases of infill configurations subjected to high-temperature and working load conditions. The findings of this study related to the critical column of the frame, pattern of infill stresses, Demand Capacity Ratio (DCR), temperature, and time to failure were obtained and compared. The major conclusions are that the middle column is critical for both the bare frame and the infilled frame (with brick masonry and cement mortar interface) under high temperature and working load conditions. The investigation of the Bare Frame showed that the maximum vertical displacement is greater than that in the infilled frame, while the DCR is greater in the fully infilled frame, under high temperature at the first-story level (directly above the ground level) combined with working load conditions. Additionally, artificial neural network (ANN) models were developed to predict the vertical and lateral displacements observed in FEA during the transient-state analysis. Despite challenges in training ANNs, the models demonstrated strong potential in capturing complex structural behaviors under transient-state conditions.