Nonlinear Finite Element Analysis and Artificial Intelligence (ANN)-Based Predictions for RC Beam Members Strengthened with CFRP Laminates

Abstract

Fibre-reinforced polymer (FRP) laminates are frequently used to strengthen reinforced concrete (RC) structural components. First phase of the present research has been focused on the behaviour of reinforced concrete beams wrapped with carbon fibre-reinforced polymer (CFRP) laminates. Nonlinear finite element approach has been followed to develop numerical models. The deflection, stresses, cracking and yielding damage are recorded for beam models with and without the CFRP laminates. It has been concluded that not only the presence of FRP laminates but also its location and area covered by it under different types of loading play an important role in guiding the overall structural behaviour of the RC beam strengthened with FRP. This kind of study requires a lot of computational time, expensive software packages, including technical personnel conversant with it. As an alternative approach, in the second phase, an ANN-based predictive model developed in MATLAB and Python has been suggested, using the data obtained from the first phase of the study, to predict the structural behaviour of RC beams strengthened with FRP laminates using different parameters as input. It has been noted that the difference between the results obtained from the finite element software and the ANN model is negligible.