Experimental and machine learning-based model for large-scale reinforced concrete shear walls strengthened with CFRP sheets and shape memory alloys

Decades of research have focused on improving the ability of structures to withstand dynamic loads. Numerous studies have established the effectiveness of Carbon Fiber Reinforced Polymers (CFRP) sheets in strengthening of existing RC walls. This research examines the effectiveness of using CFRP sheets to strengthen Nickel-Titanium (NiTi) alloy walls. Two large-scale NiTi-walls with an aspect ratio of 1.9 were repaired and strengthened by using unidirectional Sika Wrap 230-C CFRP sheets, then tested experimentally under lateral cyclic loading. Moreover, this study investigates the effectiveness of a strengthening technique that uses variable configurations of CFRP in RC-NiTi walls to improve their structural performance in terms of lateral load capacity, ductility, inter-storey drift, and hysteretic behavior. Furthermore, machine learning models (Fitting Neural Networks (FNN)) are developed to analyse the impact of various factors on shear wall strengthening techniques, including cross-sectional area, concrete strength, and CFRP sheet intensity. The proposed models’ validation demonstrates a high degree of agreement with the experimental results. The study demonstrated a remarkable improvement in shear walls strengthened with CFRP sheets, resulting in an average 38% increase in lateral load capacity and a 15% enhancement in energy dissipation.