Shear Strength of Reinforced Concrete Squat Walls

Squat shear walls are widely used in various structures to resist earthquake loads. However, the relevant design expressions found in building codes and literature do not incorporate the influence of all crucial parameters and provide inconsistent peak shear strength estimations. This study adopts the artificial neural network (ANN) to predict the peak shear strength of squat walls using an extensive database that includes the results of 487 walls with wide-ranging test parameters. The ANN models consider the effect of concrete strength, the wall aspect ratio, vertical and horizontal reinforcements, vertical reinforcement of boundary elements, and axial load ratio. These accurately predicted the available test results. They implemented it to carry out parametric and sensitivity analysis to investigate the effect of the main parameters on the peak strength and to give information about the factors that contribute most to the shear response. In addition, a softened strut and tie method is proposed, considering the variables that substantially influence the shear strength. A nonlinear regression analysis is employed to determine the coefficients of the proposed model using the available database. The performance of the proposed model is measured using the existing models, which results in the best favorable agreement with the test results.