Neural network modeling of torsional strength of reinforced concrete beams

This study presents the application of Neural Networks (NN) for modeling torsion of RC beams. The NN models are proposed for the computation of selected two different NN models. The proposed NN models are based on a wide range experimental database. The accuracy of the proposed NN models is quite satisfactory as compared to results of design codes. Moreover, the results of proposed NN formulations are compared with existing models and are found to be more accurate than the expression available in the literature. The generalization capability of proposed NN models is also verified by a set of parametric studies.