The feed forward back propagation Artificial Neural Networks (ANN) was applied to develop two models for predicting the ultimate shear strength of reinforced concrete deep beams for Normal Strength Concrete (NSC) and High Strength Concrete (HSC). Both ANN models were trained and tested using experimental results. The input layer of the models comprised beam geometry, concrete and steel reinforcement properties. The output layer for both NSC and HSC models contained one parameter representing the ultimate shear strength. The ANN models successfully predicted the ultimate shear strength of deep beams within the range of the considered input parameters. The average ratio of the experimental to the predicted shear strength is 1.04 for normal strength concrete and 1.002 for high strength concrete. The predicted shear strength values were also compared with those calculated values using the ACI code. This comparison showed that the ANN model has a higher potential in predicting the ultimate shear strength of both normal and high strength deep beams within the range of input parameters. The trained network model was also used to perform a parametric study to evaluate the effect of the input parameters on the ultimate shear capacity.
Mohammed Arafa, Mamoun Alqedra and Haytham An-Najjar, 2011. Neural Network Models for Predicting Shear Strength of Reinforced Normal and High-strength Concrete Deep Beams. Journal of Applied Sciences, 11: 266-274.