Predicting the Ultimate Load Capacity of R.C. Beams by ANN

Abstract

The present study deals with the use of artificial neural networks ANN in predicting the ultimate load capacity of reinforced concrete beams. The data is collected from the numerical solution by finite element method of the simply supported beams with various properties, under the action of two point loads, symmetrically with the center. The data were arranged in a format such that input parameters cover the geometrical, reinforcements ratio and properties of beams and the corresponding output is the ultimate (failure) load. Results were compared with the available methods in the literature. It was found that the average ratio of numerical solution (finite element) to predicted failure loads of beams was 1.018 for neural network, and 1.21 for limit state theory. It is clear that neural network provides an efficient alternative method in predicting the ultimate load capacity for R.C. beams.