Computational & Technology Resources
an online resource for computational,
engineering & technology publications 

CivilComp Proceedings
ISSN 17593433 CCP: 62
ARTIFICIAL INTELLIGENCE APPLICATIONS IN CIVIL AND STRUCTURAL ENGINEERING Edited by: B. Kumar and B.H.V. Topping
Paper V.6
Prediction of Ultimate Shear Strength of Reinforced Concrete Deep Beams using Neural Networks A. Sanad* and M.P. Saka+
*Ministry of Housing, Municipal Affairs, State of Bahrain
A. Sanad, M.P. Saka, "Prediction of Ultimate Shear Strength of Reinforced Concrete Deep Beams using Neural Networks", in B. Kumar, B.H.V. Topping, (Editors), "Artificial Intelligence Applications in Civil and Structural Engineering", CivilComp Press, Edinburgh, UK, pp 147157, 1999. doi:10.4203/ccp.62.5.6
Abstract
This study explores the use of artificial neural networks in
predicting the ultimate shear strength of reinforced concrete
deep beams. 111 experimental data collected from the literature
covers the simple case of simply supported beam with
two point loads acting symmetrically with respect to the centre
line of the span. The data is arranged in such a format
that 10 input parameters cover the geometrical and material
properties of deep beam while the corresponding output value
is the ultimate shear strength. Among the available methods
in the literature, ACI, Truss and MauHsu methods were
selected due to their accuracy and used to calculate the shear
strength of each beam in the set. Later, artificial neural network
is developed using two different software and the ultimate
shear strength of each beam is determined from these
networks. It is found that the average ratio of predicted and
actual shear strength was 1.01 for the neural network, 0.48
for ACI method, 1.17 for Truss method and 1.19 for MauHsu method. It is apparent that neural networks provide an
efficient alternative method in predicting the shear strength
capacity of reinforced concrete deep beams where several
equations exist, none of them producing an accurate result.
purchase the fulltext of this paper (price £20)
go to the previous paper 
