Prediction of the Torsional Strength of Reinforced Concrete Beams Subjected to Pure Torsion

2002 ◽  
Vol 14 (6) ◽  
pp. 1010-1021 ◽  
2012 ◽  
Vol 214 ◽  
pp. 306-310
Author(s):  
Han Chen Huang

This study proposes a artificial neural network with genetic algorithm (GA-ANN) for predicting the torsional strength of reinforced concrete beam. Genetic algorithm is used to the optimal network structure and parameters. A database of the torsional failure of reinforced concrete beams with a rectangular section subjected to pure torsion was obtained from existing literature for analysis. This study compare the predictions of the GA-ANN model with the ACI 318 code used for analyzing the torsional strength of reinforced concrete beam. The results show that the proposed model provides reasonable predictions of the ultimate torsional strength of reinforced concrete beams and offers superior torsion accuracy compared to that of the ACI 318-89 equation.


Structural beams in construction are subjected to significant torsional moment which affects the design of structures.Eight beams were produced with two distinct grade of concrete with two ratios of longitudinal as well as transverse reinforcements.An experiment for evaluation of torsional strength of reinforced concrete beams is presented in this paper.The main objective of this study is to access the role of varying percentage of transverse and longitudinal reinforcements on torsional strength of beams.Concrete grades of M 20 and M 30 beams were cast with 0.56% and 0.85% of longitudinal reinforcement as well as 50 mm and 75 mm spacing of stirrups.The experimental results are presented.The reported results include the behavioural curves and the values of torsional moment and angle of twist for entire 8 beams


Sign in / Sign up

Export Citation Format

Share Document