KNUSTSpace >
Research Articles >
College of Engineering >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/11338

Title: Development of shear prediction models for rc beams without stirrups: artificial neural network approach
Authors: Owusu Afrifa, Russell
Adom-Asamoah, Mark
Issue Date: Sep-2016
Abstract: Using a shear database of 310 beams, a simple equation to predict shear was developed from regression analysis. The test to predicted ratio had a mean value of 1.36 and standard deviation of 0.4. Artificial Neural Network with a back propagation algorithm of five inputs (longitudinal reinforcement ratio, concrete strength, shear span to depth ratio, beam breadth and beam depth), with two hidden layer of seven and nine node respectively and having only one output layer (ANN 5791) provided a better prediction of shear strength of low strength reinforced concrete beams, recording a mean strength ratio of 1.03 and a standard deviation of 0.11. The ANN model was more accurate than the regression model in predicting shear capacity of beams without web reinforcement. For design purpose often to obtain conservative results, a reduction factor of 0.85 has been derived to multiply ANN values.
URI: http://hdl.handle.net/123456789/11338
Appears in Collections:College of Engineering

Files in This Item:

File Description SizeFormat
KNUSTDCESTRENG2016TR4.pdf796.55 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback