Please use this identifier to cite or link to this item: http://nopr.niscair.res.in/handle/123456789/14145
Title: Predicting the flexural behaviour of reinforced concrete and lightweight concrete beams by ANN
Authors: Kamanli, Mehmet
Kaltakci, M Yasar
Bahadir, Fatih
Balik, Fatih S
Korkmaz, H Husnu
Donduren, M Sami
Cogurcu, M Tolga
Keywords: Mechanical properties;Aggregates;Concrete;Modelling;Reinforced concrete;ANN
Issue Date: Apr-2012
Publisher: NISCAIR-CSIR, India
Abstract: In this study, artificial neural network (ANN) method is used to predict the deflection values of beams and compared with the experimental results of a testing series. For this purpose six reinforced concrete beams with constant rectangular cross-section are prepared and tested under pure bending. The concrete of the test specimens is casted using the lightweight aggregates obtained from volcanic sediments. The lightweight concrete has some advantages comparing the traditional concrete, such as less self weight, less earthquake forces due to decreased mass, good sound and thermal insulation. The use of lightweight concrete in the construction industry is popular due to various advantages. The neural network procedure is applied to determine or predict the deflection values of 1/1 scaled model beams. The analytical results are compared with the test results and further predictions, including different mix designs can be possible at the end of the study. As a result, while the statistical values RMSE, R2 and MAE from training in ANN model are found as 0.266, 99.2% and 0.216, respectively, these values are found in testing as 0.370, 96.47% and 0.419, respectively.
Page(s): 87-94
URI: http://hdl.handle.net/123456789/14145
ISSN: 0975-1017 (Online); 0971-4588 (Print)
Appears in Collections:IJEMS Vol.19(2) [April 2012]

Files in This Item:
File Description SizeFormat 
IJEMS 19(2) 87-94.pdf441.1 kBAdobe PDFView/Open


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