Techno Press
You logged in as Techno Press

Computers and Concrete
  Volume 9, Number 6, June 2012 , pages 469-486

Neuro-Fuzzy modeling of torsional strength of RC beams
A. Cevik, M.H. Arslan and R. Saracoglu

    This paper presents Neuro-Fuzzy (NF) based empirical modelling of torsional strength of RC beams for the first time in literature. The proposed model is based on fuzzy rules. The experimental database used for NF modelling is collected from the literature consisting of 76 RC beam tests. The input variables in the developed rule based on NF model are cross-sectional area of beams, dimensions of closed stirrups, spacing of stirrups, cross-sectional area of one-leg of closed stirrup, yield strength of stirrup and longitudinal reinforcement, steel ratio of stirrups, steel ratio of longitudinal reinforcement and concrete compressive strength. According to the selected variables, the formulated NFs were trained by using 60 of the 76 sample beams. Then, the method was tested with the other 16 sample beams. The accuracy rates were found to be about 96% for total set. The performance of accuracy of proposed NF model is furthermore compared with existing design codes by using the same database and found to be by far more accurate. The use of NF provided an alternative way for estimating the torsional strength of RC beams. The outcomes of this study are quite satisfactory which may serve NF approach to be widely used in further applications in the field of reinforced concrete structures.
Key Words
    reinforced concrete beam; neuro-fuzzy; torsional strength; building code.
A. Cevik: Department of Civil Engineering, University Of Gaziantep, Gaziantep, Turkey
M.H. Arslan: Department of Civil Engineering, Selcuk University, Konya, Turkey
R. Saracoglu: Department of Electronic and Computer Education, Selcuk University, Konya, Turkey

Techno-Press: Publishers of international journals and conference proceedings.       Copyright © 2022 Techno Press
P.O. Box 33, Yuseong, Daejeon 305-600 Korea, Tel: +82-42-828-7996, Fax : +82-42-828-7997, Email: