Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 64
COMPUTATIONAL ENGINEERING USING METAPHORS FROM NATURE
Edited by: B.H.V. Topping
Paper I.3

Prediction of Moment-Rotation Characteristic for Saddle-like Connections using FEM and BP Neural Networks

A. Kaveh+, D. Fazel-Dehkordi* and H. Servati+

+Department of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
*Mazadaran University of Science and Technology, Babol, Iran

Full Bibliographic Reference for this paper
A. Kaveh, D. Fazel-Dehkordi, H. Servati, "Prediction of Moment-Rotation Characteristic for Saddle-like Connections using FEM and BP Neural Networks", in B.H.V. Topping, (Editor), "Computational Engineering using Metaphors from Nature", Civil-Comp Press, Edinburgh, UK, pp 15-24, 2000. doi:10.4203/ccp.64.1.3
Keywords: neural network, backpropagation, saddle-like connection, steel structures, semi-rigid joint, finite element, moment-rotation.

Abstract
Finite element models are suggested for the analysis of a special type of semi-rigid connections of frame structures, known as saddle-like connections. The performance of these models is verified through the available experimental results. Non-linear finite element analysis is performed for a total number of 138 connections, and M-phi diagrams are obtained. Each diagram is represented by a multi-linear relationship through seven points. A neural network is trained for the prediction of M-phi diagrams for saddle-like connections. The results obtained by the trained backpropagation neural network are found to be quite satisfactory compared to those of the finite element analysis.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description