A Backpropagation Neural Network Model for Semi‐rigid Steel Connections

Abstract

The analysis of semirigid steel structure connections based on exact theoretical modeling, which is demanding and time consuming if all the nonlinear parameters of the problem are taken into account, can be avoided provided that enough experimental measurements exist and an appropriate predictor can be constructed from them. A supervised learning backpropagation neural network approach is proposed in this paper for the construction of this model free predictor. A number of experimental momentrotation curves for single-angle and single-plate beamto-column connections are used in this paper to train the neural network. The trained network provides us with an estimator for the mechanical behavior of the steel structure connection element. 0 1995 Microcompufers in Civil Engineering. Published by Blackwell Publishers,

Behavior appraisal of steel semi-rigid joints using Linear Genetic Programming

“…A comparison of the ratio between the predicted initial rotation stiffness values by the LGP single and team solutions and likewise those obtained from European design code, ANNs [16], GP/SA [27,28], Kishi et al [3] and experimental values are illustrated in Figs. [13][14][15]. Performance statistics of models for initial rotational stiffness prediction are presented in 9.35%, 12.85% and 0.9923, 1.73%, 2.22%, respectively.…”

“…Some investigators have concentrated on predicting the beamcolumn steel joints behavior using Artificial Neural Networks (ANNs) [12][13][14][15][16]. In spite of the generally successful performance of ANNs, they are black-box models that do not give a deep insight into the process of using available information to obtain a solution.…”

A design‐variable‐based inelastic hysteretic model for beam–column connections

“…The model consists of three beam-column elements and has a total of 10 degrees of freedom. Consistent mass matrix for the beam-column elements and a lumped mass (1.2×10 4 The column base has a free horizontal DOF so that base excitation can be imposed. Tangent stiffness of the rotational spring element for the column base is derived from the phenomenological model [10] and non-pinching cyclic behavior is assumed.…”

Behavior of bolted angle connections subjected to combined shear force and moment

“…Abdalla and Stavroulakis [14] have used neural networks to predict the global moment-rotation curve of single web angle beam-to-column joints. De Lima et al [15] employed neural networks for assessment of beam-to-column joints.…”