Classification of Steel Semi-Rigid Connections by Neural Networks

Demirtaş, Betül; Santiago, Edwin; Oleary, Jeannie

doi:10.1061/40700(2004)168

Cited by 1 publication

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results were found to provide approximations to the experimental response that are satisfactory for use in structural engineering design. Recently, a neural network model was proposed by Demirtas et al [5] to obtain nonlinear moment-rotation curves for semi-rigid connections. The resulting model was then integrated into a non-linear frame analysis program to obtain nodal displacements and corresponding frame element forces.…”

Section: Introductionmentioning

confidence: 99%

An Artificial Neural Network Model for Predicting the Behaviour of Semi-Rigid Joints in Fire

Al‐Jabri¹,

Al-Alawi

Al-Saidy³

et al. 2009

Volume 5 Number 4

View full text Add to dashboard Cite

This paper presents an artificial neural networking (ANN) model developed to predict the behaviour of semi-rigid bare-steel joints at elevated temperature. Data for three flush end-plate and one flexible end-plate joints were considered. Sixteen parameters which included geometry of the joint's components, material properties of the joint, joint's temperature and the applied moment were used as the input variables for the model whilst the joint's rotation was the main output parameter. Data from experimental fire tests were used for training and testing the model. In total, fifteen different test results were evaluated with 331 and 61 cases were used for training and testing the developed model, respectively. The model predicted values were compared with actual test results. The results obtained indicated that the model can predict the moment-rotation behaviour in fire with very high accuracy. The coefficients of determination (R 2 ) for training and validation of the model were 0.964 and 0.956, respectively.Keywords: Bare-steel, flush end-plate, flexible end-plate, semi-rigid joints, artificial neural network, fire, elevated temperature, rotation INTRODUCTIONSteel loses both its strength and stiffness when subjected to fire. Fire tests on steel structures have shown that the temperature within the joints is lower compared to connecting steel members. This is attributed to the additional material around joints (column, end-plate, concrete slab, etc.) which significantly reduces the temperatures within the connections compared to those at the centre of supported beams. The experimental results on the behaviour of steel connections under fire conditions are relatively recent and limited, partly because of the high cost of the fire tests and the limitations on the size of furnaces used. Only limited joint tests have been performed and they were concentrated on obtaining the moment-rotation relationships of isolated joints (Al-Jabri et al.[1]). Therefore, experimental fire joint tests are not anticipated to be performed on many connection types with various end conditions. It is well known that even nominally 'simple' connections can resist significant moments at large rotation. At the severe deformation of structural members in fire, moments are transferred through the joints to the adjacent members, and hence, they may have a beneficial effect on the survival time of members.Accurate prediction of the structural behaviour of steel beam-to-column connections, by estimating the local deformations and induced stresses, is necessary to assess the capacity of the connections and prevent their failure. Numerical modeling presents, in principle, an alternative way to predict the response of structural steel joints in fire. Artificial neural network (ANN) modeling is an artificial intelligence-based technique that emulates the human ability to learn from the past experience and derive quick solutions to new problems. The developed ANN-based prediction model can be used by structural engineers to predict the elev...

show abstract

Section: Introductionmentioning

confidence: 99%