A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Modelling the behaviour of extended shear tab connection using artificial neural network
The extended shear tab (EST) connection is an economical option for beam-beam or beam-column web connections in which a shear tab is welded to the supporting member and is bolted to the web of the supported beam. The present work aims at modelling the behaviour of EST using artificial neural networks (ANN) for predicting its failure mode. The ANN models were developed using the data sets produced by employing a finite-element solution of EST connection. For this, a database of 48 finite element models was generated and validated with experimental test results. The input parameters for ANN were selected based on average mutual information. Failure mode of EST connection using von Mises stresses by artificial neural network model was compared with the finite element method results. The comparison indicates that the proposed artificial neural network model achieved a better prediction of stresses. The promising results have shown that ANN can be used as a supplementary tool for the detection of failure of EST connection, to be used together with more rigorous 3D finite element analysis and/or experiment.
Modelling the behaviour of extended shear tab connection using artificial neural network
The extended shear tab (EST) connection is an economical option for beam-beam or beam-column web connections in which a shear tab is welded to the supporting member and is bolted to the web of the supported beam. The present work aims at modelling the behaviour of EST using artificial neural networks (ANN) for predicting its failure mode. The ANN models were developed using the data sets produced by employing a finite-element solution of EST connection. For this, a database of 48 finite element models was generated and validated with experimental test results. The input parameters for ANN were selected based on average mutual information. Failure mode of EST connection using von Mises stresses by artificial neural network model was compared with the finite element method results. The comparison indicates that the proposed artificial neural network model achieved a better prediction of stresses. The promising results have shown that ANN can be used as a supplementary tool for the detection of failure of EST connection, to be used together with more rigorous 3D finite element analysis and/or experiment.
Modelling the behaviour of extended shear tab connection using artificial neural network
Asian J Civ Eng
Satarkar, Priti R. (author) / Londhe, S. N. (author) / Dixit, P. R. (author) / Suleiman, Mohamed F. (author)
Asian Journal of Civil Engineering ; 24 ; 2767-2782
2023-12-01
16 pages
Article (Journal)
Electronic Resource
English
Modelling the Shear Behaviour of Clean Rock Discontinuities Using Artificial Neural Networks
| British Library Online Contents | 2017
Modelling the Shear Behaviour of Clean Rock Discontinuities Using Artificial Neural Networks
| Online Contents | 2017
Modelling the Shear Behaviour of Clean Rock Discontinuities Using Artificial Neural Networks
| Online Contents | 2017
Artificial neural network modelling of granular material behaviour under repeated loading
| British Library Conference Proceedings | 2000