Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
286401 | Journal of Constructional Steel Research | 2006 | 7 Pages |
Distortional buckling of compression members usually comprises rotation and translation of each flange and lip about the flange-web connection in opposite directions. The present procedures for the calculation of elastic distortional buckling stress in the literature are very complex, cumbersome and have long expressions. In this paper a new neural network (NN) based formula is proposed for the determination of the elastic distortional buckling stress of cold-formed steel C-sections with both end sections pinned. The focus of this study is on the distortional buckling, for which existing results are for sections subjected to pure compression and/or pure bending only. The data used for training and testing NNs is taken from Schafer’s report. The NN-based estimates are compared with the experimental, numerical and analytical results of different researchers and methods. It was found that the proposed NN based-formula is practical in predicting the elastic distortional buckling stress of cold formed steel C-sections.