Investigation of cold formed steel angle compression through high throughput design FEA and machine learning.
Journal:
Scientific reports
Published Date:
Jun 1, 2025
Abstract
This research investigates the finite element analysis (FEA) of cold-formed steel (CFS) L-columns with pin-ended supports under compression. The study involves a comprehensive parametric analysis with 110 FE models to assess the effects of section thickness and material yield stress on the axial strength of CFS L- columns. On the basis of the findings from this parametric study, a new design equation for determining the compression strength of CFS L- columns was proposed. This new equation demonstrated superior performance compared with the direct strength method (DSM) equations specified in the American (AISI S100-16) and Australian/New Zealand (AS/NZS 4600:2018) standazrds. To further validate the proposed design equation, a comparison with the finite element analysis (FEA) results was conducted. In addition, machine learning (ML) techniques were employed to forecast the load-bearing capacity of cold-formed sections. The predictions made from ML models were compared, including the FEM and the proposed equation. Evaluation metrics were used to compare predictions from various models, such as XGB, AB, RF, CB, FEM, and the newly introduced design equation, were used to compare predictions accuracy. All models showed similar high R values (0.99), indicating accurate and reliable axial capacity predictions.
Authors
Keywords
No keywords available for this article.
