Local buckling behavior in high-strength steel beams: A concentrated plasticity approach

This paper introduces the first model designed to simulate local buckling in high-strength steel (HS-S) beams within moment-resisting frame systems (MRFs), especially under seismic loading conditions. Existing code limitations on section slenderness are derived from research on conventional steel, overlooking the special stress- strain characteristics of HS-S. To address this, a 3D finite element (FE) model developed in ABAQUS, calibrated with experimental data, investigates the moment-rotation hysteretic behavior of HS-S beams. Building on this, a new concentrated plasticity-based model (CPM) is proposed using an idealized tri-linear moment-rotation curve with three anchor points representing yielding, post-yield, and residual strength stages. Gene Expression Programming (GEP), a state-of-the-art AI tool, refines and formulates the mechanism of these anchor points, achieving prediction accuracies of 95 % for peak strength, 98 % for initial stiffness, and 90 % for dissipated energy. The partially mechanical-based CPM is implemented in SAP2000, offering engineers an accessible tool for analyzing HS-S beams with improved simulation accuracy. This model significantly enhances the prediction of local buckling, addresses the limitations of current design codes, and provides valuable insights for both new designs and the assessment of existing structures.