Answers are grounded in this paper and cite their sources.
Traditional approaches to the design of steel structures involve the determination of internal forces and moments through structural analysis, followed by a series of strength and stability checks on individual structural members. Recent advances in computational power and tools, as well as standardisation, have facilitated the adoption of design by GMNIA (Geometrically and Materially Nonlinear Analyses with Imperfections), enabling the direct capture of the key behaviour characteristics of structural elements and systems. This approach eliminates the need for subsequent, often cumbersome member checks typically required in conventional design practice. The present study extends the framework of a recently proposed advanced design method that employs computationally efficient beam finite element (FE) models to perform GMNIA and utilises strain limits, defined by the Continuous Strength Method (CSM), to capture cross-section failure, to the in-plane design of structural members composed of homogeneous and hybrid high strength steel (HSS) welded I-sections. The proposed advanced design method provides a rational consideration of material nonlinearity (i.e. the spread of plasticity and material strain hardening), while duly accounting for the effects of moment gradients in determining ultimate resistances. The accuracy of the proposed advanced design method for homogeneous and hybrid HSS welded I-section members is evaluated through comparisons with a substantial dataset of benchmark numerical results obtained from shell FE models, considering a range of steel grades, loading scenarios and cross-section and member slendernesses. In comparison to the traditional design approaches provided in the current European Standard Eurocode 3 (EC3), the proposed advanced design method offers more accurate and less scattered resistance predictions, providing opportunities for more streamlined and sustainable structures.
No takes yet. Share an insight, caveat, or question.
Zhu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: