Fitting of mechanical material behaviour of structural steel for numerical modelling

Abstract Numerical simulations are widely used in structural engineering research, and validated models enable the efficient expansion of datasets without costly additional testing. They allow the investigation of alternative geometries and loading scenarios beyond those covered by experiments. Their reliability, however, depends strongly on the accuracy of the material modelling. In structural steel, elastic behaviour is well captured by Hooke's law, but the plastic range remains challenging to describe realistically. Existing hardening models provide mathematical representations of the plastic portion of the true stress–strain curve, yet numerous studies have shown that these models require extensive calibration to reproduce experimental behaviour satisfactorily. This calibration is typically iterative, time‐consuming, and often yields results that still offer potential for refinement. To address this, an automated optimization tool has been developed to fit flow curves directly from uniaxial tensile test data. This article outlines the fundamentals of material modelling for steel in numerical simulations and presents the development and validation of the proposed optimization algorithm, which allows realistic numerical modelling on a scientific level and aligns with the ongoing standardization efforts in prEN 1993‐1‐14.