Numerical design calculation of the fire resistance of steel members and their connections

The proposed doctoral project will develop a numerical design framework based on the reliability assessment of thin-walled steel structures, integrating advanced finite element modeling (FEM), digital twins, machine learning, and fire resistance analysis. Although geometrically and materially nonlinear analysis with imperfections (GMNIA) enables realistic simulations of structural behavior, its sensitivity to modeling assumptions limits its direct applicability in engineering decision-making, particularly under elevated temperatures.

The research will be based on the concept of a design-consistent digital twin, in which numerical models are continuously updated using data from structural health monitoring and experiments, in compliance with Eurocode requirements. The study will focus on welded and bolted connections of thin-walled steel members, taking into account their thermo-mechanical behavior under fire conditions.