ABSTRACT
The traditional way for evaluating the fire resistance of the steel frame load-bearing structures, based on the conceptual isolation of the particular structural components of the considered frame (such as the beams and the columns), with more or less precisely modeled boundary conditions assigned to these elements, and on the subsequent specification of the sought fire resistance for each of these components separately, usually does not lead to the sufficiently credible estimates. During the analysis of such a type one is incapable of correctly estimating the mutual structural interactions between the components identified beforehand. It is essential that those interactions can often determine the behaviour of the whole structure in fire conditions. This will be shown in a current paper. Thus, it is postulated that, if only possible for problems of the type considered here, the whole structure, or at least structurally and functionally homogeneous subcomponents of it, be analyzed as a whole. For instance the paper (Maślak & Snela 2013) represents an attempt at such, in a sense holistic, approach, in which in order to estimate the fire resistance of a beam in a frame, the action of adjoining columns is taken into account in an indirect manner. This allowed for the verification of the importance of a fire specific risk, consist-
affecting the forecast redistribution of internal forces, may not suffice to obtain the fire resistance estimate ensuring the reliability sought by the evaluator. In the formal model submitted for the analysis one has to sufficiently precisely imitate the nonlinear response of the considered structure to a fire exposure. This behaviour is due to the physical nonlinearity prescribing the response of the structural steel to high temperature influence and due to the geometrical nonlinearity induced by the large displacements and thus the large deformations of the structural components are considered. Thus, an attempt to attain a sufficient precision in forecasting of the real fire resistance for the considered frame forces an application of a time consuming iterative computational algorithm. The evaluator has at his hand a plethora of widely known computational approaches. But, as all these approaches are very time consuming, an application of the numerical routines is very convenient. The authors of this paper intend to apply such the routines to verify the independently proposed simplified analytical approaches. Such a verification should allow for the determination of the applicability limits and constraints for the simplified computational models popularly recommended for engineering practice as well as for the quantitative evaluation of the magnitude of an error inherent in the application of such the simplified models. A detailed analysis of the behaviour of a classical substructure in the typical steel frame subjected to a fire, focused on the observation of the process of the redistribution of internal forces induced by a direct fire exposure, constitutes an objective of this paper. The link between this process and the deformation level of the frame components, observed and changing during fire, is of special interest here. The potential risks, which may result in the considered structure reaching the limit state of a fire resistance will be identified as a result of a current work. This in turn will enable the authors to indicate the weak points of the structure threatened by a fire and subjected to the analysis. It has to be underlined here, that such the risks may be hardly distinguishable, if only simplified analytical models are used during calculations.
