PSI - Issue 64

Margherita Autiero et al. / Procedia Structural Integrity 64 (2024) 1798–1805 Author name / Structural Integrity Procedia 00 (2019) 000–000

1800

3

the local–distortional interaction (Dinis et al. (2014)). Methods to evaluate the buckling behaviour of cold-formed sections (CFS) are still an actual topic. The latest trend is to move from simplified design models to design procedures based on “whole section” analysis, avoiding the use of the effective width method (Schafer. 2000). This paper provides an assessment of a self-supporting automated multi-depth warehouse under fire conditions. The first part focused on establishing the fire modelling necessary for assessing the mechanical response and analyzing the collapse mechanism in the second part of the study. 2. Design of ARSWs in fire conditions In the context of the new Italian fire regulation, the fire resistance is defined as a passive fire protection measure to guarantee load bearing and compartmentation capabilities to the structures according to performance levels, selected by the designer to achieve the defined fire safety objectives. Five performance levels (PL) depending on the importance of the building, are defined, the PLI could be required for ARSWs, indeed, among the criteria required to fall into the PL I, there is the request that the building is not involved in activities concerning the presence of people, except for the occasional and short-term activity of highly trained workers. The latter is generally the condition of every ARSW. Different design solutions can be chosen to comply with the PL, based on prescriptive or performance based approaches, PA and PBA respectively. The analytical evaluation the bearing capacity, in case of fire, can be divided into six phases, from the i) definition of fire scenarios; ii) evaluation of fire action; iii) evaluation of the thermal response (thermal analysis); iv) evaluation of the mechanical action (load combinations), v) evaluation of the mechanical response (type of structural analysis and mechanical analysis); vi) verification of fire resistance, that is the level of safety expressed by the structure. These steps are the same for both PA and PBA approaches, with some simplification for the prescriptive one; the main difference between the prescriptive and the performance-based approaches is that the first one requires nominal fire curves. On the other hand, the PBA considers the complexity of structures using specific natural fire curves, which can be generally obtained by through simplified or advanced models. In ARSWs structures to consider the PBA, is essential because they generally require a minimum fire resistance performance for structural elements that lead to the use of traditional passive fire protection systems, which in the case of these metal profiles are difficult to apply, because of their high section factors (A m /V). Under fire conditions, the thin thickness of the CFS profiles, combined with the high thermal conductivity of the steel, induces a fast increase in the steel temperature with a significant loss in material stiffness and strength. At present, fire design methods for CFS members are not as investigated as for hot-rolled ones. At elevated temperatures, to account for local buckling the actual EC3 part 1-2 suggests for Class 4 cross-sections a default critical temperature of 350 °C is considered, if no fire design is conducted, which means that even for a requirement of R15, passive fire protection should normally be used for current profiles. Alternatively, the informative Annex E of the EC3 part 1-2 suggests ( i ) using an effective cross-section (A eff ) calculated with the effective width method, according to EC3 part 1-5, by considering the steel properties at ambient temperature, which means that the effective properties of a steel plate should be kept unchanged as the one as at ambient temperature, ( ii ) taking the 0.2% proof strength (f 0.2p,θ ) for the design yield strength of steel instead of the stress at 2% total strain (f y ), as normally used in the fire design of other cross-sectional classes. This means that the design buckling resistance of a compressed member with a non-uniform temperature distribution for the actual Eurocode can by evaluated as follow: (1) where, the reduction factor for flexural buckling in the fire design situation provided in EC3 part 1-2; effective cross-sectional area; , is the partial safety factor for the relevant material property, for the fire situation, taken as equal to 1; k 0.2p ,θ is the reduction factors for the 0.2% proof strength at elevated temperatures. Couto et al. (2014) proposed new expressions to determine the effective width of steel sections at high temperatures. This design curve has been calibrated as close as possible to the existing design curve by introducing the factors α θ and β θ on the expressions of EC3 part 1-5, hence the influence of the imperfections is taken into account as in the original formulas developed by Winter (1947) and additionally the non-linear steel constitutive law at elevated temperatures is also acco unted for, furthermore by using the factor ε θ steel grade is also taken into account in this new proposal. Moreover, by using these expressions, the strength at a total strain of 2% (f y,θ ) can be used to