Compagnone, Alberto (2020) Probabilistc approach for simplified verification methods of resistance of steel frames in fire. [Tesi di dottorato]

[img]
Anteprima
Testo
COMPAGNONE.pdf

Download (6MB) | Anteprima
[error in script] [error in script]
Tipologia del documento: Tesi di dottorato
Lingua: English
Titolo: Probabilistc approach for simplified verification methods of resistance of steel frames in fire
Autori:
AutoreEmail
Compagnone, Albertoalberto.compagnone@unina.it
Data: 12 Marzo 2020
Numero di pagine: 204
Istituzione: Università degli Studi di Napoli Federico II
Dipartimento: Ingegneria Civile, Edile e Ambientale
Dottorato: Ingegneria strutturale, geotecnica e rischio sismico
Ciclo di dottorato: 32
Coordinatore del Corso di dottorato:
nomeemail
Rosati, Lucianorosati@unina.it
Tutor:
nomeemail
Nigro, Emidio[non definito]
Bilotta, Antonio[non definito]
Andreini, Marco[non definito]
Data: 12 Marzo 2020
Numero di pagine: 204
Parole chiave: Probabilistic approach; Fire resistance; Simplified methods; Fire Safety Engineering; Steel Structures
Settori scientifico-disciplinari del MIUR: Area 08 - Ingegneria civile e Architettura > ICAR/09 - Tecnica delle costruzioni
Depositato il: 19 Mar 2020 07:49
Ultima modifica: 10 Nov 2021 09:43
URI: http://www.fedoa.unina.it/id/eprint/13093

Abstract

As alternative to formulation provided in Eurocode 3 for analyses on single members, this thesis shows the development of a probabilistic model to estimate critical temperature to take in account the parameters that influence the behaviour of framed steel structure in fire. A new simplified formulation has been developed to estimate the critical temperature of a steel member in the context of FSE. In other words the simplified model allows to use a general fire curve instead of the ISO standard fire also for single members, because takes into account the nonlinearities of the structure and the indirect actions generally considered with structural analysis through nonlinear FEM models. To develop the probabilistic regression model, results of FEM simulations were used and thirteen parameters were identified to describe the behaviour of the structure in fire, taking in account fluid dynamic aspects. Simple Monte Carlo simulation is not suitable to develop the regression model due to the large variability of the parameters affecting the problem. Therefore, the probabilistic study was conducted using two techniques in order to reduce the size of the sample and to preserve the quality of the results. With the Latin Hypercube Sampling (LHS) we generated a small number of samples with high meaningfulness and with the Simulated Annealing optimization method we achieved diagonal correlation matrix. The analyses were handled through a in house developed software that, starting from sampling data (i) automatically generates the input files for fluid dynamic, thermal and thermomechanical analysis, (ii) performs all analyses, (iii) stores the results in a database and (iv) generates the report file to analyse the results. The probabilistic model, created using the Bayesian updating, has been used to estimate collapse temperature of beams and columns, separately. The significant parameters have been defined through a step deletion process in order to simplify the formulation. Therefore the proposed formulations for the critical temperature take in account the effective stiffness of the structure, the non-linear structural behaviour and indirect actions due to thermal expansions and allow to exploit the resistance reserves due to the internal forces redistribution in steel framed structures. Moreover define the critical temperature as a random variable, to be easily used in reliability analysis and to develop fragility curves.

Downloads

Downloads per month over past year

Actions (login required)

Modifica documento Modifica documento