Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 75
PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY Edited by: B.H.V. Topping and Z. Bittnar
Paper 104
Numerical Studies on the Ultimate Behaviour of Steel Cold-formed Purlins A. Joó and L. Dunai
Department of Structural Engineering, Budapest University of Technology and Economics, Hungary , "Numerical Studies on the Ultimate Behaviour of Steel Cold-formed Purlins", in B.H.V. Topping, Z. Bittnar, (Editors), "Proceedings of the Sixth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 104, 2002. doi:10.4203/ccp.75.104
Keywords: cold-formed purlin, full-scale test, ultimate behaviour, finite element model, buckling analysis, imperfections, non-linear analysis, advanced design.
Summary
This paper presents the numerical modelling and analysis of steel cold-formed
purlins with the aim to introduce an advanced design method. In the typical roofs of
steel industrial type buildings cold-formed purlins, as secondary structures, are
applied to support the load-bearing elements of the cladding system. The main
characteristic of the behaviour of the typically used C-, Z- or -profiles is coming
from the coupled plate and distortional local buckling and lateral torsional global
buckling modes. These phenomena are highly influenced by the structural
arrangements of the purlin and cladding system [1]. The lateral and torsional
supporting effect of the cladding to the purlin is influenced by several structural
parameters what can be hardly considered without experimental studies. In the
existing design codes significant simplifications and limitations are used to follow
this effect.
The standard design methods are studied and a new advanced methodology is under development in the frame of an ongoing research at the Department of Structural Engineering, Budapest University of Technology and Economics. It is proved that the purlin design method of Eurocode 3 [2] is conservative and benefits can be gained by improving the analysis model [3]. To support the development of the new design method the ultimate behaviour of a special purlin-cladding system is studied by a full-scale experimental program [4]. In the tests the ultimate behaviour of different purlin-cladding arrangements is determined. The obtained failure modes are mainly related to the distortional buckling phenomenon. In this paper the test results are used as experimental background of the finite element model development, what is the bases of the improved design tool. The sophisticated finite element model is developed with the requirements to follow the complex behaviour of the complicated structural arrangements due to the interaction phenomena. The aim of the model on the other hand to be efficient enough to serve not just research but practical design purposes, too. The model is worked out in Ansys FE program with the help of the newly developed Matlab software, called Purlin-FED. The Purlin-FED program is able to build automatically the model, to control the solution and to evaluate the results, in the function of varying parameters as cross section, span, overlap length, bridge system, and anti- sag bars. This program provides with the opportunity to apply the FE program in the background of the design procedure. By the application of Purlin-FED finite element models are built and analysed in parallel with the experimental tests. The model verifications are carried out by the results of linear, buckling and non-linear analyses. The comparison is done by the displacement and stress distribution in the linear range of loading, and by the features of the ultimate behaviour -– buckling shapes, combined failure modes -– for higher load intensity. As a conclusion of the obtained results it is stated that the developed numerical model can follow the behaviour of Z-purlins and can be integrated into the design method. References
purchase the full-text of this paper (price £20)
go to the previous paper |
|