Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 91
PROCEEDINGS OF THE TWELFTH INTERNATIONAL CONFERENCE ON CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING COMPUTING
Edited by: B.H.V. Topping, L.F. Costa Neves and R.C. Barros
Paper 18

The Effect of Local Geometric Imperfections on the Non-Linear Analysis of Stub Columns

M.M. Pastor, M. Casafont, F. Roure and M. Ferrer

Department of Strength of Materials and Structural Engineering, School of Industrial Engineering of Barcelona (ETSEIB), Universitat Politècnica de Catalunya (UPC), Barcelona, Spain

Full Bibliographic Reference for this paper
M.M. Pastor, M. Casafont, F. Roure, M. Ferrer, "The Effect of Local Geometric Imperfections on the Non-Linear Analysis of Stub Columns", in B.H.V. Topping, L.F. Costa Neves, R.C. Barros, (Editors), "Proceedings of the Twelfth International Conference on Civil, Structural and Environmental Engineering Computing", Civil-Comp Press, Stirlingshire, UK, Paper 18, 2009. doi:10.4203/ccp.91.18
Keywords: geometric imperfections, local buckling, cold-formed members, pallet-rack profiles, non-linear analysis, finite element.

Summary
This research focuses on short columns subject to compression. The purpose of this paper is to analyse sensitivity to different geometric imperfection magnitudes of the non-linear finite element analysis. Twenty open cold-formed steel sections have been analysed, with a nominal thickness range of 1.0 to 2.5 mm.

Experimental tests on the 20 sections have been carried out in our laboratory. As a general rule, five tests are needed for each section in order to find the position of the effective centre of gravity [1].

As local buckling usually occurs within tested lengths, the typically recommended geometric imperfection values for this mode have been applied to the finite element models. Three different magnitudes have been analysed: w/200, 0.006w, and w/100.

The study is carried out with the intention of seeing how the finite element analyses are affected depending on which imperfection value has been chosen.

A parametric finite element model has been created [2]. On a particular node of the outer face of both load plates, the displacements are prescribed. The material behaviour has been reproduced by means of an elastic-plastic bi-linear model. In the first step, an elastic (linear) buckling analysis was carried out on a perfect mesh to obtain its deformed shape (eigenvalue) in the critical buckling mode. This deformed shape indicates the possible buckling mode of the specimen and is introduced as a starting point for the second step: the non-linear analysis. The displacement is increased in successive increments until the maximum load is reached and it clearly begins to decrease or remains constant for a significant increment of displacement. The simulation is repeated, displacing the load node along the symmetry line in increments of 1 mm. The following conclusions have been drawn from this study:

  1. The failure mode of the member does not always match up with the buckling mode shape.
  2. Buckling mode shapes (eigenvectors) should be split into pure local and distortional modes.
  3. Higher buckling modes should be investigated.
  4. Both w/200 and 0.006w imperfections, defined in terms of the web, have proved to be appropriate for local buckling.
  5. The ultimate load experimentally obtained is always higher than the one obtained using the finite element analysis, except for three sections.
  6. If the failure mode is local, the imperfection magnitude has little effect on the non-linear analysis results.

References
1
F. Roure, M. Casafont, M.M. Pastor, M.R. Somalo,"Effective cross-sectional area of steel thin-walled U shaped sections, obtained by different methods", 4th European Conference on Steel and Composite Structures, June 2005.
2
M. Casafont, "Behaviour of perforated cold-formed sections subject to combined compression and bending", PhD Thesis, UPC (Universitat Politècnica de Catalunya), 2003 (in Spanish).

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)