Buckling experiments on steel silo transition junctions. II: Finite element modeling

Y. Zhao, Jinguang Teng

Research output: Journal article publicationJournal articleAcademic researchpeer-review

14 Citations (Scopus)


This paper is concerned with the finite element modeling of the experiments on cone-cylinder-skirt-ring transition junctions in steel silos under simulated bulk solid loading presented in the companion paper. Before presenting the finite element results, the issue of modeling the interaction between the stored solid and the shell wall throughout the loading process is first examined. Results from nonlinear bifurcation analyses using the perfect shapes and nonlinear analyses using the measured imperfect shapes are then presented and compared with the experimental results. These comparisons show that despite the structural complexity of steel silo transition junctions, their behavior can be satisfactorily predicted by finite element analyses taking into account a number of important factors including geometric imperfections, effects of welding and the interaction between the junction and the stored solid. Next, the paper presents results of nonlinear analyses of these junctions with assumed eigenmode-affine imperfections. These results shed considerable light on the effect of ring buckling on the load-carrying capacity of transition junctions. Finally, the implications of the experimental and finite element results for the design of steel silo transition junctions are discussed.
Original languageEnglish
Pages (from-to)1803-1823
Number of pages21
JournalJournal of Constructional Steel Research
Issue number12
Publication statusPublished - 1 Dec 2004


  • Buckling
  • Experiments
  • Finite element modeling
  • Imperfections
  • Postbuckling
  • Shells
  • Stability
  • Steel silos
  • Transition junctions

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • General Materials Science


Dive into the research topics of 'Buckling experiments on steel silo transition junctions. II: Finite element modeling'. Together they form a unique fingerprint.

Cite this