Behavior of Q690 high-strength steel columns: Part 2: Parametric study and design recommendations

Tian Ji Li, Si Wei Liu, Guo Qiang Li, Siu Lai Chan, Yan Bo Wang

Research output: Journal article publicationJournal articleAcademic researchpeer-review

38 Citations (Scopus)

Abstract

Prediction of the column strength, associated with buckling and material yielding, is vital in a successful design. In this paper, a completely numerical analysis technique is employed for simulating the compressive behavior of Q690 high-strength steel columns with box- and H-sections, and its accuracy is verified by the experimental results as presented in the part 1 of the companion papers. To this, the curved Pointwise-Equilibrating-Polynomial (PEP) is introduced for representing the column deformations with an explicit simulation on the member initial imperfection. Further, the refined plastic hinge model combined with the use of sectional yield surfaces is employed to reflect inelastic yielding at the critical location. A cross-sectional analysis approach allowing for residual stresses modeling is also incorporated into the numerical method. Sequentially, a parametric study is conducted extensively for the Q690 high-strength steel welded columns. At last, the buckling curves from Chinese (GB 50017-2003), European (Eurocode 3), and American (ANSI/AISC 360-10) codes are examined, and some recommendations for design of Q690 high-strength steel columns are presented.
Original languageEnglish
Pages (from-to)379-394
Number of pages16
JournalJournal of Constructional Steel Research
Volume122
DOIs
Publication statusPublished - 1 Jul 2016

Keywords

  • Design
  • High-strength steel
  • Imperfections
  • Numerical analysis
  • Residual stress
  • Welded sections

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanics of Materials
  • Metals and Alloys

Fingerprint

Dive into the research topics of 'Behavior of Q690 high-strength steel columns: Part 2: Parametric study and design recommendations'. Together they form a unique fingerprint.

Cite this