Access Type

Dissertation/Thesis

Date of Award

January 2019

Degree Type

Thesis

Degree Name

M.S.

Department

Civil and Environmental Engineering

First Advisor

Fatmir . Menkulasi

Abstract

This study presents a numerical investigation of the out-of-plane buckling behavior of trapezoidal singly symmetric free-standing circular arches with a hollow cross-section. The behavior of such arches is investigated in terms of the critical elastic buckling load and ultimate load-carrying capacity when such arches are subject to a radial pressure that creates axial compression and no bending in the arch. All investigations are conducted using validated shell finite element simulations. A parametric study is conducted to evaluate the influence of arch depth, wall thickness, subtended angle, arch radius, yield strength, initial imperfections, and residual stresses. Numerically obtained critical elastic buckling loads and ultimate load-carrying capacities are compared with existing mathematical formulations derived for arches with an I-cross-section proposed by Pi and Trahair (1998) and Bradford et al. (2018), and modified provisions for compression members provided in Eurocode 3 and Australian Standard AS4100. Numerically obtained critical elastic buckling loads match well with those obtained using the Euler column buckling equation featuring a computed over predicted average load ratio of 1.07 and COV of 6 %. However, when numerically obtained critical elastic buckling loads are compared with the loads obtained from the equations proposed by Pi and Trahair (1998) and Bradford et al. (2018) for arches with an I-cross section, the average computed over predicted load ratio is 1.33 and the COV is 11%. The failure mode of all arches was out-of-plane inelastic buckling. Arch capacities are compared once using existing formulations for critical elastic buckling loads and another time using numerically obtained critical elastic buckling loads as an input. When existing formulations for the critical elastic buckling load are used, it is concluded that, in general, Eurocode column curve “c” provides more accurate predictions compared to column curve “b” and “d” featuring an average ratio of computed over predicted capacities of 0.99 and a COV of 16%. Provisions for compression members in AS4100 feature a computed over the predicted average ratio of capacities of 0.95 and a COV of 14%. Provisions proposed by Trahair et al. (1998) for arches with an I-cross-section feature an average ratio of computed over predicted capacities of 1.14 and a COV of 14%. When numerically obtained critical elastic buckling load is used, it is concluded that, in general, Eurocode column curve “b” provides more accurate predictions compared to column curve “c” and “d” featuring an average ratio of computed over predicted capacities of 0.98 and a COV of 14%. Provisions for compression members in AS4100 feature a computed over the predicted average ratio of capacities of 0.95 and a COV of 14%. Provisions proposed by Trahair et al. (1998) for arches with an I-cross-section feature an average ratio of computed over predicted capacities of 0.90 and a COV of 15%.

Download

Share

COinS