Abstract
Despite the breadth of existing numerical studies on the shear behavior in stiffened plate girders, there has been no comprehensive evaluation on the efficacy of finite element (FE) models for capturing both the local response of the web plate (including out-of-plane deformations and stress contours) and the global response of the stiffened web panel (pertaining to the engagement of transverse stiffeners and flanges). The stages of shear load-displacement can be demarcated by three shear limit states as follows: (1) elastic stage, (2) web yielding stage, and (3) frame (stiffener and flange) engagement stage. This study examines the efficacy of two types of FE models for capturing the response of 28 large-scale shear tests on steel plate girders: the full beam model (FBM), which models the full as-tested specimen; and the extended panel model (EPM), which is proposed as a simplified representation of a single stiffened web panel. In this study, the “shear capacity,” called Vw, occurs when the web plate has developed a local yield mechanism, at which time the slope of the shear load-displacement curve first approaches zero. The FBMs capture all stages of the global response and the local response. The EPMs capture the local web response and Vw, which is defined as a shear capacity; but EPMs struggle to accurately capture the frame engagement stage past Vw due to their reliance on idealized boundary conditions. Nevertheless, the post-Vw response is not relevant to design since the web will have already yielded.
Original language | English (US) |
---|---|
Article number | 120723 |
Journal | Engineering Structures |
Volume | 340 |
DOIs | |
State | Published - Oct 1 2025 |
Externally published | Yes |
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
Keywords
- Finite element model
- Large-scale testing
- Shear capacity
- Shear mechanisms
- Steel plate girder
- Web shear buckling