Simulating Local Buckling-Induced Softening in Steel Members Using an Equivalent Nonlocal Material Model in Displacement-Based Fiber Elements

Abstract

Fiber-based elements are commonly used to simulate steel beam-columns, due to their ability to capture P-M interactions and spread-of-plasticity. However, when mechanisms such as local buckling result in effective softening at the fiber-scale, conventional fiber models exhibit mesh dependence. To address this, a two-dimensional nonlocal fiber-based beam-column model is developed and implemented numerically. The model focuses on hot-rolled wide flange (W-) sections that exhibit local buckling-induced softening when subjected to combinations of axial compression and flexure. The formulation up-scales a previously developed nonlocal formulation for "single-fiber" buckling to the full frame element. The formulation incorporates a physical length scale associated with local buckling, along with an effective softening constitutive relationship at the fiber level. To support these aspects of the model, 43 Continuum Finite Element (CFE) test-problems are constructed. These test-problems examine a range of parameters including the axial load, cross-section, and moment gradient. The implemented formulation is validated against CFE models as well as physical steel beam-column experiments that exhibit local bucklinginduced softening. The formulation successfully predicts post-peak response for these validation cases in a mesh-independent manner, while also capturing the effects of P-M interactions and moment gradient. To enable convenient generalization, guidelines for calibration and selection of the model parameters are provided. Limitations are discussed along with areas for future development.

Proposed Updates to the ASCE 41 Nonlinear Modeling Parameters for Wide-Flange Steel Columns in Support of Performance-Based Seismic Engineering

“…In that respect, columns experiencing varying axial load and lateral drift demands may be modeled based on the procedures outlined in this paper considering only the gravity-induced axial load ratio, B / "# and neglecting the transient effects. Ideally, numerical models that explicitly capture the axial force-bending interaction within the cross-section should be employed for this purpose (e.g., Krishnan 2010;Suzuki and Lignos 2017;Do and Filippou 2018;Kolwankar et al 2018). Global instability modes shall also be considered within a simulation framework.…”

Proposed methodology for building‐specific earthquake loss assessment including column residual axial shortening

“…These could either be continuum finite‐element models or fiber‐based models with effective stress‐strain formulations that trace softening over a buckling length (eg). Nonlocal formulations have also been proposed to tackle the issue of spurious mesh dependency in stress‐strain formulations with softening . Alternatively, empirical formulations can facilitate the computation of Δ axial of a steel column .…”

“…While deterioration models that enable explicit quantification of column axial‐shortening have evolved, the current state of the art in vulnerability assessment of structures employs point‐hinge deterioration models (eg,). Although these models may efficiently predict story‐based EDP s and inform earthquake‐induced risk and loss assessments, they cannot explicitly appraise local EDP s of interest.…”

Experimental Evaluation and Numerical Modeling of Wide-Flange Steel Columns Subjected to Constant and Variable Axial Load Coupled with Lateral Drift Demands

“…10b. To the best of the authors knowledge, the data presented herein is unique and can facilitate the calibration of mechanics-based numerical models that explicitly capture axial stiffness degradation as well as column axial shortening (Suzuki and Lignos 2017;Do and Filippou 2018;Kolwankar et al 2018). Figure 11a shows the measured plastic hinge length, LPH, for all the column specimens.…”