Empirical Equations for the Direct Definition of Stress–Strain Laws for Fiber-Section-Based Macromodeling of Infilled Frames

Abstract

Equivalent strut macromodels are largely used to model the influence of infill walls in frame structures due to their simplicity and effectiveness from a computational point of view. Despite these advantages, which are fundamental to carrying out seismic simulation of complex structures, equivalent struts are phenomenological models and therefore have to conventionally account for the influence of really large amounts of geometrical and mechanical variables with a relatively simple inelastic response. Mechanical approaches, generally used to evaluate the force–displacement curve of a strut, are based on hypothesizing the damage mechanism that will occur for an infill–frame system subject to lateral forces. This assumption has a great impact on the response of the system and is affected by large uncertainties that also propagate in seismic assessment results. Based on the aforementioned unknowns, this paper proposes a new stress–strain relationship to be used for fiber-section modeling of equivalent struts. The definition of the stress–strain model is based on an empirical approach rather than a mechanical one. The four parameters defining the stress–strain law (peak stress, peak strain, ultimate stress, and ultimate strain) are directly linked to geometrical and mechanical features of an infilled frame through analytical correlation laws, which are determined from an experimental data set enlarged with data from refined finite-element (FE) simulations. The analytical correlations provided in the paper are proposed as a tool for direct modeling of an infilled frame. Validation tests were carried out with experimental results different from those used to build the data set. The extension of the model to the prediction of cyclic behavior is finally proposed.