Validation of non-linear equivalent-frame models for irregular masonry walls

This article compares the results obtained from the analyses of seven two-story masonry walls with an asymmetric arrangement of the openings. The walls are modelled with finite elements detailed micromodels and equivalent frame models. The equivalent frame model reliability is evaluated by comparing the results obtained with the two modelling strategies assuming the detailed micromodel's results as reference. Namely, the equivalent frame models of the walls, subjected to nonlinear static analyses, were developed in Scientific Toolkit for OpenSees (STKO) and the OpenSees framework using a macroelement, widely accepted by the literature. The macroelement is based on the idealization of the equivalent frame model with fiber modeling of wall panels to capture bending behavior and a phenomenological law to describe shear behavior. It also provides a model for comparison and validation able to correctly describe the nonlinear behavior of fragile materials that exhibit softening in the response. The micromodeling approach was used as the reference as it is particularly efficient from a computation point of view. The numerical models were validated by comparing pseudo-static experimental tests present in the literature concerning walls with a symmetric arrangement of openings. After the preliminary calibration, the numerical comparisons of the irregular walls generally demonstrate a good correspondence between the equivalent frame model and finite element results. The most relevant differences arise for the walls with masonry panels that exhibit a mixed compression-shear damage mechanism because the simplified frame model cannot capture both behaviors, as it privileges the mechanism that activates first. The differences between the maximum shears, however, range from approximately 1% to 12%.