Shake-table test on a historical masonry aggregate: prediction and postdiction using an equivalent-frame model

Abstract Modeling the seismic response of historical masonry buildings is challenging due to many aleatory and epistemic uncertainties. Additionally, interaction between structural units further complicates predictions behavior unreinforced aggregates found throughout European city centers. This motivated experimental campaign on half-scale, double-leaf stone within SERA (Seismology Earthquake Engineering Research Infrastructure Alliance for Europe)—Adjacent Interacting Masonry Structures project. The included a blind prediction competition that provided participants with data materials, geometry, construction details, input. After test, actual input all recorded processed accelerations, base-shear, displacements were shared participants. Instead single analysis phase, we performed broader stochastic incremental dynamic analyses answer whether common assumptions aggregate modeling either fully coupled or completely separated yield safe behavior. We modeled as equivalent frames in OpenSEES using newly developed macroelement, which captures both in-plane out-of-plane failure modes. To simulate two units, implemented new material model applied it zero-length elements connecting units. Our results demonstrate importance explicitly non-linear connection probabilistic approaches when evaluating response. Although simplifications unit deterministic might produce conservative predicted peak ground acceleration, these simplified overlook likely damage stress calibrating parameters from quasi-static cyclic tests appropriate damping models.