Mechanisms detection by nonlinear finite and distinct element simulations of a historical religious masonry complex

Abstract The detection of collapse mechanisms in masonry structures poses a critical challenge in structural engineering, particularly when dealing with complex historical buildings under seismic loading. Masonry structures exhibit highly non-linear mechanical behaviour due to their composite nature, characterized by discontinuities, weak tensile strength, and anisotropy. Accurately capturing these failure mechanisms, which include cracking, crushing, and sliding along joints, is essential for evaluating their seismic vulnerability. This paper focuses on the mechanical challenges of simulating collapse mechanisms in a masonry historical religious complex, significantly damaged during the 2016 Central Italy earthquake. Nonlinear numerical simulations are carried out to model the structure’s response to seismic loads implementing both the Finite Element Method, based on Concrete Damage Plasticity and the Distinct Element Method, studied using two different approaches: Discrete Element Method (DEM) and the Non-Smooth Contact Dynamics (NSCD). Despite their advanced properties in numerical simulation, neither method can fully capture the complexity of masonry collapse mechanisms. Instead, the combined and controlled use of both Finite and Distinct element methods enhances the predictive accuracy of the simulations. Therefore, this study aims to propose a benchmark approach for damage analysis: through a methodological cross-assessment of their respective displacement behaviours, the time-step activations corresponding to local collapse mechanisms are identified. It is then demonstrated that together, these methods offer a more comprehensive approach to detecting collapse mechanism, with reciprocal compensating for the limitations of the other. This synergistic application is essential to address the inherent complexity of masonry mechanics, including material heterogeneity and non-linear failure progression.