Evaluation and Comparison of Capability of Finite Element and Discrete Element Methods for Predicting In-plane Behavior of Unreinforced Masonry Wall

In general, masonry structures are divided into two categories in terms of the connection between components: mortar-free masonry, which is called dry joint, and structures that use mortar to connect the components. Due to environmental conditions, when the mortar loses its adhesion over time, the components of the wall can be considered as discrete elements. The poor performance of masonry structures leads to the failure of a significant number of these structures even in moderate earthquakes. In particular, in-plane failure in the walls of masonry structures is very common. The results of laboratory research show that shear failure is very likely in lateral loading. One of the characteristics of this failure mode is its ability to resist high shear force and low ductility, which leads to brittle failure in the wall. Previous numerical studies on masonry structures have been mainly based on the finite element method in which the wall components are considered as continuous elements and the local behavior of the constituent parts of the walls is ignored and causes their real function to be misunderstood. In this paper, first, the specifications of unreinforced masonry wall specimens were determined based on the results obtained from previous experiments, and then numerical models were calibrated and analyzed using two software ABAQUS (finite element software) and 3DEC (discrete element software). Then, behavioral diagrams of walls with different characteristics were extracted under different loading conditions. Finally, a comprehensive comparison between the results obtained from the discrete element and finite element methods was presented. Finally, it was concluded that the discrete element approach provides a more accurate prediction of unreinforced masonry walls than finite element one.