PSI - Issue 66

10

Author name / Structural Integrity Procedia 00 (2025) 000–000

Domentico Ammendolea et al. / Procedia Structural Integrity 66 (2024) 350–361

359

Fig. 6. A depiction of the phase field distribution (  ) corresponding to the peak force of Fig. 5-a.

In the beam under investigation, these areas can be identified as the zones at 1/4 L b and 3/4 L b , which are characterized by small values of the phase field variable (  ), as observed from the results of DA, and for which the behavior can be considered as linearly elastic. This is confirmed by the fact that the extension of the coarser region for  = 1.4 is more significant than that obtained for  = 1 and  = 1.2. 5 Conclusions This work presents a new adaptive concurrent multiscale model for the failure analysis of masonry structures. The proposed model adopts a domain decomposition strategy, whose key idea is to assign to the computational domain different resolutions ( i.e. , different scales) of material according to the evolution of the material's constitutive behavior. More precisely, the computational domain is divided into coarser and fine regions. In coarser regions, the microstructure of the masonry is replaced by an equivalent linear elastic homogenized material, whose description of constitutive behavior is obtained by performing numerical homogenization methodology. Specifically, a Repeating Cell (RC) for the masonry is identified, and a numerical homogenization is performed by assuming periodic boundary conditions. In finer regions, a micro-modeling approach is used, representing masonry as a two-phase material formed by the combination of brick units and mortar joints. Such a representation of the masonry ensures an accurate reproduction of the failure mechanisms. In particular, the Phase Field Cohesive Zone Model (PF-CZM) is used to reproduce the failure mechanisms at the microscale level. In the proposed strategy, the computational domain is progressively refined during the numerical simulation until damage mechanisms start to affect coarser regions. More precisely, finer regions are activated according to the conditions dictated by an activation criterion. Comparative results between direct micromechanical analysis and the proposed multiscale model have demonstrated that the latter is highly accurate in replicating the failure mechanisms of masonry structures and computationally efficient. This confirms that the proposed methodology provides a powerful tool for the detailed analysis of masonry failure, offering significant advantages in terms of both precision and computational cost.