PSI - Issue 41

Fabrizio Greco et al. / Procedia Structural Integrity 41 (2022) 576–588 Author name / Structural Integrity Procedia 00 (2019) 000–000

586

11

The results show that the curves are approximately similar, thus denoting that the shape of q ( 1 2 , influence the efficiency and the accuracy of the present method. Finally, the second parametric study concerns the auxiliary field adopted in the computation of the M -integral method. Basically, two suitable options are available to analyze dynamic fracture processes: the first deals with the analytical solution developed by Rise concerning an infinite plate affected by a growing crack at a constant velocity (Rice (1979)); the second refers to the classic asymptotic solutions developed by Williams (Williams (1956)). Fig. 7 compares the value of normalized Dynamic Stress Intensity Factors (i.e., K I /K 0 ) extracted using Rise’ and Williams’ asymptotic solutions. The results denote that both auxiliary fields provide comparable predictions, thus highlighting that the choice for the auxiliary field is arbitrary. c c x x ) does not

Fig. 7. Comparison in terms of normalized Mode-I DSIFs achieved using the Rise’ and Williams’ asymptotic solutions as auxiliary fields in the M -integral computations

5

Conclusions

This work has presented a novel modeling strategy for simulating dynamic fracture propagation phenomena in quasi-brittle materials. The proposed method is based on an FE code enhanced by a Moving Mesh strategy consistent with the Arbitrary Lagrangian-Eulerian formulation (ALE). The MM technique changes the computational grid consistently with the evolution of dynamically growing cracks. More precisely, the nodes around the crack tip region are moved according to the conditions prescribed by fracture mechanics’ criteria regarding the direction of propagation and advancing velocity of the crack front. In such a framework, the ALE ensures the consistency of the mesh motions, avoiding excessive distortions for the finite elements of the computational mesh, thus reducing the overall amount of remeshing events. The proposed approach employs the interaction integral method to evaluate Dynamic Stress Intensity Factors, which are essential variables governing the mechanics of dynamic fracture. In particular, the ALE formulation of the M -integral is used, which enables evaluating DSIFs in deforming elements, thus reducing the computational honors considerably. The reliability of the proposed strategy is assessed through comparisons with numerical data and analytical formulations reported in the literature. The results denote that the proposed modeling approach represents a computationally efficient numerical tool for reproducing dynamic fracture phenomena.