Mesh generation of complex three‐dimensional geometries for beam‐particle modeling of fracture

Abstract

In this article, we propose a three‐dimensional meshing algorithm for discrete models combining the lattice approach with the polyhedral particle approach. Our aim is to be able to leverage readily available, well‐supported meshers to handle various geometries. We use them to generate a tetrahedral mesh, which our mesher then converts to a polyhedral mesh. The input mesh serves as geometrical support to generate the nodes of the discrete mesh. The desired shape is obtained with an assembly of convex polyhedral particles—without having to clip them. We show that this approach enables meshing convex or concave geometries with sharp edges, curved features, and more. Several three‐dimensional geometries are presented to support this claim and illustrate the capabilities of the mesher. We provide a detailed analysis of its isotropy. The geometric isotropy is studied by analyzing the orientation of the generated beams. The mechanical isotropy is verified by assessing the properties of the elasticity tensor. Finally, we show that the new mesh retains its ability to be a good support for the generation of realistic cracking patterns.

A circle‐to‐edge contact detection and resolution approach for beam‐particle modeling: cohesive materials and structural mechanics applications

“…This 3-D extension is thus the first perspective of this 2-D development. The set of polyhedral particles can be generated using the 3-D discrete element mesh proposed in Rima et al 41 Whereas this work has only focused on the calculation of normal contact forces, further research, taking into account friction between particles, needs to be undertaken. In the context of crack closure, the calculation of tangential forces is not of primary importance, since a large part of the friction is already accounted for by the rugosity of the crack surfaces due to the irregular shape of the particles.…”

“…This 3‐D extension is thus the first perspective of this 2‐D development. The set of polyhedral particles can be generated using the 3‐D discrete element mesh proposed in Rima et al 41 …”

Computational Engineering Advancements: General Review of Mathematical Modeling in Computer Engineering Applications

“…Global Technological Advancements: In a world increasingly driven by technology, staying updated with the latest advancements in computational engineering is crucial for global competitiveness. This research will contribute to the dissemination of knowledge and foster cross-disciplinary collaboration [17].…”

“…They are used to process and analyse large datasets, identify patterns, and make predictions. Machine learning models, such as support vector machines, decision trees, and neural networks, are applied to solve classification, regression, and clustering problems in diverse fields like image processing, natural language processing, and predictive analytics [17]. 5.…”

“…Researchers have explored statistical models like linear regression, support vector machines, and decision trees for classification and prediction tasks. Furthermore, deep learning architectures, which heavily rely on mathematical frameworks, have shown significant improvements in various computer engineering applications [17]. 5.…”