idaholab
regular square lattice mesh in circular domain for solving Peridynamics problem
I am referring this journel "A MOOSE-Based Implicit Peridynamic Thermomechanical Model" https://doi.org/10.1115/IMECE2016-65552 . One of the problem discussed in this paper is as follows: " A circular domain with radius of 4.1 mm is used to represent a 2D cross section of a ceramic light water reactor fuel pellet. A regular square lattice is used for discretization, which is commonly used for two-dimensional peridynamics models. The discretization scheme is used here has a material point spacing delta= 0.0412 ,"
I am unable to create this square lattice mesh in a circular domain using the available options in moose. The picture they have shared in this paper is attached below. Please suggest a way to create this.
#6468
In the future, these kinds of questions are better fits for https://github.com/idaholab/moose/discussions
The peridynamics module now only accepts finite element mesh. The mesh used in the manuscript was generated using a GeneratedMeshGenerator (UPDATE: PD version) class that is now not available in MOOSE.
The peridynamics module now only accepts finite element mesh. The mesh used in the manuscript was generated using a GeneratedMeshGenerator class that is now not available in MOOSE.
Sir, the GeneratedMeshGenerator is available. With this, I can have a regular mesh on a rectangular domain. But my domain is circular and I need the regular square mesh. Though 'AnnularMesh' class gives a symmetric mesh, I cannot use it because the size of the element varies. If I am using GeneratedMeshGenerator and make a square domain of side length 8.2mm, is there a way to trim away those elements such that I can get a Circular domain of diameter 8.2mm as shown in the manuscript.
There was a peridynamics version of GeneratedMeshGenerator class for peridynamics. That one was longer useful since we can take any finite element mesh and convert it for use in the peridynamics modeling and simulation.
What I'm hearing is that if @Narayanan1999 was to use the input you used in your paper @hchen139 that it would no longer work with current MOOSE? If so, I don't really think this is a MOOSE issue. Software naturally evolves over time. We could create a new Discussion about what current objects to use to reproduce the behavior from the input used for that paper, but I think we can probably close this issue