Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
2-way FSI model with automatic re-meshing stops with inverted mesh elements
Posted 2 mai 2017, 12:35 UTC−4 2 Replies
Please login with a confirmed email address before reporting spam
Hello everyone,
I am trying to simulate the expansion of a filled magma chamber due to the injection of new magma (density= 2300 kg/m^3, dynamic viscosity= 1e5 Pa*s). I am interested in studying the flow of the magma in the chamber while it expands the chamber walls and deform the surrounding elastic material (Young´s modulus=30e9 Pa, density=2400 kg/m^3, and Poisson´s ratio=0.25).
This is a 2-way FSI problem. I am using an axisymmetric domain (see image) and the automatic re-meshing technique mainly to improve the mesh of the fluid of the magma chamber, subjected to a strong expansion.
The analysis always stops due to inverted mesh elements (the inverted element are those along the fluid-solid interface), no matter if I improve the mesh with smaller elements at the interface or if I increase the time steps. I reduced the tolerance to 0.0001, but doesn´t help. I can control the quality of the mesh by using the mesh quality as condition for re-meshing. I set the threshold for the minimum mesh quality at 0.5, but it doesn´t help. The solution doesn´t converge (and I still get inverted mesh elements) even if the mesh doesn´t look bad at all.
Does anyone have any idea about how to improve the model to solve this problem, or about what could be the reason for the elements' inversion?
Any help is very appreciated!
Thank you,
Erika
I am trying to simulate the expansion of a filled magma chamber due to the injection of new magma (density= 2300 kg/m^3, dynamic viscosity= 1e5 Pa*s). I am interested in studying the flow of the magma in the chamber while it expands the chamber walls and deform the surrounding elastic material (Young´s modulus=30e9 Pa, density=2400 kg/m^3, and Poisson´s ratio=0.25).
This is a 2-way FSI problem. I am using an axisymmetric domain (see image) and the automatic re-meshing technique mainly to improve the mesh of the fluid of the magma chamber, subjected to a strong expansion.
The analysis always stops due to inverted mesh elements (the inverted element are those along the fluid-solid interface), no matter if I improve the mesh with smaller elements at the interface or if I increase the time steps. I reduced the tolerance to 0.0001, but doesn´t help. I can control the quality of the mesh by using the mesh quality as condition for re-meshing. I set the threshold for the minimum mesh quality at 0.5, but it doesn´t help. The solution doesn´t converge (and I still get inverted mesh elements) even if the mesh doesn´t look bad at all.
Does anyone have any idea about how to improve the model to solve this problem, or about what could be the reason for the elements' inversion?
Any help is very appreciated!
Thank you,
Erika
Attachments:
2 Replies Last Post 4 mai 2017, 04:47 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hi Erika,
Solid Mechanics does not lend itself well to automatic remeshing, which is why it is not recommended to use that with FSI.
However, the moving mesh feature with help lines for mesh deformation can be used rather successfully. See this blog post for more info:
www.comsol.eu/blogs/model-translational-motion-with-the-deformed-mesh-interfaces/
You might also find this post interesting:
www.comsol.eu/blogs/deformed-mesh-interfaces-rotations-and-linear-translations/
Niklas
Solid Mechanics does not lend itself well to automatic remeshing, which is why it is not recommended to use that with FSI.
However, the moving mesh feature with help lines for mesh deformation can be used rather successfully. See this blog post for more info:
www.comsol.eu/blogs/model-translational-motion-with-the-deformed-mesh-interfaces/
You might also find this post interesting:
www.comsol.eu/blogs/deformed-mesh-interfaces-rotations-and-linear-translations/
Niklas
Hello everyone,
I am trying to simulate the expansion of a filled magma chamber due to the injection of new magma (density= 2300 kg/m^3, dynamic viscosity= 1e5 Pa*s). I am interested in studying the flow of the magma in the chamber while it expands the chamber walls and deform the surrounding elastic material (Young´s modulus=30e9 Pa, density=2400 kg/m^3, and Poisson´s ratio=0.25).
This is a 2-way FSI problem. I am using an axisymmetric domain (see image) and the automatic re-meshing technique mainly to improve the mesh of the fluid of the magma chamber, subjected to a strong expansion.
The analysis always stops due to inverted mesh elements (the inverted element are those along the fluid-solid interface), no matter if I improve the mesh with smaller elements at the interface or if I increase the time steps. I reduced the tolerance to 0.0001, but doesn´t help. I can control the quality of the mesh by using the mesh quality as condition for re-meshing. I set the threshold for the minimum mesh quality at 0.5, but it doesn´t help. The solution doesn´t converge (and I still get inverted mesh elements) even if the mesh doesn´t look bad at all.
Does anyone have any idea about how to improve the model to solve this problem, or about what could be the reason for the elements' inversion?
Any help is very appreciated!
Thank you,
Erika
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Niklas,
thank you very much. I will try to apply one of those strategies to my model.
Erika
thank you very much. I will try to apply one of those strategies to my model.
Erika