Can we PLEASE get an update on the nonspherical DEM implementation?

MFiX
#1

Dear MFiX team, thanks in advance for your response. Can we please get an update on the nonspherical DEM implementation? Are you going to follow the VT route and use a discrete surface representation or use an analytical function for superquadratics? What is the status of the coupling to the MFiX fluid solver (VT has their own in-house CFD code), and how will particle torque from both collision and fluid be calculated for an “unresolved” (fluid cell>> particle size) implementations? Thank you.

#2

I’ve been working with MFiX and NETL people productively for about 15 years. I would hope I would merit the courtesy of a simple reply. :pensive:

#3

Hi George, sorry for the late update. We are using the analytical function for superquadrics instead of discrete surface representation. The superquadric discrete element method (SuperDEM) has been coupled with the fluid solver. Please check the abstract for more information. We have several papers that are under review, once they come out, I will share with you.

https://aiche.confex.com/aiche/2020/meetingapp.cgi/Paper/600881
https://aiche.confex.com/aiche/2019/meetingapp.cgi/Paper/568832

#4

Thank you for the update :grinning: We are at the crux point of deciding how to proceed with non-spherical particles in DEM-CFD. I have a post-doc coming in mid-Oct who is already very expert in MFiX DEM-CFD. Maybe NETL would consider us to be a ‘test bed’ team member? We have always had a fantastic relationship with NETL and I think back on our times working with Jordan, Bill, Jean and everyone with great fondness. But I also appreciate that things have to proceed with quality first. Thanks again for the informative response.

#5

Thanks for your interest in our work :slightly_smiling_face:. Even though we have made some progress, we are still in the development and validation stages. However, we will add you to our list of beta testers when the time comes (middle-end of 2021 at the earliest).

#6

We are really interested in the combined contributions of viscous stresses (fluid) and granular stresses and their respective contributions to microstructure (coordination number, contact type, anisotropy of granular stress tensor, etc.) especially during the formation and evolution of shape fabric.

Will it be possible to mix superquadric particle shapes and/or sizes of a single shape?

Thanks!

#7

Yes, we will make it just like the traditional DEM, one can add N solid phases, with each solid phase with either different shape or size or density.