Questions and Resources for Transition modelling

HSK_ · September 21, 2020, 05:29

Hello everyone,

I am an undergraduate student and I'm planning on simulating a 2-D Kelvin-Helmholtz instability as a personal project.
My experience with CFD is using Fluent and writing simple Finite Difference codes on python. But I have a pretty good understanding of the mathematics of numerical methods.
Unfortunately I am unable find any suitable resources that would help me go about transition. Could someone direct me to resources for the same.

1) Should I go for a simulation on openfoam? or are there better open source softwares available for transition? (On openfoam Re-theta -gamma is the only transition model afaik)
2) Should I write my own code on python/Matlab?
3) What transition models should I use? I understand DNS and LES would give the best results but since DNS is computationally expensive. I was planning on going for LES. I have a i7 9th gen 6 core laptop with 16 gb of ram. Would that be good enough for a good 2-D LES ?
4) How would my results be validated ?

Moreover what I expect to be the learning outcomes are to be able to write/edit code for my usage, be able to solve in parallel and get a good understanding of the physics of the problem.
Suggestions to model different cases would be welcome. Since I'm new to it, I don't know much.

Thank you

pcosta · October 3, 2020, 19:08

Hi,

You can actually do a DNS on your laptop if you have a numerical algorithm (and implementation) that is fast enough. In particular, for a 2D problem it should be no problem to run it on a laptop. As for which tool to use, I think it does not matter so much for a simple 2D mixing layer to study the KH instability.

One tool that allows you for instance to simulate a turbulent channel flow on your laptop with is the DNS code CaNS (https://github.com/p-costa/CaNS). You can also easily use it for a 2D KH instability, as long as you implement the initial condition yourself

.

Disclaimer: I am the author of CaNS

.

The nature of the KH instability is inviscid and your results can be directly validated against linear stability theory (perhaps you can get access to the book of Darzin).

September 21, 2020, 05:29	Questions and Resources for Transition modelling	#1
HSK_ New Member Join Date: Sep 2020 Posts: 1 Rep Power: 0	Hello everyone, I am an undergraduate student and I'm planning on simulating a 2-D Kelvin-Helmholtz instability as a personal project. My experience with CFD is using Fluent and writing simple Finite Difference codes on python. But I have a pretty good understanding of the mathematics of numerical methods. Unfortunately I am unable find any suitable resources that would help me go about transition. Could someone direct me to resources for the same. 1) Should I go for a simulation on openfoam? or are there better open source softwares available for transition? (On openfoam Re-theta -gamma is the only transition model afaik) 2) Should I write my own code on python/Matlab? 3) What transition models should I use? I understand DNS and LES would give the best results but since DNS is computationally expensive. I was planning on going for LES. I have a i7 9th gen 6 core laptop with 16 gb of ram. Would that be good enough for a good 2-D LES ? 4) How would my results be validated ? Moreover what I expect to be the learning outcomes are to be able to write/edit code for my usage, be able to solve in parallel and get a good understanding of the physics of the problem. Suggestions to model different cases would be welcome. Since I'm new to it, I don't know much. Thank you

October 3, 2020, 19:08		#2
pcosta New Member Pedro Costa Join Date: Jul 2017 Posts: 9 Rep Power: 9	Hi, You can actually do a DNS on your laptop if you have a numerical algorithm (and implementation) that is fast enough. In particular, for a 2D problem it should be no problem to run it on a laptop. As for which tool to use, I think it does not matter so much for a simple 2D mixing layer to study the KH instability. One tool that allows you for instance to simulate a turbulent channel flow on your laptop with is the DNS code CaNS (https://github.com/p-costa/CaNS). You can also easily use it for a 2D KH instability, as long as you implement the initial condition yourself . Disclaimer: I am the author of CaNS . The nature of the KH instability is inviscid and your results can be directly validated against linear stability theory (perhaps you can get access to the book of Darzin). HSK_ likes this.