[marc.immer]

Hi!

Based on the work of
* Klein, M., Sadiki, A., Janicka, J.: A digital filter based generation of inflow data for spatially developing direct numerical simulation or large eddy simulation
and
* Xie, ZT, and Castro, IP (2008): Efficient Generation of Inflow Conditions for Large-Eddy Simulation of Street-Scale Flows, Flow, Turbulence and Combustion

I have further developed the filtered noise method for turbulence inflow generation, and made a comparison with wind tunnel PIV data.

The code is available on Github: https://github.com/ETH-BuildingPhysi...-OFTools-2.3.X

The validation and comparison is available in my Doctoral Thesis:

Immer, Marc Christian. Time-resolved measurement and simulation of local scale turbulent urban flow. ETH-Zürich (2016). http://dx.doi.org/10.3929/ethz-a-010657618
http://e-collection.library.ethz.ch/...h-49140-02.pdf

Please feel free to fork the repository and further develop the code!
A citation to my thesis is appreciated, and link to the github repository

Cheers!
Marc

[kiyo.matsui]

Hi Marc,

First of all thank-you for releasing this code, also, congratulations on your thesis, it was a really good read.

I have used this inflow generator and ran some test cases of my own.

Overall it worked great.

Only issues I had were pressure spikes at the inlet which I know is an issue mentioned by Xie and Castro. I read that in your code you seem to have fixed this issue, but it may be that I am implementing something wrong.

A few questions: any tips for initialising the LES domain in OpenFOAM?
and do you have any tips on getting inflow statistics for "real" to scale wind profiles other than running the SOWFA pre simulation?

Attached is a "wake interfering flow" case i did.

Again, thanks very much for releasing the code, it is fantastic, and with little modification I got it to work for OF4.

Kiyo

[SH_Zhong]

Originally Posted by kiyo.matsui

Hi Marc,

First of all thank-you for releasing this code, also, congratulations on your thesis, it was a really good read.

I have used this inflow generator and ran some test cases of my own.

Overall it worked great.

Only issues I had were pressure spikes at the inlet which I know is an issue mentioned by Xie and Castro. I read that in your code you seem to have fixed this issue, but it may be that I am implementing something wrong.

A few questions: any tips for initialising the LES domain in OpenFOAM?
and do you have any tips on getting inflow statistics for "real" to scale wind profiles other than running the SOWFA pre simulation?

Attached is a "wake interfering flow" case i did.

Again, thanks very much for releasing the code, it is fantastic, and with little modification I got it to work for OF4.

Kiyo

Hi Kiyo:

I am very interested in this code. Can you share your experiences in the modification to work in OF4. Many thanks!

Zhong

[marc.immer]

Dear Kiyo,

thank you, great to hear that you can use the code.

Sadly, I can't help you that much with specific problems.
For the pressure spikes, I just use a simple mass flow correction, that will eliminate large global pressure fluctiations in the domain.
The relevant code is in void Foam::inflowGenerator::updateCoeffs() after the comment //Mass flux correction.

To reducte the correction magnitude, make sure your domain is large enough to fit a multiple of the biggest scales. The outer domain can be quite coarse.

Please, when you modify the code, always fork from github, then commit your changes to the forked repository (https://help.github.com/articles/fork-a-repo/). This way, future researchers (like Zhong) can benefit from your work. DO NOT modify the code WITHOUT using git, and don't send code around in zip files etc. Use github

Good luck with your work,

Kind Regards,
Marc

[kiyo.matsui]

Hi Marc,

I agree, I will use github, hopefully sometime this week. It's a tiny change, I feel a bit guilty to call my change a contribution as its such a small change but am honored . On a side note, do you have any advice to contributors who want to verify any modified code?. I simply ran the example case and plotted the hairpin vorticies.

Kiyo

[SH_Zhong]

Hi Marc,
Thank you for sharing this code. I have some question about how to use the code. In your phd thesis, the data in the constant/boundaryData/inlet/Lx,Ly,Lz,R,ref are collected in your experiment, and you only need to give two lines points data to generate the turbulence. So if I do not have any experimental data, how can give the boundaryData. For example, to generate a isotropic turbulence.

Many thanks
Zhong

[kiyo.matsui]

Sorry it took some time, the quick fix is found in this link.

https://github.com/kiyoMatsui/ETH-OFTools-2.3.X

I hope this is ok, otherwise please let me know.

Kiyo

[Slanth]

Hi Marc,

Thanks very much for the excellent work. I have several questions, and I am wondering if you can help me.

I am trying to use the FilteredNoiseInflowGenerator. But I only have the experimental mean velocity file at the inlet. I also know that the turbulent intensity is about 5% and the lenght scale is about 3.5mm.
(1) I am wondering how can set the R, and L in your code?
(2) Can I use codeStream with your code for the ref (Mean velocity)? For exmaple

[openfoam-huster]

Hi SH_Zhong,
I have the question with you.Have you solved it?

[openfoam-huster]

Hi Slanth,
I have the question with you.Have you solved it?