Interior CFD influence by exterior pressure coefficients

Hi there,

I was wondering if there is a way to combine the effect of an exterior CFD with a series of obstructions in an urban context and an interior CFD influenced by the exterior CFD wind pressure coefficients in the building facades.
Have you done this before? how did you address it? is it required to interpret and modify by hand the discharge coefficient for every window? or the wind factor per facade?

Thanks and sorry for my ignorance in this topic

Hi again,
Did somebody face this before? How did you address it?

Thanks

Hi @julioamodia89 I have done this before using SimScale. Honestly, it’s one of the coolest things I have done. My workflow was something like this:

  1. Upload a microclimate like geometry, site, contextual buildings and topology if necessary.
  2. Run an external simulation in the multi-directional PWC analysis. I assume the pressure at each opening to be uniform, so I read the pressure at the centre of each opening.
  3. Convert all the pressures from all openings tp pressure coefficients. Usually, I have a row for each opening and a column for each direction, as a table.
  4. Import a geometry representing an inside space (room, floor etc), with clear surfaces for openings. These will be used as the boundary conditions. If you want to include openings and resistances from the window geometry, I include a volume to apply what we call a porous media. You can define it indirectly using the estimated discharge coefficient.
  5. For each scenario you wish to simulate, create a new simulation (CHTv2), defining the correct pressure values at the opening boundaries, I.e. you want to simulate an indoor space under natural ventilation with 15 degrees ambient temperature and wind at 5m/s from the southwesterly wind direction? Use the pressure values back-calculated from the pressure coefficient in your table to define it.
  6. Observe the flow fields inside the building from the results.

This does assume a few things. For example, the openings are small, and flow through the building is not that high due to internal resistances. If this was an ample open space, for example, a rooftop garden with a cover, then a different approach should be taken.

Also, if you wanted to be fancy, you could use the API to automate a lot of this or even write something natively as grasshopper components.

If you want to see our example on this its here for the indoor study or here for the external study (Login required, you should be able to sign up to a non-commercial account and get access).

I hope this helps! I would also be interested to know your solution if you already managed it.

1 Like

Both your links require a login.

Ah true, you can sign up for a community account, or alternatively, let me know what you would like to see and I can screenshot it for you?

I edited my original post to caution signup is required.

Why don’t you link to a blog post?

Downhill Skateboarding World Record Holder Simulation | SimScale

That is a good question/statement. Let me see what is the most relevant and post that. Bare with.

1 Like

Here is a blog post on the same project. I think it’s a good overview, but if you need details, the project links are still relevant above. Have a good day!

Thanks for the update on the link. I actually gave up on trying any sort of CFD myself and I’m happy to have this done by a specialist who’s capable of judging the results.

Anyway, I’m looking at many different topics on this forum and recently there’s been quite some spam with funny links. So that’s the main reason I posted here…

That is unfortunate about your experiences with CFD, but great you found a solution/person. We are aiming to put this in the hands of the less experienced, so if you ever fancy trying it yourself, you would be welcome! Likewise, it is a shame you are getting lots of spam here, the links are genuine and do lead to real projects, but I understand your concern, we have a SimScale forum also, and on occasion, similar issues.

Best,
Darren

1 Like

Hi @dlynch
I have two questions about calculating Cp values for openings:

  1. I understand the assumption of a uniform pressure distribution over each window. However, as someone not well-versed in this, I’d like to know how to calculate an average Cp when the pressure isn’t uniform. Is it correct to calculate Cp individually for each point on the opening and then find an average of them?
  2. I’m a bit confused about the Uref value, as I’ve come across conflicting information in different papers. When calculating Cp for natural ventilation, is it acceptable to set Uref to different values for each wind direction based on the weather file, or should it be a fixed value like 10 m/s, as suggested in some sources?

I’d greatly appreciate any help or suggestions to clear up these questions.
Shadab

Hi @Shadab_Mahdiyar , certainly.

  1. Since we are talking CFD, most CFD codes can take a surface/area-weighted average. This is the fairest way of evaluating it. I don’t think other methods will give you drastically different results unless you have some extreme values across the opening, but either way, the area-weighted average would ensure that the extreme values are only represented with weighting on how much of the window they cover.
  2. This is a good question, but I think this is broken down into two parts: are we measuring Cp, or are we calculating pressure from Cp? In the first part, we probably run a CFD for each direction, probably with a reference wind speed that is uniform. Usually, we take the wind speed at the key building height; however, as long as you are consistent throughout, it does not matter. Then, when working out the pressures for the internal flow, we take the wind speed and direction for the scenario (from the weather data) at the same height we calculated the Cp values for. And remember, when talking about wind speeds at different heights, this is the undisturbed wind speed, i.e. the speed at the height in the atmospheric boundary layer.

hope that helps,
Darren