Post processing: the return link from Butterfly to Honeybee about butterfly HOT 14 CLOSED

Hello everyone,

Just giving a small heads up on what we have been testing with Chris these last few weeks. The data interface between CFD and LB/HB was after all much simpler than anticipated. The file format we selected was simply csv, which is easily extracted by Paraview. Chris has since then developed/updated a couple of amazing components. This now allows us to do two important analyses:

• outdoor thermal comfort, using Chris's known workflow and with the addition of actual CFD simulated wind speeds. Chris is targeting to include the same analysis (atm done using UTCI) to PET. This is quite exciting, since we can already see a big difference on studies based on climatic vs CFD data.
• pedestrian/outdoor wind comfort. A nifty, little component, takes wind factors for each point (previously or later linked to a point in the viewfactormesh), calculated in a wind comfort study and extracted from Paraview, and uses that information to create a prediction of yearly wind velocities patterns. The component produces 8760 values of wind velocities for that specific point, at the chosen height, and within its specific geometrical context and landscape. This analysis is actually very fast (a matter of 1-2 minutes for 1000 points) and allows the user to conduct pedestrian wind comfort studies, making the calculation of wind exceedance probabilities an easy exercise. I think it is actually such a good component that I hope it is extended to include the wind comfort categories.

Quite interesting stuff I think and a nice showcase on how Butterfly would really link so well with the beast that is LB/HB. More soon I hope!

Kind regards,
Theodore.

from butterfly.

Very exciting stuff! I'm keeping an eye on all your posts and still waiting for couple of days to spend on butterfly. The other workflow that you have found looks promising #13

from butterfly.

Hello,

this is very interesting. We have been working on the exact same workflow in our office. Getting the wind data into GH via a .csv file and use a factor for each grid.

I am very interested in the method you used in ladybug. We run the results for a single day because we worked with a much larger grid and there were some complications with the indices component. Correct me if I am missing something but we could not run the analysis for the whole grid using the Ladybug_Thermal_Comfort_Indices component. We had to "trick" the component to do so, using the number of the grid points instead of 8760 hours.

It will be very interesting to see the method you used for the annual run, which seems very well optimized in order to run so fast.

keep up the nice work.

kind regards,
Byron

from butterfly.

I agree that this is all very exciting!

Byron,
I just added a microclimate map recipe for PET last night as I realized that both Theodore and a number of other people were interested in interacting with the PET model spatially:

Example file is here:
https://www.dropbox.com/s/qb93j78f8d22zj3/PET_Microclimate_Map2.gh?dl=0

I should have made this microclimate map recipe component a while ago. @stgeorges had put his code for calculating PET into Ladybug_Ladybug so many months ago in order to facilitate this.

Let use know if you have any thoughts or feedback.
-Chris

from butterfly.

@TheodoreGalanos , @chriswmackey ,

Is it possible to get those beta components related with importing cfd wind values to grasshopper, please? If not publicly, then by email? Thank you in advance.

@Vyrwn

Correct me if I am missing something but we could not run the analysis for the whole grid using the Ladybug_Thermal_Comfort_Indices component. We had to "trick" the component to do so, using the number of the grid points instead of 8760 hours.

It's certainly is. You just had to define the particular HOY_ for which you would like to conduct your analysis.

from butterfly.

@stgeorges

First of all thank you very much for these components they are very much appreciated and useful. The issue is that creating a map as Chris is showing above, is not possible by defining the HOY_ if we want to test more than one grid point. We did not have the component above before so we did a work-around by adding a Grid_Point input that if !=none then it overrides the hours of the year to the number of the grid points and runs a "fake annual" simulation instead.

@chriswmackey thank you very much for the link, I will have a go to see how it works.

-Byron

from butterfly.

Hi @Vyrwn ,
I would be very grateful if you could send me the particular .gh definition which replicates the situation you are describing. Thank you.

from butterfly.

Hi @stgeorges ,

I created a file with a small grid just to show what my issue was.

https://www.dropbox.com/s/ciys1spt6prqy0l/PET_on_Grid.gh?dl=0

I use generic data for Temp, Humidity and Wind just to make it simple. Let me know your thoughts and if there is something I am doing wrong.

-Byron

from butterfly.

@stgeorges you can find the files in this discussion. They are both parsing the .csv file.

from butterfly.

@Vyrwn ,

Glad that you find the PET recipe component useful and I agree that you should be able to use the Thermal Indices component to make simple outdoor comfort maps if you are not interested in accounting for surface temperatures similar to this simple UTCI map.

We will find some way to make this possible.

-Chris

from butterfly.

@chriswmackey,

Yes I agree that mapping PET or other indices on a grid is quite important and a much easier way of conveying the message. The process we follow is exactly the same as Theodore and you are describing above. Get the wind data from OpenFoam, then run radiation on the same grid points and then calculate comfort with some slight modifications that we did and we are still looking into.

As I said before, the main issue was the simulation of all grid points at the same time. For this I modified the component which you can find in this link. (the modification is not very scientific, but it does the trick).

https://github.com/Vyrwn/byron-master/blob/master/Ladybug_Thermal%20Comfort%20Indices_Grid.ghuser

I hope this is useful

-Byron

from butterfly.

@TheodoreGalanos , @chriswmackey

1. What does the WindFactor(10) column in the .csv file represent?
2. For what direction was the cfd simulation in openfoam been conducted? Westerly wind direction?

@mostaphaRoudsari
Thank you for pointing me to the topic. I am a bit ashamed that I am not investing a little bit of my free time to check for the new topics on forum. I even missed this topic on github.

@chriswmackey :
I agree that you should be able to use the Thermal Indices component to make simple outdoor comfort maps if you are not interested in accounting for surface temperatures similar to this simple UTCI map.

One can always supply new values to the meanRadiantTemperature_ input, which will account for the surface temperatures.

@Vyrwn
Hi Byron,
Thank you for uploading your .gh file.
I think I overcrowded this topic a bit. I opened a new topic on forum, with the reply to your PET issue. In future, it is the best to seek help on forum, as it has better chances of getting noticed, than in here. I wouldn't even see your question in here, if Chris hasn't tagged me.

from butterfly.

Hi @chriswmackey , @mostaphaRoudsari , @TheodoreGalanos ,

Is it possible to get an answer on these two questions related with the .csv Open Foam result file from the grasshopper forum:

1. What does the WindFactor(10) column in the .csv file represent?
2. For what direction was the cfd simulation in openfoam been conducted? Westerly wind direction?

Thank you.

from butterfly.

@stgeorges ,
I apologize for such a late reply but you can find a full answer to the question here:
#14

Given that the end results of this conversation have split into different github issues, it seems appropriate to close this one now. Feel free to reopen if you think I am mistaken
-Chris

from butterfly.