Hi guys,
I want to run my thermal comfort simulation to get a single set of data for each room (operative temp/pass, fail etc). I note that I am required to create a sensor grid to run the model, but I only really want one sensor per room. Is it possible to set this up? Alternatively I guess I could calculate the average for the room for each hour of the year?
Chris
Hi @chriss, try to set the grid size to a very large number and that will give you a sensor grid with a single sensor. It should work both in Grasshopper and Rhino.
Hi @mostapha, I’m drawing the rooms in Rhino. I set the grid size to 99999999, and I still seem to get quite a number of sensors in some rooms (the more oddly shaped rooms, as it happens):
Interesting, I see that with this form, and because of how the mesh is being generated this trick doesn’t work. @chriswmackey, would you consider this a bug or just how it is supposed to work?
Meanwhile, you can use the bounding box of each room to create the grids. This one should work for any form as long as you can find the aligned bounding box.
create_grid_with_single_sensor.gh (29.2 KB)
Hey @chriss ,
If you are not concerned with mapping the differences in temperature that people experience across each Room and you are only concerned about the average conditions within each room, then using the comfort mapping recipes is really overkill since you really don’t need Radiance for this. Instead, I would recommend doing a workflow like you see in the tutorial video below, where you plug the average air and radiant temperatures from EnergyPlus directly into the LB Adaptive Comfort component or the LB PMV Comfort component.
The previous and following videos in that playlist are also likely to be of interest to you. Sample files are included in the video description.
Hey @mostapha ,
I don’t think I’d consider it a bug but, even if I were, then it’s not really a bug on our end since those meshes and sensor grids are being generated using the Rhino meshing algorithms. Still, I also don’t know what else I would want from McNeel for a case like this since mesh faces can only have a maximum of 4 vertices and the very fact that I get a mesh face at all is good enough for me. This is particularly because I feel like these cases of “one sensor per room” aren’t ones where we really care about the mesh anyway. Maybe I should ask what were you expecting to get in a case like this if not the current result?
