Just to give you an update on this; I planarised the geometry and the problem has disappeared. I imagine it was due to the presence of curved surfaces.
Thanks again for your help on this!
May I ask you few other questions on this topic? Please let me know if this should be moved to another discussion.
I have a list of items on which it would be great to have your feedback:
If I input the analysis period in Microclimate Map Analysis component as in the picture below, am I sure that I am including in the assessment only the hours between 8 am and 8 pm for that week or I am considering all the hours from 8 am the 26th of April to 8 pm the 2nd of May? Basically, I want to make sure to consider only the occupied hours. I am quite sure I am only considering the occupied hours but I would prefer to confirm.
This procedure is great to evaluate comfort over a week as it takes time to run the analysis for the whole mesh. Is there a way we can set control points (similar to what we can do with radiation or illuminance) and plot for those control points the PMV or SET for every hour of the year?
How is the SET defined? What is its equation? I assume the SET accounts for direct radiation falling directly on the user considering the SolarCal model?
I would like to compare different configurations (e.g. using different glass coatings, etc.) and show the improvements for example showing that the hours with SET < 28 degC are less of a certain % compared to a previous design option. Do you have any suggestions on how to achieve that?
Is there a recommended upper and minimum threshold for the SET or this should be simply agreed according to the specific needs?
I moved you to a new topic since this is definitely a departure from your previous one. To quickly answer your questions:
Yes. Setting the analysisPeriod like that will only run the calculation for occupied hours.
There’s no input for control points but you can plug custom surfaces into the _distFromFloorOrSrf input of the “Honeybee_Indoor View Factor Calculator” component to customize the regions where you want to run the comfort study. You can also adjust the _gridSize to adjust the number of control points generated. I’ll try to add in an option to plug in a custom viewFactorMesh at some point soon and, when I re-write everything for Honeybee[+], I’ll structure the generation of points the same way that it is for daylight simulation. Also, 90% of the calculation time of the PMV recipe is running the PMV model (only 10% of the time is spent in computing the inputs of temperature, etc for the model). So, if you use an Adaptive Thermal Comfort recipe, which is a lot simpler than PMV, it will run in 10% of the time.
You can convert the result matrices that you get out of the microclimate maps to Grasshopper data trees using the “Honeybee_Matrix to Data Tree” component. Once you have the raw results of the simulation in a Grasshopper Data Tree, you can manipulate it with whatever criteria you want using the native Grasshopper math components (including finding the hours with a SET above/below given criteria).
If you’re interested in the number of comfortable hours, why not just use the ComfMtx that you get out of the component? Or look at occupied thermal comfort percent like in this example: http://hydrashare.github.io/hydra/viewer?owner=chriswmackey&fork=hydra_2&id=Comfort_Autonomy&slide=0&scale=1&offset=0,0
I think a SET of ~30 C corresponds to a PMV of 0.5, which fails the ASHRAE-55 standard. But, if what you are interested in is whether a point passes or fails a comfort standard, there’s no need to deal with the SET directly.
I know this is an old topic, but I am having a similar issue. So just wanted to cross-check. @chris, the microclimatic analysis component would not work with non-planar study surfaces?
We never said that the microclimate maps don’t work for non-planar surfaces (if they don’t, please post another issue and I will have a look).
It’s just that I know it takes a long time to calculate such that the cost / benefit is usually in favor of cleaning up your geometry into a planar format to avoid the long calculation times.
