Adaptive Comfort Maps - Simulation Parameters and Operative Temperature

Hello @chris!

I’m generating point-in-time Adaptive Thermal Comfort maps in a house with a conservatory. The output that I’m interested in at the moment is operative temperature, only for 1 hour of a specific day, on a horizontal grid at 0.8 m from the floor.

Compared to my annual simulation with E+, I am getting very different operative temperatures. I know they are not 100% comparable because the calculation with Radiance is not the same and it considers a grid instead of a point in the middle of the zone, but the difference is ~10°C (map reaches 10°C higher than zone in annual simulation). Air temperatures are close enough, so I believe that is related on how operative temperature is calculated.

For context:

  • I tried running the E+ annual simulation only for the 1 day to see if the Run Period is a problem, and got similar results with the same operative temperature.
  • Since I have some concave zones, I have been working with Pixel Counting for my annual simulations, but I couldn’t set the same in the Adaptive Map component as there is no input for Simulation Parameters (would it be possible to incorporate it?). I tried running the E+ annual simulation with Polygon Clipping instead of Pixel Counting, but the results were again very similar.
  • My model has some operational components that have 2 setpoints (and/or setpoints) which are not currently supported by HB, so I’m modifying the IDF file for my annual simulation with eppy. These edits cannot be input into the Adaptive Map recipe as it is generating its own IDF. As I am working with point-in-time Adaptive Maps, I can work around this for the moment by manipulating the setpoints for each day/hour to have everything operate in the same way as in the annual simulation, but do you think there is a way around this?

Do you have any clue on what could cause such a difference in the operative temperature calculation?

Also, some other questions in general about how the recipe works:

  • Does Radiance consider the glazing properties from my HB Window Material? Or should I be using a Glass Modifier?
  • In the results folder, I am getting 2 CSV files (named 0 and 1) with different results: some are really similar, others such as shortwave MRT are really different. If I am only simulating 1 hour, what do these 2 results mean?

Thank you in advance,
María

2 Likes

Hi @MariaDevecyan ,

This is to be expected for a glassy room:

To clarify, the comfort mapping recipes run the same exact EnergyPlus calculation that you refer to as an “annual simulation with E+”. It’s just that the comfort mapping recipes do two more things to give you a more accurate value for radiant temperature:

  1. They use Radiance to calculate view factors to the Room surfaces at each sensor point. Then, these view factors get multiplied by the EnergyPlus surface temperatures to give you long wave MRT.

  2. A fully-detailed simulation of shortwave solar irradiance is performed using Radiance and the result is plugged into the SolarCal model to give you an additional shortwave radiant temperature delta that gets added to the longwave MRT to account for sun falling directly onto the occupant (or being reflected onto them).

The second step typically adds 10-20C to the MRT if the occupant is in full sun. When EnergyPlus gives you its MRT output, this is only a longwave MRT value that is computed by doing an area-weighted average of the Room’s surface temperatures. So that is why you will consistently find the MRT in the comfort maps is higher than the MRT output from E+. The maps account for more things and are therefore a better representation of what occupants actually experience.

To answer the other questions:

All properties of materials in the comfort mapping recipes come from the energy simulation. All HB-Radiance modifiers are ignored. So change the HB-Energy Window Construction. Not the HB-Radiance Modifier.

It’s how the sensors are being split up so that they can be run in parallel within the recipe. In your case, you probably have 2 processors available so it splits them into two files (0 and 1). But this is just a sub-step of the whole recipe. All of the files that you should care about are in the results and metrics folders.

1 Like