Indoor Thermal Comfort Tool

Share your Projects! Custom Integrations
1

:ice_cube: Indoor Thermal Comfort Tool — Advanced Thermal Comfort Metrics for Home Assistant

The ultimate goal of this integration is to obtain a scientifically based “Real Feel” virtual temperature sensor for indoor conditions, i.e. the temperature as a person feels it, taking into account all the parameters of thermal comfort:

  • ta: Air temperature
  • tr : Mean radiant temperature
  • va: Air velocity
  • rh: Relative humidity
  • clo: Clothing insulation level
  • met: Metabolic rate

Indoor Thermal Comfort Tool is a custom integration for Home Assistant that provides calculated thermal comfort indicators based on ASHRAE Standard 55-2023, including:

  • PMV (Predicted Mean Vote)
  • PPD (Predicted Percentage of Dissatisfied)
  • SET (Standard Effective Temperature)
  • CE (Cooling Effect)
  • Thermal Sensation (qualitative scale)

The integration is inspired by the amazing work of UC Berkeley’s Center for the Built Environment (CBE Comfort Tool).

:thermometer: Why Use PMV, SET, and PPD?

  • PMV (Predicted Mean Vote) helps you understand how comfortable people feel in a space based on thermal sensation (from cold to hot), allowing smarter climate control.
  • PPD (Predicted Percentage of Dissatisfied) estimates how many people are likely to feel uncomfortable — useful for identifying comfort problems even when average conditions seem fine.
  • SET (Standard Effective Temperature) reflects the perceived temperature, factoring in humidity, clothing, air speed, and more — great for adaptive comfort strategies and real-world optimization.

Using these metrics helps you optimize HVAC operation, save energy, and maximize occupant comfort with science-based insights.

:link: Repository: Indoor Thermal Comfort Tool for Home Assistant

1 Like
2

Nice, I’m keen to try this out.

I’ve aborted “feels like” temperature a long time ago. It’s pretty useless.

I’ve been using this thermal comfort integration for a while (specifically the summer simmer index) for controlling my AC. As you’ve said, it’s labelled output, so it’s not obvious how to set a temperature for this, but I essentially just have standard heating and cooling settings that gets triggered on outside and inside temperatures, with conditions to check the abovementioned index together with a variety of other factors. I’ve gone through the summer testing this and now we’re in winter and I’m still tweaking some winter conditions. Hopefully I can improve on all of this.

3

Hello, I want to give it a try.

What are Clothing insulation level and Metabolic rate supposed to be?
Can you give an example.

Thank for your work and best regards.

4

HI.
For initial configuration, it is good to use tabular values (Table 5-1, 5-2 of ASHRAE Standard 55 or Annex C of ISO 7730). For example, a typing person generates 1.1met. Also, for example, the following set of clothes (Underpants, boiler suit, socks, shoes) creates 0.7clo insulation.

Next, it would be nice to refine (calibrate) these parameters at a time when you feel really comfortable changing the met/clo value to PMV=0.

5

if i try to add the repo https://github.com/1iverea9er/ha-comfort-tool from readme, it doesnt exist.
Using GitHub - 1iverea9er/Indoor-Thermal-Comfort: Indoor Thermal Comfort Tools HACS says it already exists, but i do not find the integration, already restarted HA.

I think it would help to add some lists from https://comfort.cbe.berkeley.edu/ to the readme to help people understand.

6

Thank you for your feedback. I have updated the Readme.
Please use repo: https://github.com/1iverea9er/Indoor-Thermal-Comfort
and search “Indoor Thermal Comfort”

7

@1iverea9er
Sounds cool, but which connected devices are capable of providing all these sensors?

8

Hi.
The main difficulty in making full use of all integration features lies in the parameters “va” (Air velocity) and “tr” (Mean radiant temperature). That is why these parameters are accepted as optional in the integration. Yes, indeed, for an ordinary user, measuring them using special sensors is too difficult a task. Fortunately, these parameters can be calculated and virtual sensors can be created based on these calculations to dynamically account for their changes.

The only caveat here is that these calculations are based on national building codes and are highly dependent on the individual parameters of the room, so they are not universal.