I’m looking to add some intelligence to my space and water heating automations by basing my control algorithms on usage statistics. I’m looking for a sensor that can learn the usage of a space over a period of time and let me know how its being used and when.
What I’m thinking of is something like this -
For space heating I’m thinking of something like a leaky bucket histogram with 24 * 1 hour “containers” * 7 days for each space. As a occupancy sensor in the space is triggered it increments the appropriate “container” in the histogram. The more the sensor is triggered the greater the value in the relevant “container” becomes. Over an adjustable time period the value in the relevant “container” bucket leaks away to zero. What this gives is a learned histogram of room occupancy - the most used rooms have the highest values and hourly usage is recorded. As the usage changes the histogram adapts to the new usage pattern.
What I can then do is target my heating towards rooms that have the highest usage and only heat other rooms enough to prevent dew forming (or another set back temperature).
For water heating I can determine the days with the highest levels of hot water usage and optimise my heating so I only heat what’s required.
Is there anything in Home Assistant that can do this? I’ve searched but not been able to find anything like this.
Thanks