Generally speaking, your heat pump appears to be built around something called a “combi buffer”?
This is typically not the most efficient setup, but since you have a ground source heat pump it might not be that bad.
Concerning your initial question, usually a heat pump “likes” to run slow and steady unless it is oversized. If you want to load shift away from price spikes you would first need to figure out how well your home stores heat. Once you know that you can decide how long your “breaks” can be. I live in a well insulated home with an air source heat pump and can easily turn off the heating for 4 hours without losing too much room temp.
Just for the record, I’ve resorted to a calendar-based automation. When I have an event in our shared Home calendar with a certain name.
Here’s an example of my house’s heat inertia. My indoor temperature decreases about 2°C per day with outdoors temperature fluctuating between -2°C to 2°C:
If the event summary is Sauna, it simply turns the sauna on for the duration of the event.
If the event summary is Away or Travel and the event lasts 2+ days:
Turns off the heat pump 12 hours before the event starts, as we really can’t feel the drop of ~2°C at home. The event usually starts at 00:00, so the heating is turned off at noon the day before travel.
Turns on the heat pump at the event’s end.
The script.heat_min just stores the current heatpump mode in an input_text and script.heat_restore restores it.
I find this works well enough. The power consumption by the house with the heatpump on was ~27kWh per day, and with the pump turned off (actually at this point I had just set the target temperature to 10°C) was around 9kWh per day so 3x savings, which manifested in ~€2 savings per day (Nordpool electricity prices were quite cheap during that period), totalling ~€8 for this away session. Note: the savings are without adjusting for the added power for “reheating”, but turning the pump off instead setting it to 10°C will likely cover that. As long as I’m not away for weeks and it’s not -30°C outside, there should be no risk of damage to pipes due to freezing.
Next step that I’d like to do, is to look at the energy prices e.g. 6 hours before and after the pump is scheduled to start, and re-heat when it’s the cheapest period.
I’m currently looking at how often pumps are running. It’s getting pretty cold here atm (-12C) and I might need to bring the Curve max to 55 again if the pump is not able to keep up.
The house is pretty big, today will be consuming around 50kwh I think.
I’m thinking of having some sort of dumping place for some data so that we can compare each other. Would that be something you folks would like to try?
I’m guessing a git repo might be a good place to start.
Temperatures were fluctuating quite a lot so it’s nice to see the impact.
Sadly my power consumption is the whole house, but to be fair, the main power draw is from the heating system.
Here is a regression plot that illustrates the relation between the outside temp and the consumption (hourly)
I’m currently working on a python script that will output a target temp and a thermia STATUS. The goal is to regulate the target temp based on the predicted outside temp, electricity price and “hot water” needs. I’m also going to add a peek price mode that should be able to preheat before the cost is too high and stop heating altogether when the price is really high.
Once I’m happy with the python script, I’ll try and convert it into ??? I don’t know, I would need help here. Not sure what’s the best way.