Open fabi9163 opened 3 months ago
I get your point. And i struggle to have a reasonable algo idea. The current code just checks whether the remaining predicted pv power production is larger than the missing power to reach the target SOC. In the automation we do not know the concretely predicted pv production curve for the remainder of the day. One would have to take this curve and check how long those 1.7kW (plus maybe a margin) might still be produced as excess power. Just assuming that there is time till sunset, will not be correct looking at typical pv production curves.
Great Script thanks a lot! Seems there is one problem I'm not able to solve with it:
Problem
My Battery can only charge with 1,7kw, after 50% only 900W and runs multiple "battery tests" in which it doesn't charge at all (...don't ask->Senec, it's ...dahh). My PV-Power is >8kwp.
(My example is extreme, but I thing many people have more PV-Power than the battery can handle)
The result is that the Script thinks there's enough energy (forecast) to charge the battery later and enables additionally consumers. When it decides to force charging there's not enough time to charge the battery anymore because of the slow charging rate while theoretically there would be enough energy available.
Also in "force charging"-State much power goes to the grid cause the battery can only take <1,7kw
Idea:
Nice would be additional input-values like
and some logic like:
Maybe someone have the same problems like me, has some other ideas or can optimize the script. Would be great :-) If I have understand something wrong, please let me know.
Thanks!