You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running Predbat on Solax in Sweden. From January 2027, all Swedish grid companies are legally required to implement demand-based tariffs (effekttariff) — though there's some debate around the exact timeline. In practice, many companies already have them in place today.
The way it works: you get charged based on the average of your top N hourly peaks per month (typically 3–5), weekday daytime only. It can easily add 200–500 SEK/month to the bill depending on season.
Predbat doesn't handle this today, but the foundation is already there — the simulation loop, forecasting, battery model and inverter control. What's missing is essentially a penalty term in the cost function for demand peaks, plus some real-time logic to react when an unexpected spike threatens to set a new monthly peak.
Is anyone else looking at this or thinking about how to approach it? Should be relevant across the Nordics and parts of the EU where similar tariffs are showing up. Happy to help develop something if there's interest.
Following up on my earlier post. Been thinking more about this.
For those not familiar: effekttariff charges you based on your top N hourly peaks per month (typically top 3, weekdays daytime). It's already common in Sweden and becoming mandatory. The charge can easily be 50-100 SEK/kW/month so it adds up fast if you get an unnecessary peak because of a badly timed charge session.
Right now predbat doesn't know about this, so it can schedule things that create expensive peaks even though spreading the load would have been fine.
I think the main missing piece is a penalty term in run_prediction() so the planner avoids creating new peaks during active hours. Plus some awareness of what peaks you've already set this month — if you already hit 6 kW there's no reason to constrain to 5.
The real-time side could be interesting too, checking the remaining import budget each cycle and shedding loads or forcing discharge if needed, but honestly even just the planning side would make a big difference.
Is this on the radar at all? Feels like it would matter for a lot of users.
I don't believe this is on the radar to answer part of your question.
I sort of understand the issue, but I'm not convinced that predbat won't already do a good enough job without any enhancement.
Predbat minimises your electricity costs so will charge your battery on cheap rate periods, I don't know what tariff you are on, but for most tariffs that means that predbat will charge the battery overnight, not in the daytime when rates are almost always higher.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Hey,
Running Predbat on Solax in Sweden. From January 2027, all Swedish grid companies are legally required to implement demand-based tariffs (effekttariff) — though there's some debate around the exact timeline. In practice, many companies already have them in place today.
The way it works: you get charged based on the average of your top N hourly peaks per month (typically 3–5), weekday daytime only. It can easily add 200–500 SEK/month to the bill depending on season.
Predbat doesn't handle this today, but the foundation is already there — the simulation loop, forecasting, battery model and inverter control. What's missing is essentially a penalty term in the cost function for demand peaks, plus some real-time logic to react when an unexpected spike threatens to set a new monthly peak.
Is anyone else looking at this or thinking about how to approach it? Should be relevant across the Nordics and parts of the EU where similar tariffs are showing up. Happy to help develop something if there's interest.
All reactions