Why are from roughly 14:00 to midnight those prices identical and after midnight for the next day different. Can I change some settings, that those prices are identical?
how are we supposed to answer that with that little information you give upfront?
We don’t know how you have setup your system, and we don’t know how evcc works w.r.t. fetching the prices. You don’t even explain where these originate from.
If you use the evcc binding to fetch some price channel (I’m just guessing you do), then it’s not a time series. You cannot get future values that way. The item that’s persisted is only okay for the past. Possibly not even that as then they should be equal, maybe there’s an offset of a day?
Are you sure you are displaying the inmem persisted item values in your chart and not the rrd4j ones (or whatever your default db is)?
My question was to get some insight in wich way, how often and what kind of prices the entsoe binding uses. How evcc handles it, is not a topic for this forum, I will ask it in their forum.
That’s why I did not come up with more info, because I thought it not relevant.
ok, here are some additional informations:
DB is influx V1 (yes I know, it cannot delete existing values, I replace them)
My system is a RPI5, 8GB, but I don’t know why this is relevant here.
I get the time series for the evcc prices via mqtt. It provides that time series in a json string. If this item changes, a js rule runs and converts it to a time series item, that is what the graph shows
So you have a complex way of processing evcc price data which is basically still unknown and not shown to us (those people you asked why it differs from what EntsoE gives you).
The answer to that is simple (and probably useless to you at the same time yes I know): there’s errors somewhere in your design and/or processing chain.
Try persisting the entsoe item with inmem persistence and chart that against your mqtt-evcc item right from OH builtin Apache ECharts.
Hint: use ‘service: inmemory’ in chart definition to define which db to query for persistence data for that item.
NOT from Influx, Grafana or whatever.
If they don’t match, your evcc data or your processing is bad. Like I already answered, maybe a date offset mistake by a day in your code.
It’s not important how often and when the entsoE binding fetches new data.
If the “native” entso e binding’s item chart line differs from what you have charted from your influxdb data then Influx is likely the wrong tool for that purpose.
The opposite is true. EntsoE prices are correct.
That there are times where they don’t match your values from evcc nullifies your argumentation in fact.
If during some hours your processing of values from evcc source comes to different values, then you made mistakes in that computation/processing. Even more probable since as I understand it, it’s not arbitrary mismatches but block-like. That very much sounds like a design or coding error.
If you believe that understanding what the binding does helps you with finding those errors then feel free to read the code, it’s online here. (For the record: I don’t know myself).
Maybe, someone else sees what I see: Now the entsoe graph (orange) matches the evcc graph (grey) for the rest of today. I see no change in the evcc prices for the rest of today, but in the entsoe prices.
I suggested how to proceed to evaluate that, didn’t I.
You can compare your EntsoE values with official charts from whatever source (like EPEX) to find out whether they’re correct.
Ah, okay. It cannot be your mistake of course so I must be wrong… so as rethinking that vague possibility could result in some unpleasant realization let’s better blame that hotline guy for his tone you have been disliking anyway… he also must have missed that you’re paying the premium service fee for volunteer’s advice in this forum. (I hope you read the sarcasm I put in there for you).
Good luck in finding someone else to help you with your problem so you can keep up your attitude.