victronenergy / dynamic-ess

MIT License
76 stars 5 forks source link

[Bug]: System flipping between charging /discharging forth and back from the grid within seconds each time #81

Closed SimonYoungtree closed 9 months ago

SimonYoungtree commented 9 months ago

Contact Details

siempoot@gmail.com

VRM portal ID

c0619ab31cc4

Country / region

Netherlands (nl)

B max

40

TB max

6

FB max

6

TG max

16

FG max

16

Battery costs

0.027

Buy price

(p+0.13)*1.21+0.02

Sell price

(p+0.13)*1.21-0.02

feed-in possible

yes

feed-in control

None

Version

0.1.6

What happened?

I see the system DESS switching from charging from the grid to discharging to the grid from ~4K to +2K and everything in between for several minutes 15+ already. (all around 8:40am 22-9-23). At the same same battery is being charged/discharged with more or less the same value.

There were no changes in the loads (changes in the graphs are cause because this is a calculated value, being temporary wrong when big changes happen).

The two screenshots give an impression of the 2 states the system was flipping (the extremes were hard to catch with a screenshot). See also the screenshot about the battery being charged/discharged all the time.

Screenshot 2023-09-22 084132 Screenshot 2023-09-22 084243

Screenshot 2023-09-22 085537 -2

Relevant log output

No response

Screenshots

![DESCRIPTION](LINK.png)
dirkjanfaber commented 9 months ago

It could be caused by the fact that the system does not have the battery capacity set correctly, so I'd like to know if your flow contains the following piece:

If not, description if adding it can be found here.

SimonYoungtree commented 9 months ago

The issue with battery size not being set correctly was already tackled in the beginning (0.1.2 -> 0.1.3.). At that time I had slow charging discharging. This is no longer the case. I do have the shown piece. I also found a few other time windows when this was happening. Also during mid night when there was no sun.

Screenshot 2023-09-22 102217 Screenshot 2023-09-22 102601

dirkjanfaber commented 9 months ago

Thanks for the info, we are looking further into this.

dirkjanfaber commented 9 months ago

This is probably caused by the inability to hit exactly zero current at battery idle, and an SOC indication that's right on the edge. So the slightest bit of current going back into the battery sends it back to 21% and makes it go back into discharge. We'd like to test a fix for this (increasing the used hysteris) on your system. If you enable remote support, we can make that change for you and see if this is indeed the fix that needs to be done.

SimonYoungtree commented 9 months ago

remote support has been enabled now

dirkjanfaber commented 9 months ago

Thanks. At the moment we are testing the fix on our own systems. Once we are sure that no side-effects arise, we'll put it on your system. Will let you know once that is done and when you can disable the remote support again.

dirkjanfaber commented 9 months ago

Your system has been patched with this fix. You can disable remote support again. We are scheduling this patch to be included in the next release as well. Closing this issue now; please re-open the issue (or open a new one) in case the problem comes back.

SimonYoungtree commented 9 months ago

Thanks very much. Keep up the good work. I have closed the remote support again and I will keep an eye on it.