Hi,
in a separate post under VRM beta I’ve written about a strangely, seemingly high, consumption forecast published in VRM (beta beta & production website).
I’ve been playing around with different AC Load meters and different modes on those for measuring my ground heatpump - originally I had a Shelly 3EM pro installed, but swapped that out early january of this year for a Elgris smart lan 2.0 meter. This one runs as a EM24 from Venus OS point of view and connected over LAN.
With that one I started to see higher than expected consumption forecast figures, it started suddenly around 29th or 30th of Januari of this year.
Over the last weekend (Saturday evening) I tested to set the power meter as type EV Charger in VenusOS, I regretted that change after 2-3 hours if I recall correctly and set it back to AC Load in Venus OS under the device. I also removed missing devices in Venus OS device screen (classic interface)
The day after I noticed that the consumption prediction for Monday of this week for the grid had jumped even higher and all the consecutive days prediction indicated the same, that our house should draw around 130 KwH during a 24 hour period.
So can it be so that the Venus OS fails to delete the old device type (EvCharger in this last example) from the cloud and hence the device is still counted for in the prediction model, thus adding one instance for each device type that it has seen?
I’ve checked the device list in VRM and its empty of any unknown devices.
Ideas of how to proceed to reset the prediction value back to the right amount?
Tomorrow prediction
Todays prediction with six hours already accounted for
Yesterdays real consumption (which is very typical of our house during winter)
