Today's Losses

[jdpmendes]

Hi all,
First of all, thank you for this Integration, it works very well so far.

I have noticed that in my system, the estimation for Today's Losses is implausible.

See for example the the attached picture.
The estimated energy losses are roughly the same as the solar production energy, and we see the rate of increase also follows the daylight.

For reference, the solar production for me comes exclusively from a 3 phase microinverter connected on the Gen port. I suspect this is linked to the issue. This is on a DEYE lg05lp03.

[jdpmendes]

I looked a bit at the code, and for my case, I believe the issue is the following (deye_p3.yaml):
- name: Today Losses alt: Daily Losses friendly_name: Today's Losses description: Includes today's consumption of the inverter device itself as well AC/DC conversion losses class: "energy" state_class: "total_increasing" ensure_increasing: uom: "kWh" rule: 1 digits: 1 scale: 0.1 registers: [0x0208, 0x0211, 0x0218, 0x0203, 0x0209, 0x020E, 0x0202] uint: enforce sensors: - registers: [0x0208] - registers: [0x0211] - registers: [0x0218] - registers: [0x0203] - operator: subtract registers: [0x0209] - operator: subtract registers: [0x020E] - operator: subtract registers: [0x0202]

Today Production [0x0211] and Today Generation [0x0218], have the same value in my setup, which leads to double accounting of the micro-inverter production, and therefore the incorrect calculation of the total losses.
The "Total Losses" has a similar problem.

For my case, I guess removing the "Production" type parameters is the right approach. However, I wonder how this behaves on a setup with an actual Generator.
Anyone knows?