Deviation between SolarEdge consumption data and PVOutput values


#1

The production is spot on:

e.g Aug 4, 2019

https://pvoutput.org/intraday.jsp?id=52387&sid=47624&dt=20190804

PVOutput:
1326,854 kWh

SolarEdge
1326,854 kWh

Perfect.

However, also the SolarEdge modbus energy meter is installed for the consumption;

PVOutput:
1127,778 kWh

SolarEdge:
1170 kWh

3,6% difference?

On another installation I see a similar difference.

What can it be and how to solve it?

Best regards.


#2

SolarEdge API only provides consumption at 15-minute intervals which will affect the total amount over a longer period of time.


#3

Ok, but solar production is also provided at a 15 minute interval and that one is perfect.


#4

I have the same problem:

13/02/2020
PVOutput:10.978kW
SolarEdge:12.542KW
dif: 1.564KW

12/02/2020
PVOutput:14.514kW
SolarEdge:14.838KW
dif: 0.324KW

11/02/2020
PVOutput:14.554kW
SolarEdge:14.805KW
dif: 0.251KW

10/02/2020
PVOutput:13.693kW
SolarEdge:13.734KW
dif: 0.041KW

09/02/2020
PVOutput:15.982kW
SolarEdge:15.855KW
dif: 0.127KW

PVOutput

SolarEdge


#5

How are you getting your production data to match exactly on the SE portal vs. pvoutput? PVoutput shows about 3% less on average. I donate to pvoutput so it captures data in 5 min increments (pulled every 15 min).


#6

I suspect you may have a different issue as well - looking at your daily charts, it is showing your system is producing energy even during the night. Looking at days like https://www.pvoutput.org/intraday.jsp?id=82775&sid=73393&dt=20200223&gs=0&m=0, it looks like you are uploading a combined generation+consumption into the PVO generation field.
Ideally, that chart should show the green line at precisely 0 up to ~6am and after ~5pm (dawn and dusk), yet here it is showing a typical consumption pattern.
Hope that helps…


#7

I thought that this output would be due to the consumption generated from the LG battery when there is no sun. What I have checked is that no record is generated at 00:00 valid in the program.

Pensaba que esa salida sería debido al consumo que se genera desde la batería LG cuando ya no hay sol. Lo que si he comprobado es que no se genera ningún registro a las 00:00 valido en el programa.


#8

Got my solar panels installed last month and I’ve run into a similar problem. The difference is that the deviation I’m seeing is in the output instead of consumption (because I’m not measuring that).

The deviation between solaredge monitoring and pvoutput deviates by roughly 0.5 to 1.5 kWh per day. For example, yesterday 27.02 kWh was reported by solaredge and pvoutput registered 26.214 kWh.

I don’t really understand why this deviation appears, because I figure solaredge uses the exact same data for reporting as pvoutput queries it through their API.

When I have some spare time, I’ll check what their API is reporting and make some comparisons to see why this is the case.


#9

Is one number raw production and the other number due to efficiency loss? I’m seeing substantial differences only on some days, when comparing what SolarEdge says I generated to what the utility sees going back to it.


#10

To my understanding, Solaredge overestimates energy production when compared with my electricity supplier.
In pvoutput, one can select in the auto uploader to get the data from solaredge or to allow pvoutput to estimate energy generation.
To me it seems that pvoutput is closer to the estimates of my electricity supplier.