String and Central Inverter Clipping Losses.

[vishalbhavar]

Hi Team,

I have performed two different simulation with Sungrow 3437 kW central and Huawei  215 kW string inverter.

Can you please clarify why inverter loss over nominal power is more in central inverter (0.78%) than string inverter(0.31%)?

Sungrow central inverter Project AC capacity(501.8 MW) is more at 45 Deg C than Huawei string inverter(487.5 MW) and Maximum site ambient temperature is 30.2 Deg C, therefore I believe temperature degradation should not be an issue. 

Project capacity and details as below.  If you need anything further please let me know.

Module	Inverter	Max Ambient site Temperature	Inverter rating @ 45 Deg C	Project AC Capcity @45 Deg C in MW	Project DC Capacity	Inverter Loss over nominal inv. Power	Inverter Loss during operation
Canadian Solar, 665 [Wp] Monofacial	Sungrow,  3437 kW	30.2	3437 kW	501.802	670 MWp	0.78%	1.34%
Canadian Solar, 665 [Wp] Monofacial	Huwei, SUN2000-215KTL	30.2	195 kW	487.5	670 MWp	0.31%	1.55%

Edited June 10, 2022 by vishalbhavar

[dtarin]

I dont have the OND file, but my guess would be that 1) the huawei has a wider mppt voltage range and 2) it has a higher input current rating compared to the sungrow

[André Mermoud]

This probably depends on the PNom definition as a function of the temperature. 

- This may be different in both cases. You don't provide the values of PNom and PMax of each inverter and the corresponding reference temperatures.

- It depends also on the way you have defined the inverter's temperature during operation in the dialog "Energy Management". 

 

[vishalbhavar]

Hi Andre/Dtarin,

Thanks for your response please find inverter and snapshot at different temperature as below.

- This may be different in both cases. You don't provide the values of PNom and PMax of each inverter and the corresponding reference temperatures.

If I compare Pnom and Pmax of both inverter still sungrow has max AC capacity as compare with Huwei string inverter at 45 Deg Temperature.

- It depends also on the way you have defined the inverter's temperature during operation in the dialog "Energy Management". 

As shown in below snapshot for both inverters I have considered same assumption.(Outdoor Installation).

If you need anything further please let me know.

Sungrow:-

Huwei:-

[bapossatto]

Assuming initially that in the "Energy management" area you have set "Inverter temperature for PNom evaluation" to "External ambient temperature (outdoor installation)" in the same way for both inverters, perhaps it is valid to analyze the "Grid power limitation" field, also in the "Energy management" area; if the limitation occurs at the inverter level and considering that the installed power of the systems is different, there will probably be a difference there.

Anyway, I believe that PVSyst has a great difficulty, which is the modeling considering the temporal and spatial variability of irradiance in a large-scale power plant.

See the article below:
https://www.sciencedirect.com/science/article/abs/pii/S0038092X19304530

You will see that a central inverter will suffer less clipping by overradiance events, but despite this, the consideration of hourly data generates modeling errors that can be significant.

[Michele Oliosi]

Yes we are aware of these potential differences when the sub-hourly variability of irradiance is important and the Pnom ratio/Grid limitation clipping is high as well. Thanks for the article reference by the way.

We are presenting a model to take this into account at https://www.wcpec-8.com/. The idea is to implement it into PVsyst.