IAM Profile Of a Module not yet in PVSyst Database

[Alterna - Ingeniería]

 

Hello,

We are working in a project where we plan to use a PV module whose data is not yet in the PVSyst database, and we are using a .PAN file provided by a third party (independent certified laboratory).

There is a significant difference in generation when using the default IAM curve and the "User defined IAM profile", up to 3 percentage points.

We have extensively read your post about “How to determine the IAM profile ?”, and we think the IAM profile provided has been created using outdoor measurements. Do you agree with this? Would you accept this IAM profile for a new module of the PVsyst database? Or would you consider the values differ too much?

Attached to this post you can find a picture of the IAM profiles (both default, using Fresnel’s law, and "User defined IAM profile",). It’s in Spanish, sorry about that!

Thank you very much for your answer!

 

[Lazare Fesnien]

Dear Kittitut,

 When receiving PAN files from manufacturers, these usually define some parameters – not mentioned on the datasheets – which considerably boost the performance of their modules in the simulation.

These PAN files may often be directly provided by “independent” laboratories (claimed as “certified”).

PVsyst tries to detect these anomalies, and gives you this information.  In these cases you are advised to use default parameters proposed by PVsyst, which are usually more realistic.

 There are mainly 2 classes of such uncertain parameters:

• The IAM profile. We are convinced that the IAM of most PV modules should follow the Fresnel’s physical laws, either for normal glass or for AR coatings. We received measurements from very serious laboratories which are very close to Fresnel. And we had still this confirmation recently after an informal discussion with a responsible of a well-known laboratory in the USA.  See our FAQ   https://forum.pvsyst.com/topic/1190-how-to-determine-the-iam-profile/#comment-3181
• The low-light efficiency as measured by some laboratories  is often biased by the fact that the filters used for the measurements at 200 and 400 W/m2 deliver slightly higher irradiance than nominal. I have analyzed many measurement reports from these labs,  and observed that the Isc/G ratio is not stable, as it should be. When correcting this the low-light efficiency usually stays between -3% and -2%, rarely higher. See our FAQ  https://forum.pvsyst.com/topic/1030-how-are-specified-the-pan-files-in-the-pvsyst-database/#comment-2731

When using these boosted PAN files in your simulations, you have the risk of getting over-estimated simulation results with respect to the real system..

Regards,