Innacurracy in monthly PR values due to horizon profile

[Loick Le Coz]

Hello,

I would like to point out a discrepancy between the PR calculation in PVsyst and a on-site calculation of the PR, in a environment with important horizon profile.

The monofacial PR formula used in PVsyst is as follows :  PR  =  E_Grid  /  (GlobInc * PnomPV)

 

The main issue I find is that the GlobInc (GTI) does not account Horizon Losses in PVsyst.

However, in the case of a real PV plant, with a pyranometer installed on the plane of array, the  measured GTI will consider the horizon.

This leads to an inconsistency between the estimated irradiance and the real irradiance received by the modules (unlike the production, which consider the horizon losses in its calculation). As a result, it causes the PR to drop significantly in some month when the horizon impact is high. (in my case, a height of almost 30° in some month)

 

I understand that the PR formula is based on the IEC 61724 norm wich states “[…] giving PR as the ratio of measured energy to expected energy (based only on measured irradiance and neglecting other factors) (14.3.1 – Performance Ratio) »

I belive it would be fair to consider the “measured irradiance” as irradiance that includes the horizon profile.

 

One methodology I propose is to calculate a “GlobIncHrz” which uses the Horizon losses (HrzLoss) and GTI (GlobInc), (GlobIncHrz = GlobInc – HrzLoss) and use this revised GlobIncHrz to calculate a PR value.

Would you agree with this methodology, and the accuracy it would bring to a better estimate of the performance of a plant in a site with strong horizon profile ?

 

Regards,

Loïck