PVWATTS Implementation of Feuntes Thermal Model

Hi! I am a researcher at the Electricity Research Centre at University College Dublin, Ireland.

First off, many thanks for making this tool available and for publishing the details.

I have a question about the implementation of the Fuentes thermal model in PVWATTS. I have implemented this myself in GAMS, (by basically translating the Fortran 77 code found in the Fuentes reference) and the results I get using the same climate data etc. are very close to PVWATTS, but slightly different. The total DC energy over the year is within 0.7%.

Since I can exactly replicate the results from the module and inverter models, I would like to find the source of the difference in the thermal model. The only correlation I can see is that the error I observe is larger when the wind speed is low.

In the PVWATTS implementation of the Fuentes thermal model, do you use the same module characteristics as in the referenced paper? I.e.:

Reflectivity: 0.10
Efficiency: 0.08
(Emissivity: 0.84 – derived from the two parameters above)
Thermal mass/capacitance: 11000 J/m^2 degC

Did you guys make any adaptations to the Fuentes model?

I would be grateful for any assistance you could provide.

Hello,
The PVWatts V5 cell temperature model assumes an emissivity constant of . The model uses that value in equations that calculate a ground temperature ratio and ratio of total convection to top side convection.
It assumes an initial thermal capacitance of , which it uses to calculate the module's thermal capacitance:

That's what I can glean from the source code -- it does not mention specific adaptations or modifications to the original model. 
Best regards,

Paul.