Beta distribution function is best suited for solar power modelling.

Check the paper titled:

Optimal mix of solar and wind distributed generations considering performance improvement of electrical distribution network

Refer my conference publications. 

Several authors used Beta Distributions with parameters fitted on historical data (see e.g. Risk assessment and risk-cost optimization of distributed power generation systems considering extreme weather conditions Reliab Eng Syst Saf, 136 (2015), pp. 47–61 or Statistical analysis of solar measurements in

Algeria using beta distributions F. Youcef Ettoumi a , A. Mefti b, A. Adane a,*, M.Y. Bouroubi)

If you are intrested in alternatives for characterize uncertainty in the sunlight, you might be intrested in considering e.g. analytical solutions to obtain clear-sky sun irradiance and Markovian processes for modelling of cloud index cover.

I have used the Perez model to give figures for direct and indirect irradiance, and generated a simple Monte Carlo formula to generate pseudo random light levels which correspond quite closely with observed values. If this is what you need I could send you a reference.

Hi,

@Tony: Many thanks for your answer. I would like to see your reference and look forward to receiving it.

e-mail: [email protected]

@Samuel: Many thanks for your valuable information. I did not know that HOMER has such capability!

Regards,

Morteza

I face the same problem. thanks for all useful hints.

Morteza,and Mehdi, I would only comment that in most locations the variability in solar energy due to weather vastly exceeds any variability due to matters like misalignment of the array, or even latitude within +- 10 degrees. Don't expect "accurate" results. They don't exist!

Morteza 

Which PDF did you select for your solar PV.

Beta distribution function is apt for modelling solar irrdiance and calculate exact output power from the PV module.