Solar power generation has characteristics that are similar to wind power generation, namely the fact that is a nonlinear and bounded process. It is bounded owing to the fact solar panels obviously have a limited nominal capacity, and nonlinear due to the combined complexities of all processes involved. Intuitively, solar power generation is a function of the solar radiation, in turn being dependent on the solar angle and cloud cover. Less intuitively, temperature and humidity also have a significant effect, by impacting the performance characteristics of solar panels.
Similarly to wind power management, optimal management of solar power production calls for a complete information about expected production, with forecast horizons from few minutes to few days ahead. Complete information translates to the providing of point forecasts, along with uncertainty estimates.
The research related to the wind power application deals with these various topics:
- Point forecasting of solar power generation (for horizons from few minutes to several days ahead)
- Probabilistic forecasting of solar power generation (up to few days ahead): quantile forecasting, prediction intervals and density forecasts, or alternatively scenarios
- Statistical (grey-box) clear sky modelling, accounting for the physical phenomena involved, though reckoning parameters may be better estimated from statistical approaches
- Modelling of the stochastic dynamics of a solar power farm