A statistical characterization of the long-term solar resource: Towards risk assessment for solar power projects

Abstract

In this study, a statistical characterization of annual solar irradiation series is presented which can be used as input in risk assessment for securing competitive financing for solar power projects. To perform this task, an analysis of annual Direct Normal solar Irradiation (DNI) and Global Horizontal solar Irradiation (GHI) probability density functions has been carried out, showing that annual DNI and GHI distributions are described by Weibull and normal functions, respectively. Normal fitting of annual GHI distributions yields uncertainties in mean parameter below 1%, and uncertainties in standard deviation parameter of ∼12%. Weibull fitting of annual DNI distributions yields uncertainties in scale parameter of ∼1%, and uncertainties in shape parameter of ∼15%. For each location analyzed in this study, the estimated regression coefficients (and their uncertainties) of annual solar irradiation distributions fitting are used to obtain percentile values and their respective associated uncertainties. The greatest uncertainties are associated with the lower percentiles, being 1st percentile uncertainty ∼1.6% and ∼4% for GHI and DNI respectively. Finally, according to the results obtained in this work, a minimum of 11 years of GHI and 15 years of DNI are recommended for their statistical characterization.