Relations Between the Probabilities of Exceedance of Solar Radiation and Production of Concentrating Solar Thermal Systems

Abstract

CSP feasibility analysis, bankability assessment and plant design are generally based in one-year meteorological series representative of “typical” o extreme conditions. There is a growing interest in accounting on the inherent variability of the performance of solar plants on their input variables in the plant model. In particular, for generating the probability distributions of annual energy yield, simulation tools require as input a large number of plausible meteorological years (PMYs). In this article we generate 100 synthetic years of DNI in the 1-min resolution for the location of Seville, Spain, and we estimate and evaluate the power produced by two CSP plants with similar configuration as two well-known parabolic trough and central receiver solar plants (Andasol 3 and Gemasolar) using System Advisor Model (SAM). We estimate that there is an almost linear relation between the Probabilities of Exceedance (PoE) of the DNI and the gross electricity generation of the solar plants but that for a given annual value that corresponds to a PoE of the DNI, there is a range of possible production values depending mainly on the monthly distribution of the solar radiation datasets.