22.8.1 Generation of Daily Radiation Sequences
Long series (many years) of daily irradiation data are sometimes required for particular purposes, for example, when studying the long-term reliability of stand-alone photovoltaic systems. However, long series of historical data are scarce and hard to obtain. This leads to the need for methods that are able to generate a series starting from widely available information, such as the 12 long-term average monthly mean values of the daily irradiation Gdm(0). The idea is that the generated series must preserve some statistical properties believed to be universal, as they are also found in historical data, when available. In particular, the persistence of solar radiation, that is, the dependence of today’s irradiation on the irradiation of the preceding days, is adequately described by a first-order auto-regressive process [33]. Moreover, the probability function of the daily clearness index for any given period has a form associated with only its average value for the period. Several methods for the generation of daily irradiation sequences are available in the literature [34]. The method proposed by Aguiar [35] is the most widely used today.
