Solar radiation under clear-sky conditions provides information about the maximum possible magnitude of the solar resource available at a location of interest. This information is useful for determining the limits of solar energy use in applications such as thermal and electrical energy generation. Measurements of solar irradiance to provide this information. Clear-sky conditions, on the other hand, present atmospheric conditions that produce predictable effects on solar irradiance. There is growing interest in models that predict clear-sky solar irradiance, which resulted in the development of many models that vary in complexity and accuracy of prediction [3][4][5][6][7][8][9][10][11]. A majority of these models predict broadband clear-sky irradiance, where the clear-sky atmospheric effects are accounted for by broadband attenuation parameters such as Linke-turbidity coefficient [12,13] and Angstrom coefficient [14,15]. Calibration of the models for local conditions involves an empirical process that computes the relevant broadband attenuation parameters using the clear-sky irradiance models backwards, with a selection of measured local clearsky solar irradiance data as input [16]. It is also possible to generate clear-sky irradiance from a set of astronomical and weather parameters using artificial neural networks (ANNs). The ANNs approximate the functional relationship between random input and output variables by learning from examples made up of historical data output and input variables [17]. Published applications of ANNs in the field of solar energy include time-series forecasting of solar radiation quantities [18][19][20][21] and other function approximation or regression models that map a set of input parameters like temperature into radiation quantities [22][23][24][25][26]. One major attraction of ANN methods is their ability to find relations between input and output even if the representation was intractable [19]. The ANN can, therefore, map a wide range of possible combinations of input or explanatory variables to a single desired output. This, however, does not underplay the importance of carefully selecting the variables. Koca et al. [26], for example, showed that different combinations of inputs affected the performance of ANN models that predicted global solar irradiation.Solar energy is one of the promising sources of energy in South Africa. It is therefore important to investigate the performance of solar radiation models for South African conditions. A growing database of solar irradiance data from measurements by the Southern African Universities Radiometric Network (SAURAN) [27] provides opportunities to investigate and develop clear-sky models for South African conditions. The present investigation considered four models, two of which are semi-empirical broadband models: Ineichen-Perez (I-P) [10] and European Solar Radiation Atlas (ESRA) [11], which take

This paper introduces a solar resource index that responds to site-specific sky conditions resulting from stochastic movement and evolution of clouds. The developed solar resource classification index called probability of persistence (POPD) had limited capabilities to distinguish persistent clear-sky conditions from persistent overcast-sky conditions. The metric proposed in this investigation IntroductionSolar energy is becoming an increasingly important component of the energy mix required to confront current global energy and environmental challenges. Detailed knowledge about its availability and variability over different time-scales are important for its exploitation to be cost-effective and efficient. Solar resource variability is primarily caused by earth-sun relative motion and movement and evolution of clouds. Variations induced by the apparent motion of the sun relative to the earth are visible on diurnal and seasonal scales, and can be predicted precisely from well-established astronomical equations [1,2]. Variability caused by clouds is less predictable, and manifests as short-term temporal fluctuations that modulate the otherwise uniform astronomicallydriven diurnal irradiance profiles. These stochastic fluctuations vary in amplitude, persistence (duration), and frequency of occurrence [3]. Assessment of the solar resource therefore requires a statistical approach using appropriate statistical metrics that model the variation in solar resource magnitude under the influence of local stochastic weather influences over different time-scales. Several metrics that show varied solar resource discrimination capabilities exist in available literature. These include fractal dimension (FD) of daily profile of global horizontal irradiance (GHI) [4], daily clearness index probability distribution functions [5], granulometric size distribution of GHI [6], variability index (VI) [7] and daily probability of persistence POPD [8]. The FD of GHI as proposed by Maafi and Harrouni [4] measures the amount of daily solar irradiance fluctuations that are due to changes in the state of the sky. Values of FD close to 1 indicate persistent skyconditions that are characteristic of either a clear day or an overcast day. These two extremes of the solar resource were distinguished by combining the FD with the daily clearness index K T , to present a solar resource classifier that identified three classes of solar resource days using GHI data from two sites in Algeria [4]. The approach proposed by Soubdhan et al. [5] was that the classifier discriminates daily solar resource according to daily distribution histograms of instantaneous clearness indexes k T . Four solar resource classes were identified at Guadeloupe, an island in the West Indies, from a year-long sample of irradiance data measured at a frequency of 1 Hz. The membership of each class is subject to similarities in marginal probability density functions (pdfs) that are modelled using Dirichlet distribution functions from the daily histograms of clearness i...