Illuminance prediction through statistical models

A reliable forecast of renewable energies production, like solar radiation, is required for planning, managing, and operating power grids. Besides, the short-term prediction of the climatic conditions is very useful for many other purposes (e.g., for Climate Sensitive Buildings). Data for the prediction can be produced by several sources (satellite and ground images, numerical weather predictions, ground measurement stations) with different resolution in time and space. However, the unsteadiness of the weather phenomena and the variability of the climate make the prediction a difficult task, although the data collected in the past can be used to capture the daily and seasonal variability. In this paper, several autoregressive models (namely, AR, ARMA, and ARTMA) are challenged on a two-year ground solar illuminance dataset measured in Milan, and the results are compared with those of simple predictor and results in literature.