Dynamical optimal positioning of a photovoltaic panel in all weather conditions

In this paper we develop and verify a model predictive control algorithm for photovoltaic panel orientation with the aim to maximize the photovoltaic system netto power production. Thereby we take into account local weather forecast with its uncertainty, thermal behavior of the panel, and the positioning system energy consumption with its technical constraints. The model predictive control synthesis procedure comprises two basic steps: (i) identification of solar irradiance model and development of the photovoltaic system model and (ii) development of predictive control algorithm for the photovoltaic panel active surface orientation, based on the obtained models. Performance of the developed algorithm is verified through year-scale simulations based on a large number of solar irradiance and other weather data patterns. It turns out that the proposed algorithm is fully competitive with the mostly used sun tracking or maximum irradiance seeking controls, and that it outperforms them. The other advantages of the proposed algorithm are: (i) the positioning system is controlled smoothly and (ii) prediction of energy yield one day ahead is available together with its uncertainty for easier photovoltaic system integration into the electricity distribution network.