Applying Rough Sets for the Identification of Significant Variables in Photovoltaic Energy Production with Isolated Systems

The main objective of this work is to study the state of the art of current techniques and algorithms for improving the efficiency of isolated solar photovoltaic systems. Additionally, a study will be conducted regarding the feasibility of applying rough sets (RS) for the practical identification of a minimum set of significant input variables (condition attributes) which determine the value of the output variables (decision attributes) in solar photovoltaic systems. Several experiments were carried out using a TS97 solar photovoltaic system donated by T-Solar for research purposes. The developed system was used to capture the values of input variables in different periods of time and climatic conditions (obtained through a MeteoGalicia monitoring station). The experimental prototype implemented was composed of a solar photovoltaic panel, a resistive load to dissipate the energy generated by the panel, a MPPT (Maximum Power Point Tracking) placed between the panel and the charge, a data logger equipped with a RTC (Real-Time Clock) and several sensors for measuring the values of physical and electrical variables of the system. Data captured was stored in a 2GB solid state card using a plain text format in order to facilitate later study and analysis using RS.