Dust Soiling Concentration Measurement on Solar Panels based on Image Entropy

Abstract

Solar plants are often set up in desert and arid areas, characterised by rough operating conditions (e.g. Dust soiling, rainfall scarcity). Dust soiling accumulation on Photovoltaic (PV) panels adversely influences on their efficiency, in such a way, panel power decreases considerably. Once dust soiling concentration exceeds a certain level, the triggering of the cleaning process becomes a necessity. In this paper, we propose a novel image processing-based system that allows achieving two-fold measurements, starting with the quantification of dust soiling particles concentration deposited on the outer surface of solar panels. Furthermore, the proposed system permits estimating the produced power decrease rate of the inspected PV panels as well. This estimation is mainly achieved through the use of a well-distinguished telltalle i.e., small plasticised paper of size 20×10cm. This telltale is characterised with a distinctive texture (Grid texture). By calculating the image Entropy of the evoked telltale, the concentration of dust particles deposited on the solar panels can be precisely calculated.