SIMULATION AND EXPERIMENTAL VALIDATION OF AN ELECTROSTATIC DRUM FOR CLEANING PHOTOVOLTAIC PANELS IN SAUDI ARABIA

Abstract

Saudi Arabia is dependent on oil and natural gas for energy generation and seawater desalination, which is the main source of drinking water in the country. Given Saudi Arabia’s abundant solar energy, the national renewable energy program aims to increase the share of renewable energy by 9500 MW in an effort to decrease carbon dioxide emissions and diversify the sources of government revenue. However, dust accumulation on photovoltaic (PV) modules of the region presents a challenge to solar power plants. Therefore, cleaning PV panels is essential to have the benefit of high insolation and mitigate the impact of dust accumulation. Among selfcleaning methods, electrodynamic screens (EDSs), embedded under the solar panel glass, have the highest increase of annual energy production in the dry regions of Saudi Arabia. However, EDSs can only remove 16-33% of the accumulated dust on PV panels, leaving residual dust that cannot be removed after EDS activation. This work analyzes five dust samples from near potential PV power plants locations in Saudi Arabia. The dust particles are analyzed for size distribution, shape, elemental composition, and chemical composition. After thorough cleaning with EDS, the residual dust is collected and characterized in the same manner as the original samples. Current EDS devices suffer from inherent problems such as the shadow effect of electrodes, a weak electric field between electrodes, and high initial cost. An electrostatic drum for PV cleaning is proposed to overcome the aforementioned problems. COMSOL simulation is used to identify optimum drum parameters such as drum diameter, drum to PV gap, drum thickness and applied voltage. The cleaning performance of the electrostatic drum is compared to existing EDS. Based on the optimal drum parameters from the simulation, a drum prototype was fabricated to examine the dust removal performance. The drum prototype is validated using five dust samples and the drum prototype cleaning performance and EDS cleaning performance from the literature are compared.