Abstract
A testbed was designed and built for studying the impact of brush-based dry cleaning on glass samples and photovoltaic (PV) solar panels. A sand deposition shaking system was integrated into the testbed to continuously deposit dust on the brushed surfaces simulating the real environmental dust conditions for a 20-year equivalent time of cleaning. The cleaning efficiency was evaluated for different types of brush materials of nylon, cloth and silicon rubber foam. In this piece of work, the focus was mainly on brushing real solar panels, which already have anti-reflective coating applied to measure the output short circuit current, which is the IV characteristics dominant factor. While some materials had a notable impact on the solar panels, no permanent or significant negative impact was found to affect the solar panels as a result of the brush-based dry cleaning with the other materials. In fact, an enhancement in the maximum power output of solar panels cleaned with silicone rubber was around 1% from the unbrushed initial power output, which could be attributed to the created surface geometry. The silicon rubber foam - a novel brush material, with no previous research literature describing its use as a brush material - provides a low cost material and allows for a simple brush design, which could reduce the cost of the brush used in robotic cleaning systems, while providing highly effective, nonabrasive cleaning. (C) 2017 Elsevier Ltd. All rights reserved.