An Image-Based Method for Soiling Quantification

Soiling due to dust accumulation reduces solar energy conversion efficiency in solar photovoltaic (PV) power generation. In this study, a preliminary investigation of using image analysis to quantify soiling on a surrogate surface is carried out. The effect of several camera settings is investigated in a lab environment where ground-truth dust loading can be measured gravimetrically. Specifically, a custom black-and-white pattern with various levels of dust loading is imaged by visible light photography at various shutter speed, aperture, and focus conditions. We develop an image analysis method to extract black and white regions from the photo, based on which a metric called black/white ratio (BWR) is calculated. Quantitative relations between BWR and the dust loading level, namely dust mass per unit area of the black-and-white pattern, are investigated under different conditions. The results from this study demonstrate the feasibility of using image attributes of a dust-covered solar module, or a surrogate surface, to quantify soiling loss and indicate the need for further research to understand the complex effects of various factors on this approach.