Effective Prediction of Transmission of Solar Irradiance through Dusty Solar Panels using Atmospheric Aerosol Data for Lahore, Pakistan
LUMS, LAHORE, Pakistan

The effect of soiling on performance of PV panels is difficult to predict as the process is highly dependent on the site specific climate parameters. Among various climate parameters, highest correlation with soiling losses have been found for the airborne particulate matter (PM) concentration and the duration of dry period (time between events of significant rain). Here, we study the effect of the size distribution of the PM and show that it can significantly modulate the PV soiling loss for a given mass concentration of PM. The PM size distribution from last 15 year satellite aerosol data for Lahore is used to predict the PV soiling loss using model based on Mie scattering theory and Lambert-Beer law. The soiling loss is experimentally measured for Lahore to validate the model prediction. For a moderate soiling mass density of 1.5 g/m2, we show that variations in PM size distribution can modulate the transmission loss by ~7%. A variation in the soiling across the year for Lahore is predicted to be maximum of 19% in June and 12% in January for a constant mass density of 1.5 g/m2. We further apply the model to validate some of the reported soiling losses for particle size distributions extracted from naturally soiled dust samples from PV panels. A nice match is found between the experimental results and model based predictions.