3-D numerical simulation and experimental study of PV module self-cleaning based on dew formation and single axis tracking

Despite the global acceptance and rapid market penetration of PV power plants and recent advancements in increasing panels electrical efficiency, soiling is still one of the most critical challenges that deteriorate the performance. This study proposes a unique method for PV panel self-cleaning based on dew formation on panel's active surface and with aid of Single Axis Tracking (SAT) mechanism. Both experimental study and numerical simulations were performed in this research to highlight the effectiveness of self-cleaning method. Such approach is based on rotation of panels toward ground during night in an inclined position for simultaneous dew formation and cleaning of panels via water droplets runoff and prevention of dust accumulation. Experimental results exhibit a maximum loss of 5.9% in panel efficiency during a 94 days testing period while for the fixed PV and SAT without night position losses were measured to be 33.2% and 11.9% respectively. Using a validated CFD model, a new PV glazing geometry was proposed to increase dew formation for more effective self-cleaning. CFD numerical results demonstrate that the amount of collected dew mass formed on the panel can be increased by about 86 % while dew film thickness increased by about 67% compared to conventional flat glaze configuration. Overall, the proposed system outperforms its counterparts based on a continuous self-cleaning strategy on a nightly basis which does not require any additional capital and labor costs, energy consumption and water resources.

Automated summary

This study proposes a self-cleaning method for PV panels based on dew formation and Single Axis Tracking (SAT) mechanism. Experimental and numerical results demonstrate the effectiveness of this method in maintaining panel efficiency and preventing dust accumulation. A new PV glazing geometry is also suggested to enhance dew formation for more effective self-cleaning.