Data-Driven Soiling Detection in PV Modules

Soiling is the accumulation of dirt in solar panels that leads to a decreasing trend energy yield and may be cause vast revenue losses. The effect soiling can reduced by washing panels, which is, however, procedure non-negligible cost. Moreover, monitoring systems are often unreliable or very costly. We study problem estimating ratio photovoltaic (PV) modules, i.e., real power output would produced if were clean. A key advantage our algorithms they estimate soiling, without needing train on labeled data, periods explicitly each park, relying generic analytical formulas do not take into account peculiarities installation. consider as input time series comprising minimum set measurements available most PV park operators. Our experimental evaluation shows we significantly outperform current state-of-the-art methods for ratio.