Optimized Soiling Cleaning Schedules for Photovoltaic Systems Using Weather Driven Modeling

This experimental study evaluated the combined effects of dust deposition and elevated module temperature on residential photovoltaic performance in Baqubah, Diyala. Two matched photovoltaic module strings were operated in parallel, one kept permanently clean as a reference and the other exposed to natural soiling. Solar irradiance, module temperature, electrical power, and gravimetric dust accumulation were measured at 1 min resolution, supported by weekly optical inspections. Hourly aggregation captured behavior, while optical losses were quantified from measured in-plane transmittance and associated changes in current–voltage characteristics. Dust loading from 12 to 72 mg m -2 reduced optical transmittance and produced measurable voltage losses. Thermal effects were characterized using a temperature coefficient of −0.39% for each 1 °C increase. Impacts on total energy yield, household self-consumption, and self-sufficiency were evaluated under alternative cleaning strategies, including fixed cleaning intervals of 2, 7, and 14 d and an event-driven schedule during peak dust months from April to July. A coupled response-surface model described normalized maximum power point as a function of dust density and temperature rise, with uncertainty propagation using block bootstrap and Monte Carlo methods. Results showed marked combined impacts, including reduced daily energy generation, earlier inverter clipping, deeper battery cycling, and lower self-consumption and self-sufficiency, with improved outcomes associated with more frequent cleaning and temperature-aware operation. • 8 kWp PV–6 kWh BESS tested under Diyala’s hot semi-arid climate • Dust 12–72 mg m -2 cuts MPP and raises ΔT with compounding losses • β of −0.39 % °C -1 governs temperature derating of PV power • Cleaning cadence steers E SC and E SS across seasons