Cleaning Time Prediction Method of Photovoltaic Power Station Based on Short-Term Environmental Change

In view of the cleaning problems of photovoltaic (PV) plants, existing studies focus on the optimization of cleaning frequency or cleaning period equivalent. However, few consider the possible influence of short-term weather changes on the optimal cleaning time in this period. This paper proposes a cleaning time prediction algorithm for PV plants. By predicting the comprehensive efficiency of the system, the peak sun power, and the peak sun hour, a recursive algorithm was used to calculate the difference between the expected power generation income and the cleaning cost under different cleaning time stamps, so as to determine the optimal cleaning time in a specific period. The proposed method was verified in outdoor PV plants. The comparison and analysis of the calculation results based on the measured data and predicted data show that the proposed method can accurately predict the optimal cleaning time, control the cleaning income, and provide an effective means and basis for the later operation and maintenance of PV plants.