In order to solve the problem of large error of power generation power of photovoltaic power generation system under different conditions, a new method for predicting the power generation power of photovoltaic power generation system was proposed. By analyzing the structure of the photovoltaic power generation system, the influencing factors of the power generation power of the photovoltaic power generation system were studied, the sample sources were selected with the season and weather type as historical samples, and similar data points were found in the historical database for the predicted daily time-of-time meteorological data provided by the meteorological department.
The experimental results show that the average absolute percentage errors of the proposed method in predicting the power generation power of solar photovoltaic power generation system under different conditions are 1.47% and 6.39%, respectively, and the relative change of the prediction error of power generation power of photovoltaic modules under comprehensive abnormal conditions is less than 1%, which proves that the proposed method meets the actual prediction requirements.
