抄録
Photovoltaic power generation systems are expected to be widely used because of their cleanness in the environment. One of their disadvantages is the unexpected fluctuation of generated output caused by the weather conditions. This paper proposes to forecast the power fluctuation of the photovoltaic power generation by analyzing clouds in the sky through the image processing technique. First, the velocity of the cloud in the image is calculated by the comparison between two sequential images of the cloud, and then, the weather conditions are forecasted by examining the color values of the image along the direction of the cloud movement. Experimental results, the weather forecast 60, 90 and 120 seconds ahead and the estimation of the time required for the solar radiation to fluctuate on a large scale, show the possibility of the forecast of the power fluctuation of the photovoltaic power generation.
