Cross-correlation based cloud motion estimation for short-term solar irradiation predictions

Abstract

A technique for cloud motion estimation for intra-hour Irradiance prediction using ground-based sky images is presented. A sequence of whole sky images is processed to identify cloud motion vectors. A Cross-correlation based cloud boundary tracking method was utilized to track the individual cloud boundary point movement from one image frame to next image frame. Using the cloud boundary positions in each image frame and image capturing frequency, the velocity vectors were defined. Then using the velocities, the clouds which are moving in the direction of the sun were identified. For these clouds, the time taken to reach the sun location were calculated and subsequent irradiance fluctuations were predicted 2minites before the real time.