Short-term prediction of localized cloud motion using ground-based sky imagers

摘要

Fine-scale short-term cloud motion prediction is needed for severalapplications, including solar energy generation and satellite communications.In tropical regions such as Singapore, clouds are mostly formed by convection;they are very localized, and evolve quickly. We capture hemispherical images ofthe sky at regular intervals of time using ground-based cameras. They provide ahigh resolution and localized cloud images. We use two successive frames tocompute optical flow and predict the future location of clouds. We achieve goodprediction accuracy for a lead time of up to 5 minutes.