Deep Learning Multihorizon Irradiance Nowcasting: A Comparative Evaluation of Three Methods for Leveraging Sky Images

Abstract

In this article, we investigate three distinct methods of incorporating all-sky imager (ASI) images into deep learning (DL) irradiance nowcasting. The first method relies on a convolutional neural network (CNN) to extract features directly from raw RGB images. The second method uses state-of-the-art algorithms to engineer 2-D feature maps informed by domain knowledge, e.g., cloud segmentation, the cloud motion vector, solar position, and cloud base height. These feature maps are then passed to a CNN to extract compound features. The final method relies on aggregating the engineered 2-D feature maps into time-series input. Each of the three methods was then used as part of a DL model trained on a high-frequency 29-day dataset to generate multihorizon forecasts of global horizontal irradiance up to 15 min ahead. The models were then evaluated using root-mean-squared error and skill score on seven selected days of data. Aggregated engineered ASI features as model input yielded superior forecasting performance, demonstrating that the integration of ASI images into DL nowcasting models is possible without complex spatially ordered DL architectures and inputs, underscoring opportunities for alternative image processing methods as well as the potential for improved spatial DL feature processing methods.