Deep Learning Multi-Horizon Irradiance Nowcasting: A Comparative Evaluation of Three Methods for Leveraging Sky Images

TL;DR

This study compares three methods of integrating all-sky imager images into deep learning models for short-term solar irradiance forecasting, highlighting the effectiveness of engineered features.

Contribution

It demonstrates that using engineered ASI features improves forecast accuracy without requiring complex spatial deep learning architectures.

Findings

Engineered feature maps outperform raw images in forecasting accuracy.

Aggregation of engineered features into time-series inputs yields superior results.

The approach simplifies integration of sky images into deep learning models.

Abstract

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 2D 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 2D feature maps into time-series input. Each of the three methods were then used as part of a DL model trained on a high-frequency, 29-day dataset to generate multi-horizon forecasts of global horizontal irradiance up to 15 minutes ahead. The models were then evaluated using root mean squared error and skill score on 7 selected days of data. Aggregated engineered ASI features as model input yielded superior forecasting performance, demonstrating that 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.