HelioNet‐IR: Combining Infrared and Visible Satellite Images for Solar Irradiance Forecasting in the Early‐Morning Hours
Open Access
Abstract
Forecasting of solar irradiance is crucial for integrating large shares of photovoltaics into the electricity grid. On timescales up to a few hours ahead, satellite‐based (SAT) forecasts can significantly improve upon numerical weather predictions (NWPs). Conventional SAT methods derive cloud motion vectors (CMV) from consecutive images and extrapolate these to forecast future cloud situations. A semi‐empirical version of the Heliosat method is widely used to retrieve global horizontal irradiance from visible‐range satellite images via the cloud index (CI) as key parameter. When derived from the visible spectrum, CI computation is restricted to daylight hours, and before sunrise, no SAT forecast is available for the early morning. Here, we present HelioNetIR, a convolutional neural network with UNet architecture to forecast CI derived from Meteosat second generation images without strictly relying on sun illumination. To do so, input CI is complemented with two additional infrared channels. Forecasts of HelioNetIR are benchmarked against different CMV and NWP models and its purely CI‐based predecessor HelioNetVIS over one full year (2024) with lead times (LTs) up to 4 hr and 15‐min resolution. HelioNetIR can increase SAT‐forecast availability from 22% to 100% for forecasts initiated before 8 AM. It notably outperforms NWP for , reducing root mean square error by over 40% within the first hour. During daytime, reference SAT models are outperformed for all LTs considered.Keywords
Funding Information
- Deutsche Bundesstiftung Umwelt
This publication has 31 references indexed in Scilit:
- Satellite‐Based Solar Irradiance Forecasting: Replacing Cloud Motion Vectors by Deep LearningSolar RRL, 2024
- The added value of combining solar irradiance data and forecasts: A probabilistic benchmarking exerciseRenewable Energy, 2024
- Extending intraday solar forecast horizons with deep generative modelsApplied Energy, 2024
- Omnivision forecasting: Combining satellite and sky images for improved deterministic and probabilistic intra-hour solar energy predictionsApplied Energy, 2023
- Estimating surface solar irradiance from satellites: Past, present, and future perspectivesRemote Sensing of Environment, 2019
- Estimation of all-sky instantaneous surface incident shortwave radiation from Moderate Resolution Imaging Spectroradiometer data using optimization methodRemote Sensing of Environment, 2018
- Comparing support vector regression for PV power forecasting to a physical modeling approach using measurement, numerical weather prediction, and cloud motion dataSolar Energy, 2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Satellite-Based Irradiance and Power Forecasting for the German Energy Market2013
- Cloud detection in Meteosat Second Generation imagery at the Met OfficeMeteorological Applications, 2011