SolarCrossFormer: Improving day-ahead Solar Irradiance Forecasting by Integrating Satellite Imagery and Ground Sensors
This addresses the need for high-resolution solar forecasts to integrate PV systems into the power grid, representing a domain-specific incremental improvement.
The paper tackles day-ahead solar irradiance forecasting by introducing SolarCrossFormer, a deep learning model that integrates satellite imagery and ground sensor data, achieving a normalized mean absolute error of 6.1% over a year-long dataset in Switzerland.
Accurate day-ahead forecasts of solar irradiance are required for the large-scale integration of solar photovoltaic (PV) systems into the power grid. However, current forecasting solutions lack the temporal and spatial resolution required by system operators. In this paper, we introduce SolarCrossFormer, a novel deep learning model for day-ahead irradiance forecasting, that combines satellite images and time series from a ground-based network of meteorological stations. SolarCrossFormer uses novel graph neural networks to exploit the inter- and intra-modal correlations of the input data and improve the accuracy and resolution of the forecasts. It generates probabilistic forecasts for any location in Switzerland with a 15-minute resolution for horizons up to 24 hours ahead. One of the key advantages of SolarCrossFormer its robustness in real life operations. It can incorporate new time-series data without retraining the model and, additionally, it can produce forecasts for locations without input data by using only their coordinates. Experimental results over a dataset of one year and 127 locations across Switzerland show that SolarCrossFormer yield a normalized mean absolute error of 6.1 % over the forecasting horizon. The results are competitive with those achieved by a commercial numerical weather prediction service.