SolarBench
A global benchmark for solar energy nowcasting
SolarBench combines harmonized multimodal data with a Python toolbox for reproducible data access, processing, model development, and evaluation. Our goal is to facilitate consistent comparison of solar nowcasting methods across locations and improve understanding of their performance under diverse climates, cloud regimes, and PV systems.
We envision SolarBench as a shared, extensible infrastructure that brings the community together around open data and standardized benchmarks to accelerate methodological innovation and advance reliable solar nowcasting for large-scale solar integration.
Using SolarBench in your research? Please see our citation guidelines.