Skip to main content

NowcastAI

SECTOR: Energy, Renewable Energy
TECHNOLOGY USED: HPC, Machine learning
COUNTRY: Germany

Short description of the business experiment

The NowcastAI project, led by alitiq GmbH in collaboration with Ventus Tech GmbH, aims to develop an AI-driven short-term weather forecasting model, with a focus on solar irradiance prediction. Designed for energy traders and grid operators, it will provide highly accurate, minute-level forecasts for real-time energy trading and grid optimisation. The model integrates satellite imagery, radar data, and numerical weather predictions within a machine learning framework, utilising high-performance computing (HPC) to process large datasets at kilometre-scale resolution. Unlike traditional models, NowcastAI’s data-driven approach ensures more precise, responsive forecasts. Ventus Tech’s HPC expertise is crucial, enabling efficient resource utilisation, performance optimisation, and hardware deployment—key factors for overcoming alitiq’s lack of HPC experience and ensuring the success of the project.

 

Organisations involved: 

End User: alitiq GmbH

Domain expert and HPC expert: Ventus Technologies GmbH