– NREL researchers developed Sup3rCC, an open-source generative machine learning model for simulating future energy-climate impacts
– The model enhances climate data resolution to understand climate change effects on wind, solar, and energy demand
– Sup3rCC overcomes computational challenges by using generative AI to generate detailed climate data 40 times faster than traditional methods, aiding in energy system planning and operation.
Researchers at NREL have developed Sup3rCC, an open source generative machine learning model that produces downscaled future climate data sets for understanding energy-climate impacts. This model can produce high-resolution data 40 times faster than traditional methods and will change the way energy systems are studied and planned.
The integration of energy system and climate research has been challenging due to computational limitations and lack of compatibility between existing datasets. Sup3rCC bridges this gap by leveraging generative AI techniques to incorporate climate data into energy system modeling, providing insights for decision-makers responsible for maintaining reliable energy systems.
Utilizing historical high-resolution data sets, Sup3rCC generates detailed temperature, humidity, wind speed, and solar irradiance data based on climate projections. This data can be used to study future renewable energy generation, changes in energy demand, and impacts on power system operations, enabling more accurate planning for future energy systems.
Sup3rCC increases the spatial and temporal resolution of global climate models by 25 times each horizontally and 24 times temporally, resulting in a 15,000-fold increase in total data. This model can be accessed on Amazon Web Services and used with NREL’s reV model to study wind and solar generation under various climate scenarios. Overall, Sup3rCC aims to improve energy-climate research and aid in planning future energy systems by providing fast, accurate, and detailed climate data for analysis.