As renewable energy sources like wind and solar become more common, managing the power grid has grown more complicated. Researchers at the University of Virginia have created a new solution: an artificial intelligence (AI) model that helps manage the uncertainties of renewable energy production and electric vehicle usage, making power grids more reliable and efficient.
Multi-Fidelity Graph Neural Networks: A New AI Solution
The new model uses multi-fidelity graph neural networks (GNNs), a type of AI that enhances power flow analysis, which ensures electricity is safely and efficiently distributed. The “multi-fidelity” aspect allows the AI to use a lot of lower-quality data while still benefiting from some high-quality data. This combination speeds up training and improves the model’s accuracy and reliability.
Enhancing Grid Flexibility for Real-Time Decision Making
By utilizing GNNs, this model can adjust to different grid setups and withstand changes like power line failures. It addresses the “optimal power flow” problem, which determines how much power should come from various sources. As renewable energy introduces uncertainty in production and electric vehicles increase demand unpredictability, traditional grid management methods often struggle to adapt. The new AI model combines detailed and simplified simulations to find solutions quickly, improving grid performance even under unpredictable conditions.
Negin Alemazkoor, the lead researcher, stated, “With renewable energy and electric vehicles changing the landscape, we need smarter solutions to manage the grid. Our model helps make quick, reliable decisions, even when unexpected changes happen.”
- Press release – University of Virginia
