A demo showing how AI agents can coordinate grid flexibility actions in response to predicted feeder overloads. Built for the AI Agent Hackathon x DEG finalist round.
This demo simulates a three-agent system that:
- Predicts feeder overloads
- Discovers & selects DER flexibility using a simple optimisation step
- Dispatches chosen actions through a mock Beckn-style flow It reflects the core orchestration loop described in the DEG hackathon materials and the Demand Flexibility flow (discover → confirm → status) defined in the official guide.
- Prediction Agent: forecasts load and triggers flexibility requests
- Optimisation Agent: selects the best DERs using a small knapsack-style model
- Dispatch Agent: activates chosen DERs and logs decisions
Check out the demo of Grid Flex Agent on YouTube:
Click the link to watch the full video on YouTube.