Economic Dispatch at a Combined Heat and Power Plant

Optimal solution of The Operator’s Dilemma with Artificial Intelligence.

As Distributed Energy Resources continue to penetrate the power markets at the rapid pace, a new end-to-end solution is needed to maintain system reliability and meet the demand. Our team developed a proof-of-concept to efficiently schedule generators and to maximize the profit. Take a look!

AI is given control of two 25 MW generators. It can set the
The information AI receives includes demands for high-grade heat, low-grade heat, electricity and cooling.
Reward is calculated based on the net energy cost: export electricity revenue – (gas cost + import electricity cost)
AI is roughly following the electricity price. When the price is high, it generates additional electricity for sale.
Results

Economic Dispatch

As you can see from the provided Tableau workbook, the model is roughly following the price trend. Even in this simplistic model, it was able to predict and attempt to maximize profit. Our research aimed at applying this technique for Grid Balancing with high renewables penetration.

See Tableau Workbook

* This work is based on Adam Green’s CHP model available here.