Real-time Cost Optimisation for Power Management in Microgrids Using Multi-Agent Control

This paper presents the application of Multi-Agent systems with model predictive control for an AC microgrid, implementing real-time generation cost minimisation. The MultiAgent control is tested against heuristic optimisation methods. It is shown that the control can reduce costs without the need of a central controller and in times of the order of milliseconds, making online optimisation possible. A test microgrid and the primary control were simulated in an OPAL-RT while the secondary control developed in Java manages the system over TCP/IP.