Abstract

In this paper, reinforcement learning techniques are proposed for the control of autonomous microgrids. A type of approximate dynamic programming method is used to solve the Bellman equation, namely heuristic dynamic programming. The proposed control strategy is based on actor-critic networks. The control strategy is designed using a dynamic model of islanded microgrids and makes use of an internal oscillator for frequency control. The proposed control technique is based on a value iterations algorithm which is implemented online. Using only partial knowledge of the microgrid dynamics, the simulation results showed that the proposed control technique stabilizes the system and is robust to the load disturbances.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call