Data-driven based optimal distributed frequency control for islanded AC microgrids

With the great stride of clean energy technology, an increasing number of distributed generations are connected to microgrids. However, the heterogeneity and uncertainty of distributed generation may cause frequency instability in microgrids. This paper proposes a data-driven based distributed secondary frequency controller for islanded microgrids without requiring the accurate model parameters. Motivated by iterative learning algorithm, a distributed dual heuristic dynamic programming control framework is established to derive an optimal control action. The frequency controller quickly restores frequency while guaranteeing accurate active power sharing. Furthermore, convergence of the proposed algorithm and stability of the data-based error system are analyzed. Simulation examples are carried out to validate the effectiveness of our designs. Compared with conventional distributed frequency control, simulation result shows the proposed controller has a superior dynamic performance.