Distributionally Robust Neural Control of High-Renewable Islanded Microgrids for Stability Enhancement

Robustly stable control may not exist for high-renewable islanded microgrids (IMGs). This naturally raises the question of how to control IMGs with probabilistic guarantees of stability. To this end, we develop a distributionally robust (DR) stable and safe secondary control method for high-renewable IMGs, incorporating a neural control law derived based on Lyapunov and barrier functions and DR chance-constrained optimization theory, and a data-driven implementation architecture to update the controller using up-to-date renewable uncertainty information. Numerical simulation results demonstrate the efficacy and superiority of the proposed method.