Enhancing Transient Dynamics Stabilization in Islanded Microgrids Through Adaptive and Hierarchical Data-Driven Predictive Droop Control

The transient dynamics of microgrids is primarily impacted by low-inertia power electronic interfaces, energy generation of distributed energy resources (DERs), load demand fluctuations, and the control strategies employed for system integration. This paper focuses on the enhancement of the transient dynamics to achieve a stable steady-state operation for the microgrid by minimizing the overall islanded system’s frequency deviations. A modularized physics-informed sparse identification technique is developed for system identification that can accurately predict the future states of the microgrid with interconnected DERs. The data-driven prediction model is then incorporated into the model predictive control framework to generate an optimal control input that can augment with conventional droop control for frequency stabilization. Given the inherent fluctuations in typical microgrid operations, stemming from factors such as varying load demands, weather conditions, and other variables, reachability analysis is also performed in this work. By doing so, we aim to facilitate the design of data-driven models and implement effective control strategies for microgrids subject to disturbances, and thus, ensuring the safety, reliability, and efficiency of microgrids across a wide range of operating conditions. The effectiveness of the proposed approaches is verified in this paper with numerical examples where the developed controller is tested in various worst-case scenarios generated by the reachable set computation.