Optimal Scheduling and Control of Microgrid Power Flow

The operation of renewable microgrids is undergoing rapid transformations due to increasing renewable resources penetration, making their energy management and control more crucial for economic and reliability goals. The increasing penetration of solar resources introduces a significant degree of intermittency and uncertainty. As a measure to increase reliability in the face of such uncertainty, future microgrids will require operational flexibility including ramping capacity and the ability to swiftly respond to grid conditions. Real-time feedback control is known for handling uncertainty and can play an instrumental role in real-time scheduling update along with other non-intermittent reserves. Motivated by this changing prospect, this dissertation proposes multiple techniques for optimal microgrid control, energy management, and scheduling in the presence of intermittent renewable resources, reserve non-renewable resources, and energy storage systems. The presented approaches consider microgrid optimization problems from the viewpoint of a single microgrid as well as multiple cooperating microgrids under a variety of energy cost structures and operating limitations imposed by resource constraints and the grid. Additionally, we present an approach of addressing both short-term scheduling and real-time power control in a unified model predictive control framework where the microgrid controller operates at two separate time scales. Techniques are proposed to relax the non-convex resource scheduling problem and enable solving the MPC at update rates comparable with renewable generation and demand variability time scales. To facilitate testing of different microgrid controllers with fast update rates, a remote hardware-in-the-loop microgrid testing setup is designed and utilized for testing the controllers proposed in this work. The proposed control and scheduling approaches are developed using data and models from real-world microgrids and some of the techniques are implemented in an actual microgrid in California with solar energy generation and energy storage