Distributed model predictive control for economic dispatch of power systems with high penetration of renewable energy resources

Distributed generation entities such as renewable energy sources have posed great challenges on power system economic dispatch because of their output variability and stochasticity. Accordingly, operators need to lessen unpredictable changes in scheduled generation settings by fully utilizing available forecast information in the decision-making process. This paper proposes a closed-loop algorithm for solving economic dispatch at runtime while reducing potential deviations of generation schedules. At first, a traditional centralized approach that addresses the economic dispatch problem is presented with discussions on potential enhancement enabled by model predictive control (MPC) techniques. The MPC application makes it possible for operators to address the concern of variability and stochasticity. This paper develops a dual decomposition-based distributed model predictive control (DDMPC) strategy that is compatible with consensus techniques. Different advantages of the proposed DDMPC are highlighted throughout the paper and are analyzed through simulations. The simulation results validate the advantages of the proposed DDMPC approach by comparing it with traditional techniques for economic dispatch and by another distributed method based on MPC.