Distributed Model Predictive Control for Hybrid Energy Resource System With Large-Scale Decomposition Coordination Approach

Abstract

Due to the large-scale and distributed characteristics of increasing renewable energy resources, dynamic economic emission dispatch (DEED) of hybrid energy resource system becomes more and more important in the power system operation. This paper proposes a distributed model predictive control (DMPC) method for hybrid energy resources system of dynamic economic optimal dispatch with large-scale decomposition coordination approach. First, the DEED model of hybrid energy resources is converted into predictive control model, which can provide rolling optimization mechanism for dealing with intermittent energy resources optimization. Second, predictive control model is decomposed into several subsystems with Lagrangian multipliers for coordinating those subsystems, which can greatly decrease the computational complexity. Third, due to the randomness or uncertainty of intermittent power generation, model predictive control can dynamically optimize random or uncertainty problem with rolling optimization mechanism. Furthermore, adaptive dynamic programming is utilized to solve those subsystem optimization problems, which can optimize the random or uncertain problem in real-time condition. In the optimization process, probability constraint is converted into deterministic constraint with its probability density function, and system load balance can be properly handled with coupled coarse-fine constraint-handling technique. According to the obtained results in the case studies, the proposed DMPC can optimize the DEED of hybrid energy resources well combining with the large-scale decomposition-coordination approach, while greatly decreasing the optimization complexity and computation time, which reveals that the proposed method can provide an alternative way for solving the DEED problem of hybrid energy resources.