Multi-layer distributed multi-objective consensus algorithm for multi-objective economic dispatch of large-scale multi-area interconnected power systems. (15th October 2021)

• Record Type: Journal Article
• Authors: Yin, Linfei; Sun, Zhixiang
• Is Part Of: Applied energy. Volume 300(2021)
• Publication Date: 2021-10-15
• Subjects: Multi-layer distributed multi-objective consensus algorithm -- Large-scale multi-area interconnected power systems -- Parallel -- Distributed multi-objective algorithm -- Multi-objective economic dispatch; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2021.117391
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Calculation speed, information privacy, and multiple objectives are considered. Layering, multi-objective economic dispatch and distributed algorithm are combined. A framework of multi-layer distributed multi-objective economic dispatch is proposed. The multi-layer distributed multi-objective consensus algorithm is proposed. The proposed algorithm has higher information privacy and faster calculation speed. Abstract: Traditional multi-objective economic dispatch problems all apply centralized multi-objective optimization methods. However, with the rapid development of smart grids, issues such as system calculation speed, robustness, and information privacy have been increasingly vitally. Although the distributed multi-objective optimization method can solve the problems of system robustness and privacy when the scale of the interconnected power system expands and the number of agents increases, each agent independently optimizes its sub-problems and then exchanges information to complete the entire system optimization. The pressure of system communication calculation leads to reduce the overall calculation speed greatly. In response to the above problems, this paper proposes a multi-layer distributed multi-objective consensus algorithm. This method first calculates the optimal power generation of each area of each layer through the network topology and then calculates the power of each unit in each area in parallel according to the calculated optimal power generation. The analysis of the three system simulation results of IEEE118-bus, IEEE2154-bus, and Xinjiang Turpan 117-bus shows that the proposed method can solve the problems of multi-objective and information privacy, and realize the rapid solution of economic dispatch in large-scale multi-area interconnected power systems.