Distributed event-triggering algorithm with uncoordinated step sizes for economic dispatch problem over unbalanced directed network. (February 2023)

• Record Type: Journal Article
• Authors: Ji, Lianghao; Shen, Min; Yang, Shasha; Li, Huaqing
• Is Part Of: International journal of electrical power & energy systems. Volume 145(2023)
• Publication Date: 2023-02
• Subjects: Event-triggering -- Unbalanced directed network -- Economic dispatch -- Uncoordinated step; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2022.108601
• Languages: English
• ISSNs: 0142-0615

Abstract

Abstract: This paper studies the economic dispatch problem of the smart grid based on unbalanced directed graph. An event triggering mechanism is introduced to save communication resources, and each generator broadcasts its information to its neighbors when the event-triggering condition is satisfied. A new distributed Lagrange algorithm with uncoordinated steps is proposed to accelerate the convergence of the algorithm. The row random matrix is introduced to balance the imbalance of the directed network, and convergence of the algorithm is analyzed by the small gain theorem as well. The theoretical analysis shows that when the cost function of the generator is convex and smooth, and the uncoordinated step size satisfies certain conditions. The algorithm can linearly get the optimal solution of economic dispatch problem. Finally, Simulation examples prove the effectiveness and applicability of the proposed algorithm. Highlights: A new distributed event-triggering condition is proposed for consensus-based distributed algorithms. The proposed algorithm allows step length to be variable, and the step length can be changed over time. A new distributed Lagrangian momentum algorithm based on row stochastic matrix is proposed to eliminate imbalances on a directed unbalanced network, which are easier to implement in a distributed manner than by using double stochastic matrix.