A distributed robust optimization approach for the economic dispatch of flexible resources. (January 2021)

• Record Type: Journal Article
• Authors: Chang, Xinyue; Xu, Yinliang; Sun, Hongbin; Khan, Irfan
• Is Part Of: International journal of electrical power & energy systems. Volume 124(2021)
• Publication Date: 2021-01
• Subjects: Flexible resources -- Renewable energy -- Economic dispatch -- Distributed optimization -- Robust optimization; 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.2020.106360
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A robust optimization model is proposed to handle economic dispatch problems with uncertainties. An improved distributed approach with self-adaptive step-size and a polynomial filter is proposed. It has faster convergence performances and robustness to random communication failures. Abstract: Power systems are confronted with prodigious challenges of scheduling and operation incurred by the high penetration of distributed renewable energy with intermittency and uncertainty. Therefore, this paper proposes an improved distributed robust optimization approach with self-adaptive step-sizes based on the line search method and a polynomial filter, to minimize the overall costs of flexible resources including conventional generators, energy storage systems, renewable energy curtailments, deferrable loads and tie-line power exchanges, while considering various constraints, such as supply-demand power balance, line congestion constraints and power output limits. Numerical case studies conducted in a modified IEEE 14-bus system and a modified IEEE 118-bus system demonstrate the reliability, robustness and extensibility of the proposed approach. In addition, the effectiveness and accuracy of the proposed distributed robust optimization approach are validated through comparisons with the traditional centralized gradient method and the convergence performance is better in contrast to other distributed optimization algorithm.