A distributed PageRank-based dynamic partition algorithm to improve distributed energy storages participation in frequency regulation. (August 2023)

• Record Type: Journal Article
• Authors: Lin, Yujun; Lei, Xuanang; Yang, Qiufan; Zhou, Jianyu; Chen, Xia; Wen, Jinyu
• Is Part Of: International journal of electrical power & energy systems. Volume 150(2023)
• Publication Date: 2023-08
• Subjects: Energy storage -- Frequency regulation -- Distributed partition algorithm -- Virtual power plant; 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.2023.109105
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A real-time, distributed partition algorithm is proposed to enhance the robustness. Distributed ESs are aggregated into VPPs to reduce the circulating currents. The concept of modularity is utilized to update the weight coefficients of partitioning. The electrical distance is considered in partitioning to further reduce the power loss. Abstract: In high renewable energy sources penetrated power systems, a large number of distributed small-capacity energy storages (ESs) are expected to form virtual power plants (VPPs) and provide comparable frequency regulation capacity to conventional power plants. In this paper, a dynamic and distributed partition method is proposed to cluster ESs into VPPs by selecting the electrical distance and local power demands as the index for partitioning. Utilizing the partition indices, the Personalized PageRank (PPR) scores are used to measure the similarity of all ESs with respect to other ESs in parallel. Then Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is utilized to partition the ESs into different VPPs. Within the VPPs, the secondary frequency control strategy is adopted to achieve the frequency regulation and state of charge (SoC) balancing. Finally, the performance comparison of the proposed partition method with several classic clustering algorithms is conducted to validate the efficiency and the quality of the partition. And numerical studies are performed in an IEEE 39-bus system to verify that the proposed control method can eliminate the frequency deviation and reduce power losses.