Development of a static equivalent model for Korean power systems using power transfer distribution factor-based k-means++ algorithm

Bae Geun Lee, Joonwoo Lee, Soobae Kim

Research output: Contribution to journalArticlepeer-review

4 Scopus citations

Abstract

This paper presents a static network equivalent model for Korean power systems. The proposed equivalent model preserves the overall transmission network characteristics focusing on power flows among areas in Korean power systems. For developing the model, a power transfer distribution factor (PTDF)-based k-means++ algorithm was used to cluster the bus groups in which similar PTDF characteristics were identified. For the reduction process, the bus groups were replaced by a single bus with a generator or load, and an equivalent transmission line was determined to maintain power flows in the original system model. Appropriate voltage levels were selected, and compensation for real power line losses was made for the correct representation. A Korean power system with more than 1600 buses was reduced to a 38-bus system with 13 generators, 25 loads, and 74 transmission lines. The effectiveness of the developed equivalent model was evaluated by performing power flow simulations and comparisons of various characteristics of the original and reduced systems. The simulation comparisons show that the developed equivalent model maintains inter-area power flows as close as possible to the original Korean power systems.

Original languageEnglish
Article number6663
JournalEnergies
Volume13
Issue number24
DOIs
StatePublished - 2 Dec 2020

Keywords

  • Bus aggregation
  • Critical energy infrastructure information
  • K-means and k-means++ clustering algorithm
  • Korean power systems
  • Line loss compensation
  • Power system equivalent
  • Power transfer distribution factor
  • Voltage level selection

Fingerprint

Dive into the research topics of 'Development of a static equivalent model for Korean power systems using power transfer distribution factor-based k-means++ algorithm'. Together they form a unique fingerprint.

Cite this