Keywords
wind farm; dynamic equivalent; data-driven; physical quantity; degree of similarity; spectral clustering
Abstract
Considering the characteristics of numerous wind power units, variable operating conditions, and complex collection grids as well as topological wiring in large-scale centralized wind farms, a data-driven similarity method is proposed to realize the equivalent modeling of such wind farms. Firstly, similarity is introduced to characterize the data features in the operating states of wind power generation units, and through similarity, data-driven clustering of wind power generation units is achieved. Secondly, the generation units within the same cluster are aggregated to obtain the equivalent parameters of the equivalent units, ultimately leading to the equivalent model of the wind farm. Finally, a case study of an offshore wind farm is used to simulate and verify the proposed method. The research results indicate that this method can effectively enhance the modeling efficiency and accuracy of wind farms.
DOI
10.19781/j.issn.1673-9140.2024.05.013
First Page
118
Last Page
128
Recommended Citation
WU, Yue; ZHU, Lin; HU, Yonghao; and LIU, Yang
(2024)
"Research on dynamic equivalent modeling of a wind farm using a data‑driven degree of similarity method,"
Journal of Electric Power Science and Technology: Vol. 39:
Iss.
5, Article 13.
DOI: 10.19781/j.issn.1673-9140.2024.05.013
Available at:
https://jepst.researchcommons.org/journal/vol39/iss5/13