A data mining‐based method for mining key factors affecting transient voltage stability for power systems with renewable energy sources. Issue 4 (13th October 2021)

• Record Type: Journal Article
• Authors: Huang, Dan; Sun, Huadong; Zhang, Jian; Zhao, Shanshan; Zhou, Qinyong
• Is Part Of: IET generation, transmission & distribution. Volume 16:Issue 4(2022)
• Page Start: 617
• Page End: 628
• Publication Date: 2021-10-13
• Subjects: Electric power production -- Periodicals; Electric power transmission -- Periodicals; Electric power distribution -- Periodicals; 621.3105
• DOI: 10.1049/gtd2.12314
• Languages: English
• ISSNs: 1751-8687

Abstract

Abstract: Increasing penetration of renewable energy sources (RESs) into power systems has brought new challenges to guarantee transient voltage stability (TVS) of the system, due to complex and different characteristics of the RES compared with the synchronous generator. The related theories to the TVS for power systems with RES (PSRESs) are incomplete, and it is difficult to construct accurate physical model of the PSRES by using traditional TVS analysis method. Here a novel data mining‐based approach for extracting key factors that affect the TVS of a PSRES is put forward. The original Relief algorithm is modified to deal with the imbalance of sample size between stable and unstable samples of the practical data set and improve the calculation accuracy. Then, a data mining scheme based on the modified Relief algorithm is presented to acquire key factors affecting the TVS. With the proposed scheme, the influence degrees of different factors on the TVS can be evaluated quantitatively by their weighting values, and then the key factors as well as the influence patterns can be determined. Test results which are conducted on the modified IEEE‐39 test system with RESs are presented to demonstrate the accuracy and efficiency of the proposed method.