An Improved Density Peak Clustering Method for Power Load Anomaly Detection

Han Jianfeng, Chen Xinxin, Wang Li, Fan Shurui and Zhang Yong
Additional contact information
Han Jianfeng: Tianjin University of Commerce, China
Chen Xinxin: Tianjin University of Commerce, China
Wang Li: Hebei University of Technology, China
Fan Shurui: Hebei University of Technology, China
Zhang Yong: Tianjin University of Commerce, China

International Journal of Intelligent Information Technologies (IJIIT), 2025, vol. 21, issue 1, 1-23

Abstract: Aimed at the sensitivity of the cut-off distance parameter, the difficulty in accurately determining cluster centers, and the criteria unreliable for anomalous data selection, an improved density peak clustering combined with local outlier factor (LOF) algorithm is proposed for power load anomaly detection. Firstly, the density entropy algorithm is constructed to achieve the adaptive selection of cut-off distance, and the statistical quartile method is used for threshold setting to achieve the adaptive cluster center selection. Secondly, the LOF algorithm was developed to increase the rate of anomaly detection and address the absence of criteria for selecting anomalous data. Finally, the proposed algorithm is experimentally validated on a real power dataset, and the results demonstrate that the improved algorithm could be able to effectively detect the power loads anomaly data.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJIIT.385734 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jiit00:v:21:y:2025:i:1:p:1-23

Access Statistics for this article

International Journal of Intelligent Information Technologies (IJIIT) is currently edited by Vijayan Sugumaran

More articles in International Journal of Intelligent Information Technologies (IJIIT) from IGI Global Scientific Publishing
Bibliographic data for series maintained by Journal Editor ().