EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Contrastive-Active Transfer Learning-Based Real-Time Adaptive Assessment Method for Power System Transient Stability

Jinman Zhao, Xiaoqing Han, Chengmin Wang, Jing Yang, Gengwu Zhang

Sensors · 04.08.2024 · 24 (15) · 5052

DOI: 10.3390/s24155052 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 04.08.2024. Verfasst von Jinman Zhao und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Electrical and Power Engineering, Taiyuan University of Technology, 79 Yingze West Street, Taiyuan 030024, China. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundamental Research Program of Shanxi Province, grant number, Key Laboratory of Control of Power Transmission and Conversion (SJTU), Ministry of Educa-tion, Fundamental Research Program of Shanxi Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The transient stability assessment based on machine learning faces challenges such as sample data imbalance and poor generalization. To address these problems, this paper proposes an intelligent enhancement method for real-time adaptive assessment of transient stability. In the offline phase, a convolutional neural network (CNN) is used as the base classifier. A model training method based on contrastive learning is introduced, aiming to increase the spatial distance between positive and negative samples in the mapping space. This approach effectively improves the accuracy of the model in recognizing unbalanced samples. In the online phase, when real data with different distribution characteristics from the offline data are encountered, an active transfer strategy is employed to update the model. New system samples are obtained through instance transfer from the original system, and an active sampling strategy considering uncertainty is designed to continuously select high-value samples from the new system for labeling. The model parameters are then updated by fine-tuning. This approach drastically reduces the cost of updating while improving the model’s adaptability. Experiments on the IEEE39-node system verify the effectiveness of the proposed method.

Schlagwörter

Stichwörter der Autor:innen

active learningcontrastive learningtransfer learningtransient stabilization assessment

Autor:innen und Institutionen

  • College of Electrical and Power Engineering, Taiyuan University of Technology, 79 Yingze West Street, Taiyuan 030024, China.
  • Key Laboratory of Control of Power Transmission and Conversion (SJTU), Ministry of Education, 800 Dongchuan Road, Minhang District, Shanghai 200240, China.
  • College of Automotive Engineering, Shanxi Vocational University of Engineering Science and Technology, 369 Wenhua Street, Yuci District, Jinzhong 030619, China.

Förderung

  • Fundamental Research Program of Shanxi Province, grant number
  • Key Laboratory of Control of Power Transmission and Conversion (SJTU), Ministry of Educa-tion
  • Fundamental Research Program of Shanxi Province
  • Key Laboratory of Control of Power Transmission and Conversion (SJTU), Ministry of Education

Zitieren

APA

Zhao, J., Han, X., Wang, C., Yang, J., & Zhang, G. (2024). Contrastive-Active Transfer Learning-Based Real-Time Adaptive Assessment Method for Power System Transient Stability. Sensors, 24(15), 5052. https://doi.org/10.3390/s24155052

Vancouver

Zhao J, Han X, Wang C, Yang J, Zhang G. Contrastive-Active Transfer Learning-Based Real-Time Adaptive Assessment Method for Power System Transient Stability. Sensors. 2024;24(15):5052. doi:10.3390/s24155052

BibTeX
@article{Zhao2024,
  author = {Zhao, Jinman and Han, Xiaoqing and Wang, Chengmin and Yang, Jing and Zhang, Gengwu},
  title = {Contrastive-Active Transfer Learning-Based Real-Time Adaptive Assessment Method for Power System Transient Stability},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {15},
  pages = {5052},
  doi = {10.3390/s24155052}
}
Kostenlos starten →