Paper
6 February 2024 Recognition of abnormal events in nuclear power plants based on power disturbance data
Biao Li, Haiqiang Liu
Author Affiliations +
Proceedings Volume 12979, Ninth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023); 129794K (2024) https://doi.org/10.1117/12.3015552
Event: 9th International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023), 2023, Guilin, China
Abstract
With the profound changes in the power supply type and characteristics, network topology, and load characteristics of the power system, abnormal events in nuclear power plants become more frequent. However, the current automatic diagnosis device of the nuclear power plant cannot well detect and distinguish some normal and abnormal events. Therefore, this paper proposes a nuclear power plant abnormal event recognition method based on power disturbance data feature extraction. Firstly, the relevant model is established, and the detection of abnormal events that are not easily detected by conventional protection devices is realized by calculating the continuous wavelet transform coefficient of arc current data. Then, four feature quantities are extracted. The logical judgment rules for the classification and recognition of abnormal events and the database of abnormal event characteristics are established. A matching algorithm combined with topic search is proposed to achieve classification. MATLAB/Simulink is used to identify abnormal events in nuclear power plants. The feasibility of abnormal event recognition is verified.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Biao Li and Haiqiang Liu "Recognition of abnormal events in nuclear power plants based on power disturbance data", Proc. SPIE 12979, Ninth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023), 129794K (6 February 2024); https://doi.org/10.1117/12.3015552
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Continuous wavelet transforms

Nuclear power plants

Databases

Feature extraction

Wavelets

Windows

Capacitors

Back to Top