Open AccessIdentification of current transformer saturation based on the improved gradient mathematical morphology method
Author(s) -
Duan Jiandong,
Lei Yang,
Li Hao
Publication year - 2017
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2017.0489
Subject(s) - current transformer , mathematical morphology , computer science , transformer , saturation (graph theory) , morphological gradient , relay , sliding window protocol , electronic engineering , control theory (sociology) , voltage , engineering , artificial intelligence , edge detection , mathematics , image processing , physics , electrical engineering , window (computing) , image (mathematics) , power (physics) , control (management) , combinatorics , quantum mechanics , operating system
A protective current transformer (CT) is essential to accommodate protection relays while the saturation of the CT may severely impact the security of protective relays. The saturation detection is the first step for activating procedures to ensure the satisfactory performance of the relay system. This study presents a saturation detection method based on mathematical morphology. The detection method involves the secondary current data as the input in a sliding data window and a noisy filter as the must step for morphological wavelet detection. Then the proposed extended sine form of the structural element is used for shaping the saturation detecting signals. To verify the practicality and effectiveness of the method, dynamic large CT tests, up to 48 kA, are carried out, particularly. The test aims at the P class CT installed in the 330/110 kV substation and the simulating current ranging from 6 to 48 kA. More discussions are simulated in the 330 kV substation system based on RT‐LAB to inspect the noise‐repressing and saturation detect abilities. The experimental data and simulation results show that the proposed method could accurately detect the saturation of the CT and adapt to the noise circumstance of the test.