Industry 4.0-Compliant Wavelet-Based Power Transformer Fault Classification Method During Data Missing Conditions
Document Type
Conference Proceeding
Publication Date
10-4-2024
Department
Department of Electrical and Computer Engineering
Abstract
This paper introduces a novel fault classification method for power transformers using the Real-Time Boundary Stationary Wavelet Transform (RT-BSWT), which is robust to data-missing conditions. Addressing the challenges of fault classification in power transformers, particularly under data-missing conditions, the proposed method offers a threshold-based logic approach, enhancing accuracy and speed in fault classification. The method is robust against gaps in data analysis, making it suitable for real-world applications, and aligning with Industry 4.0 concepts, especially regarding the wide use of data transmission expected in modern industries. A comprehensive evaluation of an extensive database covering various fault types and conditions demonstrated the method's efficacy, with high accuracy rates in complete data and data-missing scenarios.
Publication Title
IEEE Power and Energy Society General Meeting
ISBN
[9798350381832]
Recommended Citation
Dantas, I.,
Medeiros, R.,
Costa, F.,
Sousa, F.,
Wagner, M.,
Klar, M.,
&
Aurich, J.
(2024).
Industry 4.0-Compliant Wavelet-Based Power Transformer Fault Classification Method During Data Missing Conditions.
IEEE Power and Energy Society General Meeting.
http://doi.org/10.1109/PESGM51994.2024.10689202
Retrieved from: https://digitalcommons.mtu.edu/michigantech-p2/1161