Classification of Low Frequency Signals Emitted by Power Transformers Using Sensors and Machine Learning Methods

Daniel Jancarczyk; Marcin Bernaś; Tomasz Boczar

doi:10.3390/s19224909

Classification of Low Frequency Signals Emitted by Power Transformers Using Sensors and Machine Learning Methods

Sensors (Basel). 2019 Nov 10;19(22):4909. doi: 10.3390/s19224909.

Authors

Daniel Jancarczyk¹, Marcin Bernaś¹, Tomasz Boczar²

Affiliations

¹ Department of Computer Science and Automatics, University of Bielsko-Biala, 43-309 Bielsko-Biala, Poland.
² Institute of Electric Power Engineering and Renewable Energy, Opole University of Technology, 45-758 Opole, Poland.

Abstract

This paper proposes a method of automatically detecting and classifying low frequency noise generated by power transformers using sensors and dedicated machine learning algorithms. The method applies the frequency spectra of sound pressure levels generated during operation by transformers in a real environment. The spectra frequency interval and its resolution are automatically optimized for the selected machine learning algorithm. Various machine learning algorithms, optimization techniques, and transformer types were researched: two indoor type transformers from Schneider Electric and two overhead type transformers manufactured by ABB. As a result, a method was proposed that provides a way in which inspections of working transformers (from background) and their type can be performed with an accuracy of over 97%, based on the generated low-frequency noise. The application of the proposed preprocessing stage increased the accuracy of this method by 10%. Additionally, machine learning algorithms were selected which offer robust solutions (with the highest accuracy) for noise classification.

Keywords: classification; low-frequency noise; low-frequency sensor; machine learning; power transformer.