<abstract xmlns="http://www.w3.org/1999/xhtml"><p> This paper proposes an approach based on acoustic signals for detecting faults appearing in synchronous motors. Acoustic signals of a machine were used for fault detection. These faults contained: broken coils and shorted stator coils. Acoustic signals were used to assess the usefulness of early fault diagnostic of synchronous motors. The acoustic signal recognition system was based on methods of data processing: normalization of the amplitude, Fast Fourier Transform (FFT), method of frequency selection (MoFS), backpropagation neural network, classifier based on words coding, and Nearest Neighbor classifier. A plan of study of acoustic signals of synchronous motors was proposed. Software of acoustic signal recognition of synchronous motors was implemented. Four states of a synchronous motor were used in analysis. A pattern creation process was carried out for 28 training samples of noise. An identification process was carried out for 60 test samples. This system can be used to diagnose synchronous motors and other electrical machines.</p></abstract>

This paper proposes an approach based on acoustic signals for detecting faults appearing in synchronous motors. Acoustic signals of a machine were used for fault detection. These faults contained: broken coils and shorted stator coils. Acoustic signals were used to assess the usefulness of early fault diagnostic of synchronous motors. The acoustic signal recognition system was based on methods of data processing: normalization of the amplitude, Fast Fourier Transform (FFT), method of frequency selection (MoFS), backpropagation neural network, classifier based on words coding, and Nearest Neighbor classifier. A plan of study of acoustic signals of synchronous motors was proposed. Software of acoustic signal recognition of synchronous motors was implemented. Four states of a synchronous motor were used in analysis. A pattern creation process was carried out for 28 training samples of noise. An identification process was carried out for 60 test samples. This system can be used to diagnose synchronous motors and other electrical machines.

Recognition of Acoustic Signals of Synchronous Motors with the Use of MoFS and Selected Classifiers

AGH University of Science and Technology, Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering, Department of Automatics and Biomedical Engineering, Al. A. Mickiewicza 30, 30-059 Kraków, Poland

Measurement Science Review

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

This paper proposes an approach based on acoustic signals for detecting faults appearing in synchronous motors. Acoustic signals of a machine were used for...

Recognition of Acoustic Signals of Synchronous Motors with the Use of MoFS and Selected Classifiers

Data publikacji: 27 sie 2015

Zakres stron: 167 - 175

Otrzymano: 20 cze 2014

Przyjęty: 24 lip 2015

DOI: https://doi.org/10.1515/msr-2015-0024

© by Adam Glowacz

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.