Abstract

PRIETO, Rommel S.; BRAVO, Diego A.  and  RENGIFO, Carlos F.. Fault detection in brushless motors through audio processing and machine learning. Ciencia en Desarrollo [online]. 2025, vol.16, n.1, pp.52-61.  Epub Jan 20, 2025. ISSN 0121-7488.  https://doi.org/10.19053/uptc.01217488.v16.n1.2025.16963.

Electromechanical devices tend to wear off with use; Early fault detection is an important tool to reduce operating costs and improve the life of an industrial device. This work deals with fault detection of brushless DC motors, using audio signal processing and extracting statistical and spectral features to train classical Machine Learning models as k-Nearest Neighbors, Decision Trees and Máquinas de soporte vectorial (SVM). The trained models are then deployed to an IoT application built using Django. The implemented methodology shows a success rate of up to 92 % accuracy for fault detection in brushless motors using audio processing and machine learning.

Keywords : Fault Detection; Machine Learning; Brushless Motors; Audio Processing.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )