Comparação entre sinais de vibração e de áudio aplicados à detecção de falhas em motores de indução trifásicos

Abstract

This work seeks to make a comparison between the audio and vibration signals applied to fault detection in three-phase induction motors. Concepts are presented to understand how the used equipment works, as well the methods to be used to the signal processing. The audio signals were collected by a microphone and the vibration signals by an accelerometer. In the work, it is proposed to find the frequency spectrum of the signals and visually compare them, as well to use self-organizing maps, an artificial neural network architecture, to group the data, and make a comparison between the two resulting clusters. Finally, the results and a discussion about them are presented.