Análise de falha em rolamentos de motores de indução trifásico através do som utilizando densidade de máximos

Abstract

Induction motors are the most utilized electric machines in commercial and industrial applications and 90% of them are three-phase induction motors (TIM). Faults in three-phase induction motors (TIM) can lead to the shutdown of important industry sectors, causing financial and operational safety losses. The most common faults in these motors are the bearing ones, and most of the fault analysis techniques are based on fast Fourier transform, Wavelet and training algorithms, whose classifiers demand high computational effort. A large part of the studies developed in the analysis of failures in bearings through sound uses the combination of multisensory for comparison and synchronization between acquired signals. This paper presents an approach based on quantification of the chaotic behavior for the characterization of rigid ball bearing failure of a three-phase induction motor through the method called Signal Analysis based on Chaos using Density of Maxima (SAC-DM) using the sound signal emitted by the TIM. This technique is based on an algorithm that counts peaks of the motor acoustic signal in the time domain to detect faults using only a sensor and an algorithm with a low computational cost. For validation of the technique, healthy ball bearing experiments were performed, with internal and external race failure under three load conditions (0%, 50%, and 100%). The results demonstrated that the SAC-DM was able to detect the presence of bearing failures even with the TIM under variable load.