Manutenção preditiva de motores trifásicos de indução através da análise de assinatura de corrente e fluxo magnético.

Abstract

This paper presents the study of a three-phase induction motor predictive maintenance method based on the frequency spectrum analysis of current and magnetic flux of motors from a petrochemical industry company. The main proposal was to study the methods that the company uses in relation to the mentioned techniques, compare them with the methods and procedures studied and proposed by the main researchers from the subject, and then, sugest improvements in the three-phase induction motors predictive maintenance system of the company. The objective of the application of the predictive maintenance by current and magnetic flux analysis is to detect faults and defects in the motor that may cause or has already caused damage to the motor. Such faults can be: windings magnetic unbalancing, broken rotor bars, airgap and rotor eccentricities, and other electromechanical faults. The studied technique is based on the acquisition and comparison of the motors current and magnetic flux signatures, in which are used sensors conected to an signal processing and gathering equipment, which makes the signals reading in the time domain and makes its processing to the frequency domain, through the Fast Fourier Transformer, in order to detect the characteristic electromagnetic frequencies of the motor. The graphic location, the amplitude and the amplitude difference (in dB) of these frequencies in relation to the motor fundamental frequency (60 Hz) are what defines the presence or not of faults in the motor. At the end of this paper, improvements were suggested and it was elaborated an implementation scheme of the improvements, based in the comparison of the studied techniques with its practical applicability. Thus, it was possible to notice the evolution of the predictive analysis techniques of electrical motors.