Implementación de un sistema automatizado de análisis vibracional para el mantenimiento predictivo en motores eléctricos.

Abstract

The objective of this work was to implement an automated system of vibrational analysis as a diagnostic tool for predictive maintenance in electric motors through signal processing using a test bench. This system allows to diagnose the main failures caused by vibrations in electric motors. The vibrations are acquired by using a triaxial accelerometer installed in the motor and a data acquisition card (NI-DAQ USB 6009). The acquired data is sent to a computer, and through the LabVIEW software the signal processing is performed in order to obtain the waveform over time, conditioning and transforming the signal into a frequency spectrum using the Fast Fourier Transform (FFT). To this is added a Global System for Mobile Telecommunications (GSM) using a programmable logic controller (PLC), allowing the user to be informed about the state of motor and its vibration levels without the need of continuously verifying the levels on the Human-Machine Interface (HMI). The obtained results were compared through a visual and statistical analysis with the VIBRAcheck 200 equipment. From the vibration global values RMS errors less than 2% were obtained, while with respect to the values of the frequency components a similar average value was obtained between the two equipment’s. This analysis was used to carry out a predictive maintenance plan based, which consists of the spectra diagnosis according to what is established in the Charlotte charts and based on the vibration levels, the severity is established based on the ISO 2372 standard. The use of the implemented system as an alternative to the diagnostic processes and development of electrical motor maintenance plan is recommended.