Eventos Anais de eventos
COBEM 2017
24th ABCM International Congress of Mechanical Engineering
Hybrid System Using Acoustic Emission and Electric Current for Fuzzy Diagnosis of Surface Wear in Gearbox Coupled to Induction Motor
Submission Author:
Alexander Sena , PB
Co-Authors:
Leonardo José Cavalcante Vasconcelos, ARLESON KENNEDI FRANÇA DOS SANTOS, Dalton Valadares, Alexander Sena
Presenter: Leonardo José Cavalcante Vasconcelos
doi://10.26678/ABCM.COBEM2017.COB17-2812
Abstract
This work proposes a hybrid system that uses data collected by current sensors and acoustic emission for the diagnosis of incipient failures in gearboxes coupled to Three-phase Induction Motors, searching for an alternative and non-invasive analysis by signs of vibration. The decomposition of the signals from the Wavelet Packet Transform (WPT) made it possible to evaluate the frequency bands characteristic of the failure by the spectral analysis and the calculation of the entropy. The spectral patterns were initially used to verify the effects caused by surface wear on the signals. In the sequence, the signals were processed by the wavelet transform, to be used as input of a Fuzzy inference system. Experimental results were obtained by testing the system at different wear levels due to lack or insufficient lubrication. The electromechanical set was driven directly from the electrical network and by frequency inverter at 20Hz, 30Hz, 40Hz and 50Hz in order to evaluate the sensitivity of the strategy in a wide rotation range. For each drive at 60Hz, the conditions of nominal load, at no load, with 40% and 70% of the load were imposed in order to evaluate the effects of the load in the calculation of the wavelet entropy.
Keywords
failure analysis, surface wear, fuzzy inference, gearbox, Wavelet
