Abstract

this paper addresses the comparison of features, extracted in the time domain, from vibration, acoustic emission, and current signals, for the identification of eight levels of severity of pitting in a gearbox. The vibration, acoustic emission, and current signals were first acquired using a gearbox lab experimental test bed. Then, twenty features were extracted in the time domain from each signal; these features are ranked by Chi squared and entered into the KNN classifier, which allows the evaluation of the classification accuracy for each acquired signal and performing an analysis of the features. The results indicate that the vibration and AE signals identified the pitting level better than the current signal. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.