Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Comparison of Traditional Vibration Analysis Techniques and Machine Learning Models for Bearing Fault Detection
Submission Author:
Victor Bauler , SC
Co-Authors:
Victor Bauler, Julio Cordioli, Danilo Braga, Danilo Silva
Presenter: Victor Bauler
doi://10.26678/ABCM.COBEM2023.COB2023-1635
Abstract
Vibration analysis is a widely used technique for fault detection in rotating machinery. In recent years, machine learning techniques have been increasingly applied to this field to improve the accuracy and efficiency of fault detection. This paper compares the effectiveness of traditional vibration analysis techniques, such as bandpass filtering and Hilbert transform envelope analysis, with machine learning models, namely Support Vector Machine, Logistic Regression, and Random Forest, for bearing fault detection. The study utilizes an experimental dataset collected in the Laboratory of Vibration and Acoustics (LVA) at the Federal University of Santa Catarina (UFSC). The models are trained and tested on this dataset, and the effectiveness of each technique is evaluated based on their ability to detect bearing faults. The results of this study will contribute to the literature by providing a more objective and comprehensive evaluation of the effectiveness of different vibration analysis techniques. Furthermore, the study also explores the potential of machine learning models in improving the accuracy and efficiency of fault detection. This study provides insights into the effectiveness and limitations of each approach by comparing the performance of traditional vibration analysis techniques and machine learning models.
Keywords
Rolling element bearings, Envelope analysis, Vibration Analysis, Fault Detection, Machine learning models
