2009/08/21 by Jie Liu, W. Wang, Farid Golnaraghi · 1 citation
Engineering · Mathematics · #Artificial intelligence #Bearing (navigation) #Classifier (UML) #Computer science #Condition monitoring #Data mining #Digital signal processing #Electrical engineering #Electronic engineering #Engineering #Fault Detection and Control Systems #Gear and Bearing Dynamics Analysis #Machine Fault Diagnosis Techniques #Mathematics #Reliability (semiconductor) #Reliability engineering #Rolling-element bearing #Scheme (mathematics) #Signal processing
paper · doi:10.1109/tim.2009.2023814
openalex publication_date 2009/08/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
Rolling-element bearings are widely used in various mechanical and electrical facilities; accordingly, a reliable real-time bearing condition-monitoring system is very useful in industries to detect bearing defects at both incipient and advanced levels to prevent machinery performance degradation and malfunctions. The objective of this paper is to develop an enhanced diagnostic (ED) scheme for bearing fault diagnostics. This scheme consists of modules of classification and prediction. A neurofuzzy (NF) classifier is proposed to effectively integrate the strengths of several signal-processing techniques (or resulting representative features) for a more positive assessment of bearing health conditions. A multistep NF predictor is employed to forecast the future states of the bearing health condition to further enhance the diagnostic reliability. The effectiveness of this ED scheme is verified by experimental tests that correspond to different bearing conditions.