Bearing Fault Recognition Based on Feature Extraction and Clustering Analysis

In this paper, the clustering analysis is used to distinguish bearing fault pattern. Some time domain feature parameters are extracted from vibration signal, and the combination of three feature parameters are chosen from these feature parameters for the clustering analysis. The Euclidean distance is used to calculate the distance of point-to-center. After validation, the effect of clustering analysis is effective to distinguish the bearing fault pattern, and the best combination of feature parameters for fault pattern recognition by clustering analysis is found.