Life state recognition of slewing bearing with heavy load and low speed based on point density function with fuzzy C means

Abstract

The slewing bearing with heavy-load and low-speed, as the key part of large rotating machine, has been widely applied in engineering. This paper presents an improved method to recognize the life state of slewing bearing, which guarantees the machine work efficiently and stably. Firstly, point density function with fuzzy C means (D-FCM) was proposed as an unsupervised method to recognize the life state of the slewing bearing. Then, an experiment on the full life test of slewing bearing was conducted based on a home-made test platform to demonstrate the effectiveness of the proposed method. Finally, the principal component analysis (PCA) was used as a supervised method for comparison. The results indicate D-FCM in unsupervised classification can recognize normal state, degeneration state and failure state of slewing bearing, which is more accurate and clear than traditional FCM and PCA, and lays the foundation of real-time maintenance.