Fault diagnosis of motor bearing using self-organizing maps

This paper focuses on the application of self-organizing maps (SOM) in motor bearing fault diagnosis and presents an approach for motor rolling bearing fault diagnosis using SOM neural networks and time/frequency-domain bearing analysis. The SOM is a neural network algorithm which is based on unsupervised learning and combines the tasks of vector quantization and data projection. The objective of this paper is to detect and diagnose faults to motor adoptively, with emphasis on faults occurred in the bearing part of the motor. The experiment results show that the SOM is an efficient tool for the fault visualization and diagnosis of motor bearing