ABSTRACT
Fault diagnosis is as old as the use of machines. With simple machines, manufacturers and users relied on simple protections to ensure safe and reliable operation, but with the increasing complexity of tasks and machines, improvements were required in fault diagnosis. It is now critical to diagnose faults at their inception, as unscheduled machine downtime can upset deadlines and cause heavy financial losses. Diagnostic methods to identify faults involve many fields of science and technology and include the following: (a) electromagnetic field monitoring, search coils, coils wound around motor shafts (axial flux-related detection), (b) temperature measurements, (c) infrared recognition, (d) radio frequency (RF) emissions monitoring, (e) noise and vibration monitoring, (f) chemical analysis, (g) acoustic noise measurements, (h) motor current signature analysis, (i) modeling, artificial intelligence (AI), and neural network-based techniques (Li, 1994).
