Condition-based maintenance of rolling element bearings

Since bearing is one of the most important and commonly used parts in almost all of rotating machinery, the unexpected breakdown of rolling element bearing presents a major concern to the utilization of rotating machinery. In many situations, such as in helicopters, transportation vehicles, automatic precision processing machines, etc., bearing failure can result in catastrophic consequences. Therefore, knowing bearings' real running conditions under the above circumstances is very important and critical to the maintenance schedule of the rotating machinery. Based on previous research, it is known that the major challenges of this optimization hinge on the capability to detect initial bearing defect, reliably assess inline bearing defect severity, and prognosticate in real-time its remaining life under practical operating conditions. In order to achieve these objectives, this paper addresses the following aspects: 1) Selection of monitoring sensors and an effective data acquisition system 2) Methods of bearing initial defect detection 3) Evaluation of diagnostic models of inline bearing defect severity 4) Prognostic models of real-time bearing lifetime prediction under practical operating conditions. In the end, the strategy of how to optimize bearing maintenance schedules with the application of condition-based monitoring techniques - which can save the cost and productivity or even prevent catastrophic consequences - is proposed.