Diagnostics and prognostics of a single surface defect on roller bearings

Bearing condition monitoring has been the focus of a wide range of studies over the past years. Current monitoring techniques that focus on the identification of faults present in a bearing have various limitations. Typically they are applicable only under well-defined, specific and precalibrated operating conditions, thereby preventing continuous monitoring of a system operating in a variant environment. They are often limited in damage severity estimation and prognostic capability. This, in turn, prevents the development of optimal maintenance scheduling in favour of overall system safety and productivity. Research presented in this paper has yielded results that have extended bearing diagnostics and prognostics to address these limitations and to achieve optimal machinery maintenance scheduling. This paper discusses the current research status on the development of a new signal processing method with noise cancellation capability to provide early defect detection, the establishment of a diagnostic model to estimate bearing defect severity under variant conditions and the formulation of an adaptively tuned defect propagation model to track the time-variant nature of defect growth for the forecasting of bearing remaining utility.