“Introduction:

Prognosis is the ability to predict or forecast the future condition of a component and/or system of components, in terms of failure or degraded condition, so that it can satisfactorily perform its operational requirement. In this paper, we will specifically focus on prognosis technologies that can enable the early detection and prediction of critical rotating machine components such as bearings and gears.

A specific aircraft engine bearing prognostic application is described that utilizes available measurement information, including rotor speed, vibration, tube system information and aircraft maneuvers to calculate remaining useful life for the engine bearings. Linking this sensed data with fatigue-based damage accumulation models associated with bearing remaining useful life is used to provide the predictive assessment. The combination of health monitoring data and model-based techniques provides a unique and knowledge-rich capability that can be utilized throughout the components entire life, using model-based estimates when no diagnostic indicators are present and using the measured features at later stages when failure indications are detectable, thus reducing the uncertainty in model-based predictions.

Due to the inherent uncertainties in such prognosis systems, achieving the best possible prediction on a machine component’s health is often implemented using various data fusion techniques that can optimally combine sensor data, empirical/physics-based models and historical information. Implementation of component-level, machinery prognostics will be illustrated on the engine bearing application.

Integrated Prognosis Framework

The approach described next provides an integrated prognosis framework as applied to rotating machinery components and it is generic in nature. In this case, it will be applied to rolling element bearings to fully describe the concept. The architecture builds upon existing or enabling technologies such as advanced oil debris/condition monitoring, high/low frequency vibration analysis, thermal trend analysis and empirical/physics-based modeling to practically achieve its objectives. An aspect of this approach is the development and implementation of an integrated prognostic architecture that is flexible enough to accept input from many different sources of diagnostic/prognostic information in order to contribute to better fault isolation and prediction on bearing remaining useful life.

The block diagram shown in Figure 1 illustrates this generic representation of the integrated prognosis system architecture for machinery component. Within this architecture, measured parameters from the health monitoring system such as oil condition/debris monitor outputs and vibration signatures can be accommodated within the anomaly detection/diagnostic fusion center. Based on these outputs, specific triggering points within the prognostic module can be processed so that effective transition associated with various failure mode models (i.e. spall initiation model to a spall progression model) can be accomplished.”