The health of a mechanical component deteriorates over time and its service life is randomly distributed and can be modeled by a stochastic deterioration process. For most of the mechanical components, the deterioration process follows a certain physical laws and their mean life to failure can be determined approximately by these laws. However, it is not easy to apply these laws for mechanical component prognostics in current health monitoring applications. In this paper, a stochastic modeling methodology for mechanical component prognostics with condition indicators used in current health monitoring applications is presented. The effectiveness of the methodology is demonstrated with a real shaft fatigue life prediction case study.

This content is only available via PDF.
You do not currently have access to this content.