Prognostics and condition-based maintenance: A new approach to precursive metrics

The assumptions used in the design basis of process equipment have always been as much art as science. The usually imprecise boundaries of the equipments' operational envelope provide opportunities for two major improvements in the operations and maintenance (O&M) of process machinery: (a) the actual versus intended machine environment can be understood and brought into much better alignment and (b) the end goal can define O&M strategies in terms of life cycle and economic management of plant assets. Scientists at the Pacific Northwest National Laboratory (PNNL) have performed experiments aimed at understanding and controlling aging of both safety-specific nuclear plant components and the infrastructure that supports essential plant processes. In this paper we examine the development of aging precursor metrics and their correlation with degradation rate and projected machinery failure. Degradation-specific correlations have been developed at PNNL that will allow accurate physics-based diagnostic and prognostic determinations to be derived from a new view of condition-based maintenance. This view, founded in root cause analysis, is focused on quantifying the primary stressor(s) responsible for degradation in the component of interest and formulating a deterministic relationship between the stressor intensity and the resulting degradation rate. This precursive relationship between the performance, degradation, and underlying stressor set is used to gain a first-principles approach to prognostic determinations. A holistic infrastructure approach, as applied through a conditions-based maintenance framework, will allow intelligent, automated diagnostic and prognostic programming to provide O&M practitioners with an understanding of the condition of their machinery today and an assurance of its operational state tomorrow.