Pacific Northwest National Laboratory (PNNL) scientists are performing research under the Department of Energy Nuclear Energy Research Initiative (NERI) program, to develop a methodology for accurate identification and prediction of equipment faults in critical machinery. The 3-year project, on-line intelligent self-diagnostic monitoring system (SDMS) for next generation nuclear power plants is scheduled for completion at the end of FY 2002. The research involves running machinery to failure in the Laboratory by the introduction of intentional faults. During testing, advanced diagnostic/prognostic sensors and analysis systems monitor the equipment stressor levels, correlate them with expected degradation rates, and predict the resulting machinery performance levels and residual lifetime. Application of a first principles physics-based approach is expected to produce prognostic methodologies of significantly higher accuracies than are currently available. This paper reviews the evolution and current state of the maintenance art. It presents a key measurement philosophy that results from the use of condition based maintenance (CBM) as a fundamental investigative precept, and explains how this approach impacts degradation and failure measurement and prediction accuracy. It then examines how this measurement approach is applied in sensing and correlating pump stressors with regard to degradation rate and time to equipment failure. The specifics are examined on how this approach is being applied at PNNL to cavitation and vibration phenomena in a centrifugal pump. Preliminary vibration analysis results show an excellent correspondence between the (laser) motor position indication, the vibration response, and the dynamic force loading on the bearings. Orbital harmonic vibratory motion of the pump and motor appear to be readily correlated through the FFTs of all three sensing systems.

L'invention concerne des procedes et des systemes d'identification, de comprehension, et de prediction de la degradation et de la defaillance de systemes mecaniques. Ces procedes consistent a mesurer et a quantifier les agresseurs responsables de l'activation des mecanismes de degradation dans le composant de machine d'interet. L'intensite de l'agresseur peut etre en correlation avec la vitesse de degradation physique selon certaines fonctions determinables de sorte qu'il existe une relation derivee entre la performance de la machine, la degradation de celle-ci et l'agresseur sous-jacent. La relation derivee peut etre utilisee pour effectuer des calculs de diagnostic et de pronostic a propos de la performance et de la duree de vie envisagee de la machine. Ces calculs peuvent etre executes en temps reel de facon a permettre a l'operateur de la machine de regler rapidement les parametres de fonctionnement de la machine afin de minimiser, voire d'eliminer, les effets du mecanisme de degradation, et de prolonger ainsi la duree de vie de la machine. Plusieurs systemes utilisant ces procedes sont decrits.

Prognostics is the process of predicting the future state of a system. Prognostics systems comprise sensors, a data acquisition system, and microprocessor-based software to perform sensor fusion, analysis, and reporting/interpreting of results with little or no human intervention in real-time or near real-time. It offers the promise of minimizing failures (especially failures “in the field”), extending the time between maintenance overhauls, and reducing life-cycle costs. But prognostics is still in a research and development phase, and implementing prognostics is a monumental task on several levels—the technical challenges involving hardware and sensor technologies, the analytical challenges involving predictive methods, and the logistical challenges centering on how to make use of prognostic information.

The remains of the nuclear fuel that was severely damaged in the 1986 Chernobyl unit 4 accident lie in large masses in the premises under the reactor. The fuel debris exists in the form of dusts, chunks, and lavas, and the quantities are substantial--some rooms contain several tons of fuel. Since there is a possibility of water entering these rooms, there is an obvious concern over criticality safety. Incidents of increased neutron count rates have been noted in the vicinity of nuclear fuel debris. Pacific Northwest National Laboratory (PNNL), under a program funded by the US Department of Energy, responded to this safety concern by assembling a new monitoring system to characterize the radiation environment in the vicinity of major fuel deposits. The new monitoring system will measure the gamma and neutron radiation fields in several locations. The measurement data can be tracked over time to determine the characteristics of the radiation fields and better understand the nuclear safety conditions in the vicinity of the fuel. The monitoring system was designed to provide information that will allow a better interpretation of any future events.