Age-Dependent Risk Quantification Using Standard Maintenance Records
A methodology for the analysis of standard maintenance records in order to detect and quantify changing failure rates is presented. The methodology is applied to the records for components from one system of a nuclear power plant and the results of the analysis are then used to quantify core damage frequencies and uncertainties. The first step of the methodology is to determine the times of failure occurrences from plant maintenance records. Next, inferences about the time dependent nature of the data are made following the methods of Cox and Lewis.3 The specific aging model assumes a hazard function of the form λ(t) = exp(a + βt) which we refer to as the exponential aging model. The data is first tested to see if a null hypothesis of no increasing failure rate should be rejected. If rejected, maximum likelihood approaches are employed for joint inference about α and β. Diagnostic tests are developed and used to confirm that assumptions made regarding component data pooling appear to be correct. Constant equipment failure rate is employed when the null is not rejected. Using the parameters and joint confidence region, the time-dependent hazard function is evaluated at various times. The time-dependent results for various components are used as basic event inputs to a probabilistic risk assessment model to determine the increase in core damage frequency as a function of time.
KeywordsAge-dependent PRA assumptions model methodology
Unable to display preview. Download preview PDF.
- 1.Ghahramani, PRA Data Base Development: An Integrated Approach, Proceedings of the International Topical Meeting on Probability, Reliability and Safety Assessment, p. 882, Vol. 2, Pittsburgh, Pennsylvania, April 2–7, 1989.Google Scholar
- 2.R. C. Bertucio et al., Analysis of Core Damage Frequency from Internal Events: Surry Unit 1, Draft, NUREG/CR-4550, Rev. 1, Vol. 3, Part 1 (1988).Google Scholar
- 3.D. R. Cox and P. A. W. Lewis, The Statistical Analysis of Series of Events, Halsted Press, United States and Chapman and Hall, London (1966).Google Scholar
- 4.A. J. Wolford et al., A Survey of the Nuclear Plant Reliability Data System and Licensee Event Reports to Determine Linear Failure Rates, EGG-SSRE-8603 (1989).Google Scholar
- 5.A. J. Wolford et al., Aging Risk Assessment Methodology: Development and Demonstration Study, NUREG/CR-5378, draft (1990).Google Scholar
- 6.C. L. Atwood, Estimating Hazard Functions for Repairable Components (1990).Google Scholar