Analysis of the qualified life extension of nuclear safety systems through modulated Poisson point processes

The Modulated Power Law Process (MPLP) is a three-parameter stochastic point process that can be used to describe the failure times of repairable systems. While the Non-Homogeneous Poisson Process (NHPP) or the Renewal Process (RP) implies that a system is exactly on the same condition just after a repair as just before a failure, the MPLP allows for the system to be affected both by failure and repair. For this reason this model is adequate for evaluating the recurrent events that incorporate both time trends and effects of past events such as the renewal type behavior. The purpose of this paper is to evaluate the MPLP as a model for the rate of occurrence of failures (ROCOF) of repairable systems to decide for an extension of qualified life in the context of a license renewal of a nuclear power plant. The analysis was carried out considering some field data spanning a 2300 calendar day period, which is approximately equivalent to four burn-up cycles and refueling periods. The reliability is estimated, and maintenance strategies are discussed, concerning the results of the case study presented. As a conclusion, the MPLP is adequate for modeling the rate of occurrence of failures that are time dependent, and can be used where aging mechanisms are present in the operation of repairable systems. This means that equipment characteristics that are important may be inserted into the model and the results can help make decisions in the context of maintenance programs, as is the case with the maintenance rule concept that has been proposed by the U.S. Nuclear Regulatory Commission and has just been implemented in nuclear power plants in Brazil, for instance. The difficulty with data acquisition for applying point processes is discussed throughout.