Premium
An application of genetic algorithms to surveillance test optimization of a PWR auxiliary feedwater system
Author(s) -
Lapa Celso M. F.,
Pereira Cláudio M. N. A.,
Frutuoso e Melo P. F.
Publication year - 2002
Publication title -
international journal of intelligent systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.291
H-Index - 87
eISSN - 1098-111X
pISSN - 0884-8173
DOI - 10.1002/int.10051
Subject(s) - unavailability , reliability engineering , boiler feedwater , probabilistic logic , computer science , nuclear power plant , component (thermodynamics) , common cause failure , genetic algorithm , algorithm , engineering , operations management , boiler (water heating) , machine learning , artificial intelligence , common cause and special cause , physics , nuclear physics , thermodynamics , waste management
Nuclear power plant systems are comprised of both on‐line and standby components. Standby componentsdiffer from on‐line ones, as they might be unavailable due to unrevealed failures. The usual procedureemployed to reveal failures before real demands is to submit the component to surveillance tests. Surveillance testpolicies might deal with two conflicting scenarios: the test frequency must be sufficiently high in order to revealfailures before demands, but, on the other hand, it must be low enough due to its influence on the componentunavailability. Standard surveillance test policies for typical nuclear power plants usually consist of periodic tests for whichthe frequencies are often higher than necessary for obtaining the optimal availability. In this work, a newsurveillance test policy optimization method, based on genetic algorithms, is applied to the Angra‐I(Brazilian PWR) auxiliary feedwater system. The new probabilistic model has been developed in order tocomprise the following features: (1) aging effects on standby components when they undergo surveillancetests; (2) revealing failures during the surveillance tests implies corrective maintenance, and,consequently, increasing outage times; (3) components are distinct (i.e., each has distinct testparameters, such as outage time, aging factors, etc); (4) tests are not necessarily periodic.The results, when compared to those obtained by standard test policies, show improved overall availability at thesystem level. © 2002 Wiley Periodicals, Inc.