Accounting for demand failures in Markovian analyses

Conference · Thu Jan 03 04:00:00 EST 1980

OSTI ID:5425040

Wood, A. P.

Reliability has become a fundamental concern in the development of nuclear power plants. The use of Markov Analysis in reliability evaluations is progressing quickly since it offers methods for dealing with repair considerations, common cause failures, time-dependent failure rates, cyclic failures, and availability calculations. A Markov process is a stochastic or time-dependent process allowing time-dependent failure processes to be modelled easily. In redundant safety systems, however, there are often passive system components whose failure to start on demand is higher than their probability of failure during the time for which the safety system must function. It may be impossible to model this failure process as a system with standby failure rates because the actual failure may be caused by the demand itself. This paper deals with a method for extending Markov Analyses to include demand failures.

• Research Organization: General Electric Co., Sunnyvale, CA (USA). Advanced Reactor Systems Dept.
• DOE Contract Number: EY-76-C-03-0893-032
• OSTI ID: 5425040
• Report Number(s): GEFR-SP-185; CONF-800607-19
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
220200* -- Nuclear Reactor Technology-- Components & Accessories
ENGINEERED SAFETY SYSTEMS
FAILURES
MARKOV PROCESS
NUCLEAR FACILITIES
NUCLEAR POWER PLANTS
POWER PLANTS
RELIABILITY
STOCHASTIC PROCESSES
THERMAL POWER PLANTS