Systems analysis approach to probabilistic modeling of fault trees

Conference · Tue Jan 01 04:00:00 EST 1985

OSTI ID:5697527

Bartholomew, R J; Qualls, C R

A method of probabilistic modeling of fault tree logic combined with stochastic process theory (Markov modeling) has been developed. Systems are then quantitatively analyzed probabilistically in terms of their failure mechanisms including common cause/common mode effects and time dependent failure and/or repair rate effects that include synergistic and propagational mechanisms. The modeling procedure results in a state vector set of first order, linear, inhomogeneous, differential equations describing the time dependent probabilities of failure described by the fault tree. The solutions of this Failure Mode State Variable (FMSV) model are cumulative probability distribution functions of the system. A method of appropriate synthesis of subsystems to form larger systems is developed and applied to practical nuclear power safety systems.

• Research Organization: Los Alamos National Lab., NM (USA); New Mexico Univ., Albuquerque (USA). Dept. of Mathematics and Statistics
• DOE Contract Number: W-7405-ENG-36
• OSTI ID: 5697527
• Report Number(s): LA-UR-85-494; CONF-850809-15; ON: DE85007705
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
220900* -- Nuclear Reactor Technology-- Reactor Safety
DIFFERENTIAL EQUATIONS
ENGINEERED SAFETY SYSTEMS
EQUATIONS
FAILURE MODE ANALYSIS
FAULT TREE ANALYSIS
MARKOV PROCESS
PROBABILITY
STOCHASTIC PROCESSES
SYSTEM FAILURE ANALYSIS
SYSTEMS ANALYSIS