Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Incorporating ''fuzzy'' data and logical relations in the design of expert systems for nuclear reactors

Conference ·
OSTI ID:5983568
This paper applies the method of assigning probability in Dempster-Shafer Theory (DST) to the components of rule-based expert systems used in the control of nuclear reactors. Probabilities are assigned to premises, consequences, and rules themselves. This paper considers how uncertainty can propagate through a system of Boolean equations, such as fault trees or expert systems. The probability masses assigned to primary initiating events in the expert system can be derived from observing a nuclear reactor in operation or based on engineering knowledge of the reactor parts. Use of DST mass assignments offers greater flexibility to the construction of expert systems.
Research Organization:
Oak Ridge National Lab., TN (USA); Oak Ridge Associated Universities, Inc., TN (USA)
DOE Contract Number:
AC05-84OR21400
OSTI ID:
5983568
Report Number(s):
CONF-870832-3; ON: DE87014410
Country of Publication:
United States
Language:
English