AN EXPERT SYSTEM-BASED DECISION AID FOR REACTOR TRIP REDUCTION AND POST-TRIP ANALYSIS

THIS PROJECT WILL FOCUS ON IMPROVING NUCLEAR POWER PLANT AVAILABILITY BY EMPLOYING AN EXPERT SYSTEM-BASED OPERATOR DECISION AID TO PROVIDE GUIDANCE TO THE OPERATION AND MAINTENANCE OF THE PLANT SO THAT BOTH THE INDIVIDUAL COMPONENTS AND THE PLANT REMAIN IN THE REGIME OF SAFE OPERATION TO MINIMIZE INADVERTENT TRIPS. THE PRIMARY FOCUS IS TO DEVELOP A SYSTEM THAT IS BOTH LIVING AND LEARNING; THAT IS IT MUST BE EASILY UPGRADABLE AND MUST ALLOW FOR A TESTING PROCESS THAT WILL ENABLE LEARNING. THE SYSTEM WILL HAVE BOTH THE CAPACITY TO PROVIDE TRIP AVOIDANCE INFORMATIONAND POSTMORTEM EVALUATION USING AN IDENTICAL KNOWLEDGE BASE,THEREBY PROVIDING A STRONG AND ACTIVE FEEDBACK MECHANISM TO ENABLE THE GROWTH OF THE SYSTEM. THE LOGIC FLOWGRAPH METHOD, CREATED IN AN OBJECT-ORIENTED ENVIRONMENT, WILL PERFORM THE KNOWLEDGE REPRESENTATION FUNCTION OF THE PHYSICAL SYSTEM. THE LOGIC EMPLOYED IN THE REASONING PROCESS WILL BE FUNCTION DEPENDENT, EITHER TRIP AVOIDANCE, SCENARIO EVALUATION, OR TRIP DIAGNOSTIC, AND WILL UTILIZE SOME COMBINATION OF DEDUTIVE, INDUCTIVE, ABDUCTIVE, EVIDENTIAL, AND MODEL-BASED REASONING. IN TOTAL, THE CAPABILITIES IMPLICIT IN THE ENVISIONED SYSTEM SHOULD BE OF MAJOR BENEFIT TO PLANT OPERATIONS.