An expert system that aids in severe accident management

Operations personnel at commercial nuclear power plants receive extensive training on the recognition of and proper response to design base accidents and plant transients. However, operators receive little or no training on the recognition and response to severe accidents. A severe accident is any event that produces an uncontrolled loss of reactor coolant. To assist with this problem a study was performed to determine the feasibility of developing an expert system that would act as an aid to reactor operations personnel in the event of a severe accident. The study centered around the development of a prototype expert system called SAIM (Severe Accident Information Manager). The development of the SAIM prototype was achieved in a Lisp environment using the expert system shell PC Plus by Texas Instruments. This is a rule-based system using confidence factors and fuzzy logic to account for uncertainty. Verification of the prototype was based on results of Severe Accident research and data from the Three Mile Island Unit 2 accident. The prototype testing and subsequent feasibility study revealed that development of a full-scale SAIM system would be achievable and beneficial. However, issues like regulatory acceptance, maintenance, and signal validation still remain unresolved, and may prove to be hindrances to full-scale development.