The applicability of expert systems to the control and operation of nuclear power generating stations

The authors present the principal issues that relate to the viability of expert system control in the nuclear power arena. It is noted that bringing a nuclear power generating station under the complete control of an expert system will require an extensive program of coordinated research, development, prototype testing, regulatory review, and just plain selling the idea. This is especially true for those plants already licensed and operating for over a decade. The problem is far beyond that of a straightforward, though lengthy, exercise in computer programming. Conversion of nuclear energy to electrical energy carries with it a myriad of unique and complicated issues. Some will be purely technical in nature; these the programmer must build into the expert system's rules and subroutines. Others will be less than technical, and more esoteric in nature; however, these impose constraints on the programmer, as well as on the utility, the ultimate user, that are just as real as the constraints of existing technology and the constraints of nature itself.<>