Use of neurals networks in nuclear power plant diagnostics

A technique using neural networks as a means of diagnosing transients or abnormal conditions in nuclear power plants is investigated and found to be feasible. The technique is based on the fact that each physical state of the plant can be represented by a unique pattern of sensor outputs or instrument readings that can be related to the condition of the plant. Neural networks are used to relate this pattern to the fault, problem, or transient condition of the plant. A demonstration of the ability of this technique to identify causes of perturbations in the steam generator of a nuclear plant is presented. 3 refs., 4 figs.