Maximum Likelihood Factor Analysis of the Reactor Coolant Pump System

In today’s operating environment of nuclear power plants, setpoints are established for key plant parameters, such as temperature, pressure, and flow rate. Reducing excursions beyond these setpoints would save millions of dollars as a result of improved plant availability and improve plant safety as well. The statistical method of maximum likelihood factor analysis is presented, and the results of two computer runs are given. The results of the statistical analysis indicate that it is possible to consistently rank order the eleven tracked variables of the reactor coolant system. Implementation of the maximum likelihood factor method would permit the decision maker to predict unanticipated transients and reduce plant unavailability.