A system for unattended surveillance of nuclear reactor behavior

Abstract

A multivariate statistical pattern recognition system for reactor noise analysis is presented. The basi3 of the system is a transformation for decou-' pling correlated variables and algorithms for inferring probability density functions. The system is adaptable to a variety of statistical properties of the data, and it ha3 learning, tracking, updating, and dlmensionelity reduction capabilities. System design emphasizes control of the false-alarm rate. Its abilities to learn normal patterns and to recognize deviations from these patterns were evaluated by experiments at the OBIIL High-Flux Isotope Reactor. Power perturbations of less than 0.1)5 of the mean value in selected frequency ranges were readily detected by the pattern recognition system.