Core diagnostics using noise analysis: From proof-of-principle to industrial demonstration

Paper in proceeding, 2017

Reactor noise analysis, based upon the monitoring of the deviations of typically the neutron flux from its mean value, has for objectives a) the early detection of anomalies before they have any inadvertent effect on plant safety and availability and b) the determination of dynamical core parameters. Because noise analysis is a non-intrusive technique, it can be used on-line while the reactor is running at nominal full power conditions. One of the challenges of noise diagnostics is nevertheless to be able to recover from very few neutron detector signals the nature and characteristics of the driving perturbation, localize it, and classify the severity of the anomaly. This requires competences in many areas, such as reactor physics and dynamics, reactor modelling, stochastic processes, signal analysis, and measurement techniques. This paper represents an attempt to pay a tribute to Dr. Pázsit’s seminal work on power reactor noise at the occasion of the special session organized in his honor at M&C 2017. Emphasis will be put on the development of innovative methods that resulted in industrial demonstrations at commercial nuclear power stations. The subjects covered hereafter are: the monitoring of control rod vibrations, the characterization and localization of anomalies, the diagnostics of BWR instabilities and the diagnostics of core barrel vibrations.