Signal processing applied in cortex project: From noise analysis to OMA and SSA methods

In the framework of the CORTEX EU project, several measurements from different KWU and VVER reactors have been analyzed. The goal of this project is to develop and test core monitoring techniques to detect possible anomalies and better characterize the information provided by the neutron detectors. In this article, three different signal processing methods are presented: coolant velocity estimation, Singular Spectrum Analysis (SSA) and Operational Modal Analysis (OMA). The first method is based on detecting travelling perturbations by cross correlating different detectors and obtaining the impulse response function. The second method, SSA, reduces the noise and allows a better estimation of the frequency content with a high resolution. SSA has been tested both in simulated and plant data. The last method permits distinguishing closely spaced resonances and a multivariate analysis of all the detectors. In this work, the details of each methodology are explained step by step, and the results on each reactor are also presented. Testing these methods during CORTEX project gives valuable information and orientation for further investigations in reactors diagnostics.(c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).