A deep learning approach to anomaly detection in nuclear reactors
Paper in proceedings, 2018

In this work, a novel deep learning approach to unfold nuclear power reactor signals is proposed. It includes a combination of convolutional neural networks (CNN), denoising autoencoders (DAE) and k-means clustering of representations. Monitoring nuclear reactors while running at nominal conditions is critical. Based on analysis of the core reactor neutron flux, it is possible to derive useful information for building fault/anomaly detection systems. By leveraging signal and image pre-processing techniques, the high and low energy spectra of the signals were appropriated into a compatible format for CNN training. Firstly, a CNN was employed to unfold the signal into either twelve or forty-eight perturbation location sources, followed by a k-means clustering and k-Nearest Neighbour coarse-to-fine procedure, which significantly increases the unfolding resolution. Secondly, a DAE was utilised to denoise and reconstruct power reactor signals at varying levels of noise and/or corruption. The reconstructed signals were evaluated w.r.t. their original counter parts, by way of normalised cross correlation and unfolding metrics. The results
illustrate that the origin of perturbations can be localised with high accuracy, despite limited training data and obscured/noisy signals, across various levels of granularity.

signal processing

deep learning

denoising autoencoders

nuclear reactors

convolutional neural networks

clustering trained representations

unfolding

anomaly detection

Author

Francesco Calivá

University of Lincoln

Fabio De Sousa Ribeiro

University of Lincoln

Antonios Mylonakis

Chalmers, Physics, Subatomic and Plasma Physics

Christophe Demaziere

Chalmers, Physics, Subatomic and Plasma Physics

Paolo Vinai

Chalmers, Physics, Subatomic and Plasma Physics

Georgios Leontidis

University of Lincoln

Stefanos Kollias

University of Lincoln

2018 International Joint Conference on Neural Networks (IJCNN2018)
Rio de Janeiro, Brazil,

Core monitoring techniques and experimental validation and demonstration (CORTEX)

European Commission (Horizon 2020), 2017-09-01 -- 2021-08-31.

Subject Categories

Computer and Information Science

Other Physics Topics

Areas of Advance

Energy

DOI

10.1109/IJCNN.2018.8489130

ISBN

9781509060146

More information

Latest update

12/19/2018