Anomaly Detection in Liquid Sodium Cold Trap Operation with Multisensory Data Fusion Using Long Short-Term Memory Autoencoder

Akins, Alexandra; Kultgen, Derek; Heifetz, Alexander

doi:10.3390/en16134965

Anomaly Detection in Liquid Sodium Cold Trap Operation with Multisensory Data Fusion Using Long Short-Term Memory Autoencoder

Journal Article · Mon Jun 26 00:00:00 EDT 2023 · Energies

DOI:https://doi.org/10.3390/en16134965· OSTI ID:1986565

Akins, Alexandra; Kultgen, Derek;

Sodium-cooled fast reactors (SFR), which use high temperature fluid near ambient pressure as coolant, are one of the most promising types of GEN IV reactors. One of the unique challenges of SFR operation is purification of high temperature liquid sodium with a cold trap to prevent corrosion and obstructing small orifices. We have developed a deep learning long short-term memory (LSTM) autoencoder for continuous monitoring of a cold trap and detection of operational anomaly. Transient data were obtained from the Mechanisms Engineering Test Loop (METL) liquid sodium facility at Argonne National Laboratory. The cold trap purification at METL is monitored with 31 variables, which are sensors measuring fluid temperatures, pressures and flow rates, and controller signals. Loss-of-coolant type anomaly in the cold trap operation was generated by temporarily choking one of the blowers, which resulted in temperature and flow rate spikes. The input layer of the autoencoder consisted of all the variables involved in monitoring the cold trap. The LSTM autoencoder was trained on the data corresponding to cold trap startup and normal operation regime, with the loss function calculated as the mean absolute error (MAE). The loss during training was determined to follow log-normal density distribution. During monitoring, we investigated a performance of the LSTM autoencoder for different loss threshold values, set at a progressively increasing number of standard deviations from the mean. The anomaly signal in the data was gradually attenuated, while preserving the noise of the original time series, so that the signal-to-noise ratio (SNR) averaged across all sensors decreased below unity. Results demonstrate detection of anomalies with sensor-averaged SNR < 1.

View Journal Article

Research Organization:: Argonne National Laboratory (ANL), Argonne, IL (United States)

Sponsoring Organization:: USDOE Advanced Research Projects Agency - Energy (ARPA-E)

Grant/Contract Number:: AC02-06CH11357

OSTI ID:: 1986565

Alternate ID(s):: OSTI ID: 2397335

Journal Information:: Energies, Journal Name: Energies Journal Issue: 13 Vol. 16; ISSN 1996-1073; ISSN ENERGA

Publisher:: MDPI AGCopyright Statement

Country of Publication:: Switzerland

Language:: English

References (32)

A multiple-architecture deep learning approach for nuclear power plants accidents classification including anomaly detection and “don’t know” response Santos, Marcelo C.; Pereira, Claudio M. N. A.; Schirru, Roberto Annals of Nuclear Energy, Vol. 162 https://doi.org/10.1016/j.anucene.2021.108521	journal	November 2021
A review of inherent safety characteristics of metal alloy sodium-cooled fast reactor fuel against postulated accidents Sofu, Tanju Nuclear Engineering and Technology, Vol. 47, Issue 3 https://doi.org/10.1016/j.net.2015.03.004	journal	April 2015
Predictive maintenance architecture development for nuclear infrastructure using machine learning Gohel, Hardik A.; Upadhyay, Himanshu; Lagos, Leonel Nuclear Engineering and Technology, Vol. 52, Issue 7 https://doi.org/10.1016/j.net.2019.12.029	journal	July 2020
Eigendecomposition model of resistance temperature detector with applications to S-CO2 cycle sensing Heifetz, Alexander; Vilim, Richard Nuclear Engineering and Design, Vol. 311 https://doi.org/10.1016/j.nucengdes.2016.11.015	journal	January 2017
RNN-Based online anomaly detection in nuclear reactors for highly imbalanced datasets with uncertainty Kim, Minhee; Ou, Elisa; Loh, Po-Ling Nuclear Engineering and Design, Vol. 364 https://doi.org/10.1016/j.nucengdes.2020.110699	journal	August 2020
Sequential detection of a total instantaneous blockage occurred in a single subassembly of a sodium-cooled fast reactor Nikiforov, Igor; Harrou, Fouzi; Cogranne, Rémi Nuclear Engineering and Design, Vol. 366 https://doi.org/10.1016/j.nucengdes.2020.110733	journal	September 2020
Sodium purification systems for NPP with fast reactors (retrospective and perspective views) Kozlov, F. A.; Sorokin, A. P.; Konovalov, M. A. Nuclear Energy and Technology, Vol. 2, Issue 1 https://doi.org/10.1016/j.nucet.2016.03.002	journal	March 2016
Two neural network based strategies for the detection of a total instantaneous blockage of a sodium-cooled fast reactor Martinez-Martinez, Sinuhe; Messai, Nadhir; Jeannot, Jean-Philippe Reliability Engineering & System Safety, Vol. 137 https://doi.org/10.1016/j.ress.2014.12.003	journal	May 2015
Anomaly detection and diagnosis for wind turbines using long short-term memory-based stacked denoising autoencoders and XGBoost Zhang, Chen; Hu, Di; Yang, Tao Reliability Engineering & System Safety, Vol. 222 https://doi.org/10.1016/j.ress.2022.108445	journal	June 2022
3D convolutional and recurrent neural networks for reactor perturbation unfolding and anomaly detection Durrant, Aiden; Leontidis, Georgios; Kollias, Stefanos EPJ Nuclear Sciences & Technologies, Vol. 5 https://doi.org/10.1051/epjn/2019047	journal	January 2019
France–Japan synthesis concept on sodium-cooled fast reactor review of a joint collaborative work Rodriguez, Gilles; Varaine, Frédéric; Costes, Laurent EPJ Nuclear Sciences & Technologies, Vol. 7 https://doi.org/10.1051/epjn/2021014	journal	January 2021
A robust method for detection and classification of permanent magnet synchronous motor faults: Deep autoencoders and data fusion approach Sri Lal Senanayaka, Jagath; Robbersmyr, Kjell G. Journal of Physics: Conference Series, Vol. 1037 https://doi.org/10.1088/1742-6596/1037/3/032029	journal	June 2018
A Deep Learning Approach to Anomaly Detection in Nuclear Reactors Caliva, Francesco; De Ribeiro, Fabio Sousa; Mylonakis, Antonios 2018 International Joint Conference on Neural Networks (IJCNN) https://doi.org/10.1109/IJCNN.2018.8489130	conference	July 2018
Long Short-Term Memory Hochreiter, Sepp; Schmidhuber, Jürgen Neural Computation, Vol. 9, Issue 8 https://doi.org/10.1162/neco.1997.9.8.1735	journal	November 1997
A Survey on Deep Learning for Multimodal Data Fusion Gao, Jing; Li, Peng; Chen, Zhikui Neural Computation, Vol. 32, Issue 5 https://doi.org/10.1162/neco_a_01273	journal	May 2020
Multi-sensor data collection and fusion using autoencoders in condition evaluation of concrete bridge decks Pashoutani, Sepehr; Zhu, Jinying; Sim, Chungwook Journal of Infrastructure Preservation and Resilience, Vol. 2, Issue 1 https://doi.org/10.1186/s43065-021-00032-3	journal	June 2021
Sodium Purification by Cold Trapping at the Experimental Breeder Reactor II Holmes, J. T.; Smith, C. R. F.; Osterhout, M. M. Nuclear Technology, Vol. 32, Issue 3 https://doi.org/10.13182/NT77-A31754	journal	March 1977
Monitoring of Thermal Mixing Tee Sensors with LSTM Neural Networks Ankel, Victoria; Pantopoulou, Stella; Weathered, Matthew 12th Nuclear Plant Instrumentation, Control and Human-Machine Interface Technologies (NPIC&HMIT 2021) https://doi.org/10.13182/T124-34423	conference	January 2021
Thermodynamic and transport properties of sodium liquid and vapor Fink, J. K.; Leibowitz, L. https://doi.org/10.2172/94649	report	January 1995
Anomaly Detection of the Brake Operating Unit on Metro Vehicles Using a One-Class LSTM Autoencoder Kang, Jaeyong; Kim, Chul-Su; Kang, Jeong Won Applied Sciences, Vol. 11, Issue 19 https://doi.org/10.3390/app11199290	journal	October 2021
Unsupervised Anomaly Detection for Time Series Data of Spacecraft Using Multi-Task Learning Yang, Kaifei; Wang, Yakun; Han, Xiaodong Applied Sciences, Vol. 12, Issue 13 https://doi.org/10.3390/app12136296	journal	June 2022
Monitoring of Temperature Measurements for Different Flow Regimes in Water and Galinstan with Long Short-Term Memory Networks and Transfer Learning of Sensors Pantopoulou, Stella; Ankel, Victoria; Weathered, Matthew T. Computation, Vol. 10, Issue 7 https://doi.org/10.3390/computation10070108	journal	June 2022
Development of a Physics-Based Monitoring Algorithm Detecting CO2 Ingress Accidents in a Sodium-Cooled Fast Reactor Kim, Hyeonmin; Kim, Jung-Taek; Eoh, Jaehyuk Energies, Vol. 12, Issue 1 https://doi.org/10.3390/en12010001	journal	December 2018
Machine Learning Schemes for Anomaly Detection in Solar Power Plants Ibrahim, Mariam; Alsheikh, Ahmad; Awaysheh, Feras Energies, Vol. 15, Issue 3 https://doi.org/10.3390/en15031082	journal	February 2022
A Qualitative Strategy for Fusion of Physics into Empirical Models for Process Anomaly Detection Al Rashdan, Ahmad Y.; Abdel-Khalik, Hany S.; Giraud, Kellen M. Energies, Vol. 15, Issue 15 https://doi.org/10.3390/en15155640	journal	August 2022
Physics-Informed Neural Network Solution of Point Kinetics Equations for a Nuclear Reactor Digital Twin Prantikos, Konstantinos; Tsoukalas, Lefteri H.; Heifetz, Alexander Energies, Vol. 15, Issue 20 https://doi.org/10.3390/en15207697	journal	October 2022
Latent-Insensitive Autoencoders for Anomaly Detection Battikh, Muhammad S.; Lenskiy, Artem A. Mathematics, Vol. 10, Issue 1 https://doi.org/10.3390/math10010112	journal	December 2021
Fault Detection and Diagnosis Using Combined Autoencoder and Long Short-Term Memory Network Park, Pangun; Marco, Piergiuseppe Di; Shin, Hyejeon Sensors, Vol. 19, Issue 21 https://doi.org/10.3390/s19214612	journal	October 2019
Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network Hu, Di; Zhang, Chen; Yang, Tao Sensors, Vol. 20, Issue 21 https://doi.org/10.3390/s20216164	journal	October 2020
Anomaly Detection of Water Level Using Deep Autoencoder Nicholaus, Isack Thomas; Park, Jun Ryeol; Jung, Kyuil Sensors, Vol. 21, Issue 19 https://doi.org/10.3390/s21196679	journal	October 2021
Lightweight Long Short-Term Memory Variational Auto-Encoder for Multivariate Time Series Anomaly Detection in Industrial Control Systems Fährmann, Daniel; Damer, Naser; Kirchbuchner, Florian Sensors, Vol. 22, Issue 8 https://doi.org/10.3390/s22082886	journal	April 2022
Anomaly Detection Using Autoencoder Reconstruction upon Industrial Motors Givnan, Sean; Chalmers, Carl; Fergus, Paul Sensors, Vol. 22, Issue 9 https://doi.org/10.3390/s22093166	journal	April 2022

Similar Records

Preliminary Development of Heat Transfer Model-Based Control Algorithms of Liquid Sodium Purification System: Advanced Sensors and Instrumentation Advanced Controls

Technical Report · Sat Sep 30 00:00:00 EDT 2023 · OSTI ID:2212452

Advanced Monitoring and Control in the ANL METL Facility Using an Engineering Digital Twin

Technical Report · Fri Sep 15 00:00:00 EDT 2023 · OSTI ID:2204284

Related Subjects

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS
anomaly detection
artificial intelligence
cold trap
liquid sodium purification
long short-term memory autoencoder
sodium cooled fast reactors

Anomaly Detection in Liquid Sodium Cold Trap Operation with Multisensory Data Fusion Using Long Short-Term Memory Autoencoder

Citation Formats

References (32)

Similar Records

Related Subjects