Journal of Nuclear Engineering

Research

Jump to: Review

9 pages, 1569 KiB

Open AccessArticle

Application of Machine Learning for Classification of Nuclear Reactor Operational Status Using Magnetic Field Sensors

by Braden Burt, Brett J. Borghetti, Anthony Franz, Darren Holland and Abigail Bickley

J. Nucl. Eng. 2023, 4(4), 723-731; https://doi.org/10.3390/jne4040045 - 6 Dec 2023

Cited by 1 | Viewed by 788

Abstract

The nuclear fuel cycle forms the basis for producing special nuclear materials used in nuclear weapons via a series of interdependent industrial operations. These industrial operations each produce characteristic emanations that can be gathered to ascertain signatures of facility operations. Machine learning and [...] Read more.

The nuclear fuel cycle forms the basis for producing special nuclear materials used in nuclear weapons via a series of interdependent industrial operations. These industrial operations each produce characteristic emanations that can be gathered to ascertain signatures of facility operations. Machine learning and deep learning techniques were applied to time series magnetic field sensor data collected at the High Flux Isotope Reactor (HFIR) to assess the feasibility of determining the ON/OFF operational state of the reactor. When data collected by the sensor near the cooling fans, position 9, are transformed to the frequency domain, it was found that both machine and deep learning methods were able to classify the operational state of the reactor with a balanced accuracy of over 90%. This result suggests that the utilized methods show promise for application as techniques to verify declared activities involving nuclear reactors. Additional effort is recommended to develop models and architectures that will more fully capitalize on the data’s temporal nature by incorporating the magnetic field’s time dependence to improve the model’s robustness and classification performance. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

19 pages, 1630 KiB

Open AccessArticle

SNM Radiation Signature Classification Using Different Semi-Supervised Machine Learning Models

by Jordan R. Stomps, Paul P. H. Wilson, Kenneth J. Dayman, Michael J. Willis, James M. Ghawaly and Daniel E. Archer

J. Nucl. Eng. 2023, 4(3), 448-466; https://doi.org/10.3390/jne4030032 - 4 Jul 2023

Viewed by 1689

Abstract

The timely detection of special nuclear material (SNM) transfers between nuclear facilities is an important monitoring objective in nuclear nonproliferation. Persistent monitoring enabled by successful detection and characterization of radiological material movements could greatly enhance the nuclear nonproliferation mission in a range of [...] Read more.

The timely detection of special nuclear material (SNM) transfers between nuclear facilities is an important monitoring objective in nuclear nonproliferation. Persistent monitoring enabled by successful detection and characterization of radiological material movements could greatly enhance the nuclear nonproliferation mission in a range of applications. Supervised machine learning can be used to signal detections when material is present if a model is trained on sufficient volumes of labeled measurements. However, the nuclear monitoring data needed to train robust machine learning models can be costly to label since radiation spectra may require strict scrutiny for characterization. Therefore, this work investigates the application of semi-supervised learning to utilize both labeled and unlabeled data. As a demonstration experiment, radiation measurements from sodium iodide (NaI) detectors are provided by the Multi-Informatics for Nuclear Operating Scenarios (MINOS) venture at Oak Ridge National Laboratory (ORNL) as sample data. Anomalous measurements are identified using a method of statistical hypothesis testing. After background estimation, an energy-dependent spectroscopic analysis is used to characterize an anomaly based on its radiation signatures. In the absence of ground-truth information, a labeling heuristic provides data necessary for training and testing machine learning models. Supervised logistic regression serves as a baseline to compare three semi-supervised machine learning models: co-training, label propagation, and a convolutional neural network (CNN). In each case, the semi-supervised models outperform logistic regression, suggesting that unlabeled data can be valuable when training and demonstrating value in semi-supervised nonproliferation implementations. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

21 pages, 1431 KiB

Open AccessArticle

Plutonium Signatures in Molten-Salt Reactor Off-Gas Tank and Safeguards Considerations

by Nicholas Dunkle, Alex Wheeler, Jarod Richardson, Sandra Bogetic, Ondrej Chvala and Steven E. Skutnik

J. Nucl. Eng. 2023, 4(2), 391-411; https://doi.org/10.3390/jne4020028 - 18 May 2023

Cited by 3 | Viewed by 2287

Abstract

Fluid-fueled molten-salt reactors (MSRs) are actively being developed by several companies, with plans to deploy them internationally. The current IAEA inspection tools are largely incompatible with the unique design features of liquid fuel MSRs (e.g., the complex fuel chemistry, circulating fuel inventory, bulk [...] Read more.

Fluid-fueled molten-salt reactors (MSRs) are actively being developed by several companies, with plans to deploy them internationally. The current IAEA inspection tools are largely incompatible with the unique design features of liquid fuel MSRs (e.g., the complex fuel chemistry, circulating fuel inventory, bulk accountancy, and high radiation environment). For these reasons, safeguards for MSRs are seen as challenging and require the development of new techniques. This paper proposes one such technique through the observation of the reactor’s off-gas. Any reactor design using low-enriched uranium will build up plutonium as the fuel undergoes burnup. Plutonium has different fission product yields than uranium. Therefore, a shift in fission product production is expected with fuel evolution. The passive removal of certain gaseous fission products to the off-gas tank of an MSR provides a valuable opportunity for analysis without significant modifications to the design of the system. Uniquely, due to the gaseous nature of the isotopes, beta particle emissions are available for observation. The ratios of these fission product isotopes can, thus, be traced back to the relative amount and types of fissile isotopes in the core. This proposed technique represents an effective safeguards tool for bulk accountancy which, while avoiding being onerous, could be used in concert with other techniques to meet the IAEA’s timeliness goals for the detection of a diversion. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

17 pages, 6858 KiB

Open AccessArticle

Fast-, Light-Cured Scintillating Plastic for 3D-Printing Applications

by Brian G. Frandsen, Michael Febbraro, Thomas Ruland, Theodore W. Stephens, Paul A. Hausladen, Juan J. Manfredi and James E. Bevins

J. Nucl. Eng. 2023, 4(1), 241-257; https://doi.org/10.3390/jne4010019 - 7 Mar 2023

Cited by 6 | Viewed by 2158

Abstract

Additive manufacturing techniques enable a wide range of possibilities for novel radiation detectors spanning simple to highly complex geometries, multi-material composites, and metamaterials that are either impossible or cost prohibitive to produce using conventional methods. The present work identifies a set of promising [...] Read more.

Additive manufacturing techniques enable a wide range of possibilities for novel radiation detectors spanning simple to highly complex geometries, multi-material composites, and metamaterials that are either impossible or cost prohibitive to produce using conventional methods. The present work identifies a set of promising formulations of photocurable scintillator resins capable of neutron-gamma pulse shape discrimination (PSD) to support the additive manufacturing of fast neutron detectors. The development of these resins utilizes a step-by-step, trial-and-error approach to identify different monomer and cross-linker combinations that meet the requirements for 3D printing followed by a 2-level factorial parameter study to optimize the radiation detection performance, including light yield, PSD, optical clarity, and hardness. The formulations resulted in hard, clear, PSD-capable plastic scintillators that were cured solid within 10 s using 405 nm light. The best-performing scintillator produced a light yield 83% of EJ-276 and a PSD figure of merit equaling 1.28 at 450–550 keVee. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

19 pages, 4489 KiB

Open AccessArticle

Data Augmentation for Neutron Spectrum Unfolding with Neural Networks

by James McGreivy, Juan J. Manfredi and Daniel Siefman

J. Nucl. Eng. 2023, 4(1), 77-95; https://doi.org/10.3390/jne4010006 - 3 Jan 2023

Viewed by 2187

Abstract

Neural networks require a large quantity of training spectra and detector responses in order to learn to solve the inverse problem of neutron spectrum unfolding. In addition, due to the under-determined nature of unfolding, non-physical spectra which would not be encountered in usage [...] Read more.

Neural networks require a large quantity of training spectra and detector responses in order to learn to solve the inverse problem of neutron spectrum unfolding. In addition, due to the under-determined nature of unfolding, non-physical spectra which would not be encountered in usage should not be included in the training set. While physically realistic training spectra are commonly determined experimentally or generated through Monte Carlo simulation, this can become prohibitively expensive when considering the quantity of spectra needed to effectively train an unfolding network. In this paper, we present three algorithms for the generation of large quantities of realistic and physically motivated neutron energy spectra. Using an IAEA compendium of 251 spectra, we compare the unfolding performance of neural networks trained on spectra from these algorithms, when unfolding real-world spectra, to two baselines. We also investigate general methods for evaluating the performance of and optimizing feature engineering algorithms. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

30 pages, 1315 KiB

Open AccessArticle

Proximal Policy Optimization for Radiation Source Search

by Philippe Proctor, Christof Teuscher, Adam Hecht and Marek Osiński

J. Nucl. Eng. 2021, 2(4), 368-397; https://doi.org/10.3390/jne2040029 - 30 Sep 2021

Cited by 9 | Viewed by 3430

Abstract

Rapid search and localization for nuclear sources can be an important aspect in preventing human harm from illicit material in dirty bombs or from contamination. In the case of a single mobile radiation detector, there are numerous challenges to overcome such as weak [...] Read more.

Rapid search and localization for nuclear sources can be an important aspect in preventing human harm from illicit material in dirty bombs or from contamination. In the case of a single mobile radiation detector, there are numerous challenges to overcome such as weak source intensity, multiple sources, background radiation, and the presence of obstructions, i.e., a non-convex environment. In this work, we investigate the sequential decision making capability of deep reinforcement learning in the nuclear source search context. A novel neural network architecture (RAD-A2C) based on the advantage actor critic (A2C) framework and a particle filter gated recurrent unit for localization is proposed. Performance is studied in a randomized

20 \times 20

m convex and non-convex simulation environment across a range of signal-to-noise ratio (SNR)s for a single detector and single source. RAD-A2C performance is compared to both an information-driven controller that uses a bootstrap particle filter and to a gradient search (GS) algorithm. We find that the RAD-A2C has comparable performance to the information-driven controller across SNR in a convex environment. The RAD-A2C far outperforms the GS algorithm in the non-convex environment with greater than

95 %

median completion rate for up to seven obstructions. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

17 pages, 394 KiB

Open AccessArticle

Neural Network Approaches for Mobile Spectroscopic Gamma-Ray Source Detection

by Kyle J. Bilton, Tenzing H. Y. Joshi, Mark S. Bandstra, Joseph C. Curtis, Daniel Hellfeld and Kai Vetter

J. Nucl. Eng. 2021, 2(2), 190-206; https://doi.org/10.3390/jne2020018 - 17 May 2021

Cited by 7 | Viewed by 3427

Abstract

Artificial neural networks (ANNs) for performing spectroscopic gamma-ray source identification have been previously introduced, primarily for applications in controlled laboratory settings. To understand the utility of these methods in scenarios and environments more relevant to nuclear safety and security, this work examines the [...] Read more.

Artificial neural networks (ANNs) for performing spectroscopic gamma-ray source identification have been previously introduced, primarily for applications in controlled laboratory settings. To understand the utility of these methods in scenarios and environments more relevant to nuclear safety and security, this work examines the use of ANNs for mobile detection, which involves highly variable gamma-ray background, low signal-to-noise ratio measurements, and low false alarm rates. Simulated data from a 2” × 4” × 16” NaI(Tl) detector are used in this work for demonstrating these concepts, and the minimum detectable activity (MDA) is used as a performance metric in assessing model performance.In addition to examining simultaneous detection and identification, binary spectral anomaly detection using autoencoders is introduced in this work, and benchmarked using detection methods based on Non-negative Matrix Factorization (NMF) and Principal Component Analysis (PCA). On average, the autoencoder provides a 12% and 23% improvement over NMF- and PCA-based detection methods, respectively. Additionally, source identification using ANNs is extended to leverage temporal dynamics by means of recurrent neural networks, and these time-dependent models outperform their time-independent counterparts by 17% for the analysis examined here. The paper concludes with a discussion on tradeoffs between the ANN-based approaches and the benchmark methods examined here. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures

Figure 1

Review

Jump to: Research

35 pages, 2591 KiB

Open AccessReview

Current and Prospective Radiation Detection Systems, Screening Infrastructure and Interpretive Algorithms for the Non-Intrusive Screening of Shipping Container Cargo: A Review

by Euan L. Connolly and Peter G. Martin

J. Nucl. Eng. 2021, 2(3), 246-280; https://doi.org/10.3390/jne2030023 - 7 Aug 2021

Cited by 9 | Viewed by 4832

Abstract

The non-intrusive screening of shipping containers at national borders serves as a prominent and vital component in deterring and detecting the illicit transportation of radioactive and/or nuclear materials which could be used for malicious and highly damaging purposes. Screening systems for this purpose [...] Read more.

The non-intrusive screening of shipping containers at national borders serves as a prominent and vital component in deterring and detecting the illicit transportation of radioactive and/or nuclear materials which could be used for malicious and highly damaging purposes. Screening systems for this purpose must be designed to efficiently detect and identify material that could be used to fabricate radiological dispersal or improvised nuclear explosive devices, while having minimal impact on the flow of cargo and also being affordable for widespread implementation. As part of current screening systems, shipping containers, offloaded from increasingly large cargo ships, are driven through radiation portal monitors comprising plastic scintillators for gamma detection and separate, typically ³He-based, neutron detectors. Such polyvinyl-toluene plastic-based scintillators enable screening systems to meet detection sensitivity standards owing to their economical manufacturing in large sizes, producing high-geometric-efficiency detectors. However, their poor energy resolution fundamentally limits the screening system to making binary “source” or “no source” decisions. To surpass the current capabilities, future generations of shipping container screening systems should be capable of rapid radionuclide identification, activity estimation and source localisation, without inhibiting container transportation. This review considers the physical properties of screening systems (including detector materials, sizes and positions) as well as the data collection and processing algorithms they employ to identify illicit radioactive or nuclear materials. The future aim is to surpass the current capabilities by developing advanced screening systems capable of characterising radioactive or nuclear materials that may be concealed within shipping containers. Full article

(This article belongs to the Special Issue Nuclear Security and Nonproliferation Research and Development)

► Show Figures