Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “detector simulation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Temperature trends and correlation between SQUID superparamagnetic relaxometry and dc-magnetization on model iron-oxide nanoparticles

Structural properties of iron-oxide nanoparticles deeply affect their magnetic performance in many applications such as with superparamagnetic relaxometry, when employed as cell-targeted magnetic nanoparticles for in vivo diagnostics. In this work, we present a detailed characterization of model nanoparticles for this application, with an average size of ~25 nm and a narrow size dispersion (σ < 7%). Considering the intrinsic structural properties of these model nanoparticles, the study of temperature dependence and correlation between dc-magnetization and superconducting quantum interference detector-relaxometry are discussed based on known theoretical predictions and computer simulations of the magnetic dipole moment and characteristic decay constants. Furthermore, computer simulations provide support in clarifying how important the overall collective magnetization is affected by particle size dispersion, which has a direct role on sustaining the magnetic relaxation signal in the temperature range required in preclinical and clinical settings.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

Elastic depths for detecting shape anomalies in functional data

This this paper, we propose a new family of depth measures called the elastic depths that can be used to greatly improve shape anomaly detection in functional data. Shape anomalies are functions that have considerably different geometric forms or features from the rest of the data. Identifying them is generally more difficult than identifying magnitude anomalies because shape anomalies are often not distinguishable from the bulk of the data with visualization methods. The proposed elastic depths use the recently developed elastic distances to directly measure the centrality of functions in the amplitude and phase spaces. Measuring shape outlyingness in these spaces provides a rigorous quantification of shape which, in turn, gives the elastic depths a strong theoretical and practical advantage over other methods in detecting shape anomalies. A simple boxplot and thresholding method are introduced to identify shape anomalies using the elastic depths. We assess the elastic depth's detection skill on simulated shape outlier scenarios and compare them against popular shape anomaly detectors. Finally, bond yields, image outlines, and hurricane trajectories are used to demonstrate our method's applicability to functional data observed on three different manifolds.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Practical considerations for measuring global spin density matrix elements of vector mesons in heavy-ion collisions

The STAR Collaboration has reported a significant 𝜙-meson global spin alignment (𝜌 00 ) signal in Au+Au collisions at $\sqrt{s_{NN}}$ ≤ 62 GeV by measuring the polar angle distribution of 𝜙-meson daughters with respect to the orbital angular momentum direction of the collision system. Here, in this paper, a new method is explored for studying vector-meson global spin alignment in heavy-ion collisions by examining the two-dimensional (2D) polar and azimuthal angle distribution. This method allows simultaneous extraction of 𝜌 00 and off-diagonal spin density matrix elements (SDMEs), providing unique access to local quark-antiquark spin correlations and spin hydrodynamics in quark-gluon plasma. The new 2D method also removes potential biases from nonzero off-diagonal SDMEs on 𝜌 00 with the 1D method. A detailed procedure to correct for detector acceptance and resolution effects is also presented and validated by simulation studies.

Wilks, Gavin [University of Illinois, Chicago, IL ↗
DOE OSTI2023

Characterizing a supernova’s standing accretion shock instability with neutrinos and gravitational waves

Here, we perform a novel multi-messenger analysis for the identification and parameter estimation of the Standing Accretion Shock Instability (SASI) in a core collapse supernova with neutrino and gravitational wave (GW) signals. In the neutrino channel, this method performs a likelihood ratio test for the presence of SASI in the frequency domain. For gravitational wave signals we process an event with a modified constrained likelihood method. Using simulated supernova signals, the properties of the Hyper-Kamiokande neutrino detector, and O3 LIGO Interferometric data, we produce the two-dimensional probability density distribution (PDF) of the SASI activity indicator and calculate the probability of detection P D as well as the false identification probability P FI . We discuss the probability to establish the presence of the SASI as a function of the source distance in each observational channel, as well as jointly. Compared to a single-messenger approach, the joint analysis results in P D (at P FI = 0.1) of SASI activities that is larger by up to ≈ 40% for a distance to the supernova of 5 kpc. We also discuss how accurately the frequency and duration of the SASI activity can be estimated in each channel separately. Our methodology is suitable for implementation in a realistic data analysis and a multi-messenger setting.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

New Measurements on the Neutrino Mass: KATRIN and Project 8 (Final Technical Report for DE-SC0014130)

The overarching goal of the funded research is to build experiments and develop techniques towards the major goal of measuring the mass of the neutrino, particularly by the method of high-resolution spectroscopy of the endpoint of tritium. The goals of this project were to make hardware and intellectual contributions towards the KATRIN experiment, an international project under construction in Germany, and towards the "Project 8" R&D program to develop new electron spectroscopy methods and apply them to tritium. A minor goal of the project was to support simulation work and R&D on future neutrino and dark matter detectors.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Reaction Rate Ratios for Recent Fast Metal Experiments with Large Plutonium Masses

Reaction rate ratios are integral responses that are used within the criticality experiments field because they contain spectral information. While these types of measurements have been utilized for nuclear data validation with historic experiments, few experiments of this type have been utilized for recent experiments, as few exist. This work focuses on measured reaction rate ratios for two nearly bare plutonium critical assemblies with different geometries: one that is cube like (with a Pu mass of 40 kg) and one that is slab like (with a Pu mass of 109 kg). Irradiations were performed with both configurations in which foils were placed near the center of the assembly. Plutonium, highly enriched uranium, depleted uranium, and Au foils were included in the irradiation and counted via high-purity germanium detectors. From these measurements, reaction rate ratios were calculated. Measured and simulated values and uncertainties are presented for the reaction rate ratios. Ratios utilizing the following reactions are given in this work: 197 Au(n, γ), 197 Au(n,2n), 235 U(n,fission), 238 U(n, fission), 238 U(n,2n), 238 U(n,γ), and 239 Pu(n,fission). Uncertainties for the measured reaction rate ratios ranged from 4% to 7%, and the contribution of various parameters to this uncertainty was investigated. The results are compared to historical experiments and should be used for nuclear data validation for future nuclear data library releases. These measurements are part of the EUCLID (Experiments Underpinned by Computational Learning for Improvements in Nuclear Data) project, which utilizes measurement responses in addition to k eff (such as these reaction rate ratios) to help reduce uncertainties in 239 Pu nuclear data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

Radioactive Source Localization via Bayesian Particle Filter

In the event of a misplaced radioactive source or other emergency situation, measuring a radiation field, mapping its distribution, and determining a source location are essential tasks to ameliorating the situation. However, radiation fields may be extremely hazardous to human surveyors and minimizing received radiation doses is just as essential. Robots appear to be a potential solution to these problems. Beyond simply measuring radiation, the robot's computer processing capabilities offer a way to apply complex data analysis methods to radiation measurements in real-time. Methods which predict likely source locations can then feed this information into other processes, potentially improving path planning and enabling more efficient measurements. Given a robot mounted with a gamma-ray detector, can we: develop a methodology to account for detector performance across a wide range of source angles, distances, and photon energies? operate an autonomously navigating robot to effectively survey and characterize an area of interest? implement a data analysis method, conventionally used in measurements of motion, for source localization purposes? An open-source TurtleBot 3 robot, running Robot Operating System (ROS) on Ubuntu 16.04 LTS, was fitted with a Kromek GR1{sup R} Cadmium Zinc Telluride (CZT) solid-state gamma-ray detector. As a part of ROS, the packages OpenSlam, gmapping, and amcl were used to perform Simultaneous Localization and Mapping (SLAM), determining the robot's position and mapping the surrounding area. Data was acquired via Lidar mounted on top the TurtleBot 3. Detector Calibration Fit: The equation was fit to 365 counts of various energies, distances, and angles. A MATLAB{sup R} program was written to simulate measurements taken a robot on a random walk, with count data and positions discretized into finite element pixels. Using this program, a sample of 100 runs was performed on a map with a simulated source at the center, with a total of 200 of 2 pixels each. Similarly, multiple runs of the filter were performed on recorded robot measurement data. In both simulation and real tests, when corrected for errors (particles placed outside of bounds or on the robot, and simulation-specific errors), corresponding t-tests of predicted x and y-coordinates were within a 95% confidence interval of the actual position. For the real trial, these positions are slightly skewed right in the x-axis as the robot remained mainly to the left side of the source within the sample area. These simulations demonstrate potential validity for the usage of a particle filter as method of radioactive source localization. In the future, true real-time implementation and data fusion may further augment the performance of the robot to localize lost sources. Additionally, identification of multiple sources, determination of source types, and usage of a collimator are areas to potentially be explored.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2026

A compact, low-power epithermal neutron counter for lunar water detection

The detection and characterization of lunar water are critical for enabling sustainable human and robotic exploration of the Moon. Orbital neutron spectrometers, such as instruments on Lunar Prospector and the Lunar Reconnaissance Orbiter, have revealed hydrogen-rich regions near the poles but are limited by coarse spatial resolution and low counting efficiency. We present a compact, lightweight, and low-power epithermal neutron detector based on boron-coated silicon imagers, designed to probe subsurface hydrogen at decimeter scales from mobile platforms such as lunar rovers. This instrument leverages the high neutron capture cross-section of 10 B to convert epithermal neutrons into detectable α and 7 Li ions in a fully-depleted silicon imager, providing a unique event topology to identify neutrons while suppressing backgrounds. Monte Carlo simulations demonstrate that a 3 μm boron layer achieves optimal neutron detection efficiency, further enhanced with polyethylene moderation to improve sensitivity to the 0.4 eV–500 keV epithermal energy range. For a 10 cm 2 active area, the detector achieves sensitivity to H 2 O weight fractions as low as 0.01 wt% in a 15 minute measurement. This scalable, portable, low-mass design is well-suited for integration into upcoming Artemis and commercial lunar rovers, providing a transformative capability for in-situ resource prospecting and ground-truth validation of orbital measurements.

Detector modelling and simulations I (interaction ↗
DOE OSTI2022

High-performance and high-fidelity Monte Carlo solutions to the BEAVRS benchmark

The BEAVRS (Benchmark for Evaluation and Validation of Reactor Simulation) benchmark is solved by PRAGMA, the GPU-based continuous energy Monte Carlo code. The resulting solutions are comprised of the detailed simulation results of two cycles, each of which consists of the zero power physics test (ZPPT) and the core depletion calculations. The ZPPT consists of characteristic parameters, such as critical boron concentration (CBC), control rod bank worth, isothermal temperature coefficients, and assembly-wise detector signal, which are compared with measured data provided by the benchmark administration. The core depletion calculations were performed for both the hot full power and the load follow modes, and the comparison was made with the measured or deduced CBCs and assembly-wise detector signals. In the load follow calculations, the operating power history was approximately applied to simulate the real operation as closely as possible. PRAGMA performed the various calculations with a tremendous number of histories ranging up to hundreds of millions per cycle, exploiting GPUs' massively parallel performance. The load follow run time was shorter than 16 hours on a single rack of computing nodes mounded with 24 gaming GPUs with a remarkable agreement with the measurements for most comparisons. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Identify and Assess Technical Challenges in Safeguards Measurements of Spent Advanced Reactor Fuels

Advanced reactor (AR) designs use various nuclear fuel types that can be significantly different than conventional light-water reactor (LWR) fuels, including differences in sizes, compositions, and chemical forms (e.g., oxide, carbide, metal). Nearly all the proposed AR fuels use high-assay low-enriched uranium (HALEU), which will have higher enrichments (5–20 wt% 235 U) than LWR fuels (currently limited to <5 wt% 235 U). In advance of the wide use of these new fuel types around the world, international safeguards organizations such as the International Atomic Energy Agency (IAEA]) are working with some of the AR vendors to formulate safeguards approaches for these AR fuel cycles. As part of the overall safeguards approach, it is important to identify the potential technical challenges in performing safeguards verification measurements of these AR fuels (both fresh and spent fuels) in advance of the widespread adoption of these new fuel types, because new safeguards technologies can take several years to develop, test, and approve for use. This report documents work performed in fiscal year 2024 based on modeling and simulation to assess the performance of the existing safeguards measurement technologies for irradiated or spent AR fuel elements or items. This work is a continuation of the work performed in fiscal year 2023 that focused on fresh AR fuels. Spent AR fuels have a distinct difference from their LWR counterparts: unlike the spent LWR fuels typically stored in a water-filled pool, some spent AR fuels—such as tristructural-isotropic (TRISO)-based fuels—will most likely be stored in air-filled hot cells. Because most safeguards measurements on spent fuel performed to date have been conducted under water, the air-filled hot cell environment could present unique challenges to safeguards measurements. Fork detector (FDET) and Cerenkov viewing device (CVD) systems have been the two primary instruments used by the IAEA for several decades to measure spent LWR fuel assemblies stored in pools for safeguards verification purposes. Because the lower refractive index of air causes Cerenkov light to be of lower intensity in air than in water, existing CVDs are likely unable to perform safeguards verification measurements for spent fuel stored in an air-filled hot cell, as is the case for the TRISO-based spent fuel elements (e.g., pebbles, graphite fuel blocks). Unlike FDET measurements, CVD measurements do not require fuel be moved, so they are a simpler and faster to take than FDET measurements. The inability to perform CVD measurements on the TRISO-based AR fuel types presents a major technical challenge in the effort to use existing technology to perform safeguards measurements on spent AR fuels. This study was mainly conducted through the modeling and simulation of an FDET or an FDET-like system on five spent AR fuel types, including one metallic fuel type and four TRISO-based fuel types in both pebble and graphite block forms in their respective storage configurations and environments. Because the various AR fuel types have significantly different dimensions, FDET systems must be adapted to accommodate them. Partial defect tests were also simulated in this study to assess the FDET’s ability to detect potential fuel diversions. The FDET measures the fuel’s total passive neutron and gamma emissions. The simulated FDET results from spent AR fuel items are compared against results from a typical spent pressurized water reactor (PWR) assembly. High-purity germanium (HPGe) gamma detector measurements were also simulated for the spent AR fuel types and the PWR assembly because the signature photopeaks have been used in LWR safeguards verifications, although HPGe is usually not used to detect diversions because of the fuel’s self-attenuation effects on those photopeaks. The results indicate that these detectors have significant challenges in performing safeguards measurements of the spent AR fuel items, including incompatibilities between AR fuel items and existing FDETs, lower neutron count rates, lower sensitivities to fuel diversions in certain AR fuel items, and significantly higher interference from a neighboring fuel item when the measurement is performed in air. These results suggest that an alternative technology or significant and timely technology development is needed to perform adequate safeguards measurements of some of these AR fuel items.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Local structure of zinc–indium–tin oxide films via grazing-incidence x-ray pair-distribution functions and theoretical methods

A detailed experimental and theoretical study on the local (r ≤ 4.5 Å) atomic structure of amorphous and crystalline zinc–indium–tin oxide (ZITO) thin films using grazing-incidence x-ray Pair-Distribution Functions (PDFs), ab initio Molecular Dynamics (MD), and Empirical Potential Structure Refinement (EPSR) Monte Carlo simulations is presented. High-energy synchrotron x rays, a two-dimensional detector, and different incident angles were used to probe the depth uniformity of five (ZnO) 0.15 (In 2 O 3 ) 0.70 (SnO 2 ) 0.15 films that were deposited via pulsed-laser deposition at growth temperatures (T G ) ranging from 25 to 300 °C. Films deposited at T G ≤ 150 °C were amorphous. The partially crystalline (T G = 200 °C) and fully crystalline (T G = 300 °C) films were highly textured. Both crystalline and amorphous structures were investigated using ab initio MD and EPSR Monte Carlo simulations. The density of the amorphous films determined from the experimental data agreed with MD calculations. Coordination numbers, bond lengths, and distortion for metal–oxygen and for both the edge- and corner-shared In–metal shells up to 4.5 Å obtained from PDF analysis closely agreed with MD and EPSR simulations. There is a pronounced decrease in the edge- and corner-shared In–Zn distances arising from the shorter Zn–O bond length, Zn–O tetrahedral coordination, and In–O–Zn angle in amorphous ZITO compared to its crystalline counterpart. A maximum in electrical mobility was observed for the amorphous film just before crystallization occurred. While the peak is broad, consistent with nearly unchanged overall cation–oxygen coordination in the amorphous films, ESPR results indicate that the tetrahedral coordination follows the conductivity trend.

Grazing Incidence X-ray↗
DOE OSTI2024

Smart pixel sensors: towards on-sensor filtering of pixel clusters with deep learning

Highly granular pixel detectors allow for increasingly precise measurements of charged particle tracks. Next-generation detectors require that pixel sizes will be further reduced, leading to unprecedented data rates exceeding those foreseen at the High- Luminosity Large Hadron Collider. Signal processing that handles data incoming at a rate of $\mathcal{O}$(40 MHz) and intelligently reduces the data within the pixelated region of the detector at rate will enhance physics performance at high luminosity and enable physics analyses that are not currently possible. Using the shape of charge clusters deposited in an array of small pixels, the physical properties of the traversing particle can be extracted with locally customized neural networks. In this first demonstration, we present a neural network that can be embedded into the on-sensor readout and filter out hits from low momentum tracks, reducing the detector's data volume by 57.1%–75.7%. The network is designed and simulated as a custom readout integrated circuit with 28 nm CMOS technology and is expected to operate at less than 300 μW with an area of less than 0.2 mm 2 . The temporal development of charge clusters is investigated to demonstrate possible future performance gains, and there is also a discussion of future algorithmic and technological improvements that could enhance efficiency, data reduction, and power per area.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Demonstration of low-mode shape control in indirect-drive double shell implosions at the NIF

Double shell inertial confinement fusion is a concept for achieving robust thermonuclear burn that uses dense metal shells to compress deuterium-tritium (DT) fuel to fusion conditions. Double shell implosions are typically indirectly driven and involve a target that consists of a low-Z ablator, a foam layer, and a high-Z pusher surrounding the DT fuel. The goal of the campaign is to achieve a volumetric burn as radiation losses from the DT fuel are trapped by the opaque high-Z shell. The overall performance of double shell implosions relies on the efficient collisional transfer of kinetic energy between layers. The efficiency of this transfer (and therefore the overall performance of a given implosion) is degraded by the presence of low-mode asymmetries. P2 asymmetries are often observed in spatially resolved 2D radiographs of nominal double shell implosions. This work discusses three such experiments: one with an oblate P2 asymmetry, one with a prolate P2 asymmetry, and one with an approximate spherical symmetry. After performing a shape analysis of the oblate and prolate implosions to quantify asymmetries, these experimental results were compared with the results of hydrodynamic simulations for the two experiments. Differences between the experiment and simulation were then used to design an approximately spherical implosion by altering the incident laser cone fraction. Radiographs from the experiment that implemented the modified cone fraction show evidence of an implosion that is approximately spherical until bang time. This design is intended to serve as a point design for future studies that will seek to optimize various aspects of the double shell target.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

DUAL-PARTICLE DOSEMETER BASED ON ORGANIC SCINTILLATOR

Abstract Traditionally available handheld dosemeters are generally sensitive to only one type of radiation: neutrons or photons. Some dosemeters also rely on very specific attenuation correlations between response and dose, are not scalable in size and multiple dosemeters are required to characterise mixed-particle fields. The research presented here serves as a proof-of-concept for a method to simultaneously measure dose rates from neutrons and photons using a particle discriminating organic scintillation detector without the need for spectral deconvolution. The method was compared with traditional instruments and to simulation. Isotopic photon dose rates measured with this method were within 4% of simulated truth, whereas fission spectrum neutron dose rates were measured within 21%. Measurements of dose rates from both particles agree with simulated truth better than traditional instruments. This new method allows for measurement of dose equivalent from both neutrons and photons with a single instrument and no reliance on spectral deconvolution.

Miller, C. A.↗
DOE OSTI2021

Muon Radiography with the NOvA Near Detector

Information on the geological and man-made structures, directly contributing to the overburden of the NOvA Near Detector, located 100 m underground at Fermilab, are extracted from the cosmic muon data accumulated in the period April 2015 - October 2016. Results of this radiographic analysis of cosmic data are obtained by exploiting the acceptance symmetries of the NOvA detector. Radiographic maps of the overburden variations are acquired without subtracting the surface muon flux (free sky data) and without MC simulation of the cosmic muons attenuation in the homogeneous and flat overburden. By directly subtracting the angular muon track data measured by the NOvA Near Detector, arriving from directions having the same attenuation length (in the flat-surface approximation) and belonging to the acceptance regions with the same reconstruction efficiency, one can efficiently obtain two-dimensional differential radiographic maps.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Three-dimensional reconstruction of inertial confinement fusion hot-spot plasma from x-ray and nuclear diagnostics on OMEGA

Multidimensional effects degrade the neutron yield and the compressed areal density of laser-direct-drive inertial confinement fusion implosions of layered deuterium–tritium cryogenic targets on the OMEGA Laser System with respect to 1D radiation-hydrodynamic simulation predictions. A comprehensive physics-informed 3D reconstruction effort is under way to infer hot-spot and shell conditions at stagnation from four x-ray and seven neutron detectors distributed around the OMEGA target chamber. Neutron diagnostics, providing measurements of the neutron yield, hot-spot flow velocity, and apparent ion-temperature distribution, are used to infer the mode-1 perturbation at stagnation. The x-ray imagers record the shape of the hot-spot plasma to diagnose mode-1 and mode-2 perturbations. A deep-learning convolutional neural network trained on an extensive set of 3D radiation-hydrodynamic simulations is used to interpret the x-ray and nuclear measurements to infer the 3D profiles of the hot-spot plasma conditions and the amount of laser energy coupled to the hot-spot plasma. A 3D simulation database shows that larger mode-1 asymmetries are correlated with higher hot-spot flow velocities and reduced laser-energy coupling and neutron yield. Three-dimensional hot-spot reconstructions from x-ray measurements indicate that higher amounts of residual kinetic energy are correlated with higher measured hot-spot flow velocities, consistent with 3D simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Measurement of muon-induced high-energy neutrons from rock in an underground Gd-doped water detector

In this paper, we present a measurement of the rate of correlated neutron captures in the WATCHBOY detector, deployed at a depth of approximately 390 meters water equivalent (m.w.e.) in the Kimballton Underground Research Facility (KURF). WATCHBOY consists of a cylindrical 2 ton water target doped with 0.1% gadolinium, surrounded by a 40 ton undoped water hermetic shield. We present a comparison of our results with the expected rate of correlated neutron captures arising from high-energy neutrons incident on the outside of the WATCHBOY shield, predicted by a hybrid FLUKA/GEANT4-based simulation. The incident neutron energy distribution used in the simulation was measured by a fast neutron spectrometer, the 1.8-ton Multiplicity and Recoil Spectrometer (MARS) detector, at the same depth. We nd that the measured detection rate of two correlated neutrons is consistent with that predicted by simulation. The result lends additional confidence in the detection technique used by MARS, and therefore in the MARS spectra as measured at three different depths. Confirmation of the fast neutron flux and spectrum is important as it helps validate the scaling models used to predict the fast neutron fluxes at different overburdens.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Rapid neutron and gamma-ray source localization using machine learning

Rapid localization of radiation sources is critical for applications including nuclear emergency response, safeguards, and security. However, conventional imaging systems such as neutron scatter cameras and Compton cameras depend on rare coincidence events, which often result in long acquisition times. In this work, we address the challenge of rapid source localization by developing a machine learning approach to predict the direction of a single radiation source using only count rates from an array of neutron and gamma-ray detectors. The proposed model is a fully connected neural network (FCNN) trained using Monte Carlo simulation data from a 252 Cf source. The model hyperparameters are optimized with a small set of routine 252 Cf measurements. We benchmarked the performance of the trained and optimized machine learning model using additional 252 Cf , 137 Cs , and PuBe measurements under laboratory conditions with varying source-detector configurations. For these measurements, the machine learning model achieved a mean localization error smaller than 30° with 3 x 10 3 system counts, corresponding to 8 s measurement time for the imaging system used in this work. In this low-statistics regime, the method outperformed traditional scatter-based imaging by more than 75% in localization accuracy for the evaluated measurement configurations. These results demonstrate that a machine learning-based approach can significantly reduce the time required for accurate single-source localization, providing a robust and computationally efficient alternative to traditional imaging systems in time-critical nuclear security and emergency response scenarios.

Gamma-ray imaging↗