Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data inference”

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 1,045 records · Page 58
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

An ontology-based knowledge graph for representing interactions involving RNA molecules

The "RNA world" represents a novel frontier for the study of fundamental biological processes and human diseases and is paving the way for the development of new drugs tailored to each patient's biomolecular characteristics. Although scientific data about coding and non-coding RNA molecules are constantly produced and available from public repositories, they are scattered across different databases and a centralized, uniform, and semantically consistent representation of the "RNA world" is still lacking. We propose RNA-KG, a knowledge graph (KG) encompassing biological knowledge about RNAs gathered from more than 60 public databases, integrating functional relationships with genes, proteins, and chemicals and ontologically grounded biomedical concepts. To develop RNA-KG, we first identified, pre-processed, and characterized each data source; next, we built a meta-graph that provides an ontological description of the KG by representing all the bio-molecular entities and medical concepts of interest in this domain, as well as the types of interactions connecting them. Finally, we leveraged an instance-based semantically abstracted knowledge model to specify the ontological alignment according to which RNA-KG was generated. RNA-KG can be downloaded in different formats and also queried by a SPARQL endpoint. A thorough topological analysis of the resulting heterogeneous graph provides further insights into the characteristics of the "RNA world". RNA-KG can be both directly explored and visualized, and/or analyzed by applying computational methods to infer bio-medical knowledge from its heterogeneous nodes and edges. The resource can be easily updated with new experimental data, and specific views of the overall KG can be extracted according to the bio-medical problem to be studied.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Short-Term Groundwater Level Fluctuations Drive Subsurface Redox Variability

As global change processes modify the extent and functions of terrestrial-aquatic interfaces, the variability of critical and dynamic transitional zones between wetlands and uplands increases. However, it is still unclear how fluctuating water levels at these dynamic boundaries alter groundwater biogeochemical cycling. Here, we used high-temporal resolution data along gradients from wetlands to uplands and during fluctuating water levels at freshwater coastal areas to capture spatiotemporal patterns of groundwater redox potential (E h ). We observed that topography influences groundwater E h that is higher in uplands than wetlands; however, the high variability within TAI zones challenged the establishment of distinct redox zonation. Declining water levels generally decreased E h , but most locations exhibited significant E h variability, which is associated with rare instances of short-term water level fluctuations introducing oxygen. The E h -oxygen relationship showed distinct hysteresis patterns, reflecting redox poising capacity at higher E h , maintaining more oxidizing states longer than dissolved oxygen presence. Surprisingly, we observed more frequent oxidizing states in transitional areas and wetlands than in uplands. We infer that occasional oxygen entering specific wetland-upland boundaries acts as critical biogeochemical control points. Finally, high-resolution data can capture such rare yet significant biogeochemical instances, supporting redox-informed models and advancing the predictability of climate change feedback.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Kronecker-structured covariance models for multiway data

Many applications produce multiway data of exceedingly high dimension. Modeling such multi-way data is important in multichannel signal and video processing where sensors produce multi-indexed data, e.g. over spatial, frequency, and temporal dimensions. We will address the challenges of covariance representation of multiway data and review some of the progress in statistical modeling of multiway covariance over the past two decades, focusing on tensor-valued covariance models and their inference. We will illustrate through a space weather application: predicting the evolution of solar active regions over time.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

DETECTING FIRE WITH MACHINE LEARNING-ENABLED VISUAL MONITORING FOR NUCLEAR POWER PLANT ENVIRONMENTS

Nuclear power plants are experiencing significant cost challenges to remain competitive with other energy-generation utilities. Unlike other industries, the cost of operation and maintenance activities is mostly attributed to workforce costs. To mitigate this, nuclear power plant stakeholders are increasingly interested in the development and deployment of machine learning methods to potentially automate or augment manually intensive tasks to reduce costs, especially for monitoring activities. One monitoring function that is visually demanding and that can occur frequently to meet the requirements of a fire protection program is visually monitoring an area for fire occurrence. Currently, fire watch activities consist of a worker physically stationed at a given location with the sole responsibility of observing a given area to ensure a fire is detected and mitigated promptly. This effort focused on the development and evaluation of a suitable deep convolutional neural network to classify individual video frames at a sub-second frequency for the occurrence of “fire” and “no fire” in varying industrial environments similar to nuclear power plants. It is believed that a trained neural network model could be integrated with existing facility video surveillance camera feeds to generate alerts when fire inferences occur in individual frames captured at sub-second temporal resolutions. Extensive effort was dedicated to identifying and curating suitable imagery training data representing varying environments and scene settings with and without flame features to maximize generalization in nuclear power plant environments. The data collection effort resulted in the aggregation of a large, labeled image library exceeding 12,000 images to support model training for diverse industrial environments. A deep neural network model incorporating parallel multi-scale capabilities was developed and trained to support accurate image-based detection of flame incidents of varying sizes and spectral feature properties within heterogeneous scenes. Analysis results show that the trained model can achieve high inference accuracy despite heterogeneous scene environments and components. Testing accuracy exceeded 95.0 percent with very low false positive and false negative inferences.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

A measurement of stellar surface gravity hidden in radial velocity differences of comoving stars

The gravitational redshift induced by stellar surface gravity is notoriously difficult to measure for non-degenerate stars, since its amplitude is small in comparison with the typical Doppler shift induced by stellar radial velocity. In this study, we make use of the large observational data set of the Gaia mission to achieve a significant reduction of noise caused by these random stellar motions. By measuring the differences in velocities between the components of the pairs of comoving stars and wide binaries, in this work we are able to statistically measure the combined effects of gravitational redshift and convective blueshifting of spectral lines, and nullify the effect of the peculiar motions of the stars. For the subset of stars considered in this study, we find a positive correlation between the observed differences in Gaia radial velocities and the differences in surface gravity and convective blueshift inferred from effective temperature and luminosity measurements. The results rule out a null signal at the 5σ level for our full data set. Additionally, we study the subdominant effects of binary motion, and possible systematic errors in radial velocity measurements within Gaia. Results from the technique presented in this study are expected to improve significantly with data from the next Gaia data release. Such improvements could be used to constrain the mass–luminosity relation and stellar models that predict the magnitude of convective blueshift.

(stars:) binaries: general↗
DOE OSTI2020

Monte Carlo control loops for cosmic shear cosmology with DES Year 1 data

Weak lensing by large-scale structure is a powerful probe of cosmology and of the dark universe. This cosmic shear technique relies on the accurate measurement of the shapes and redshifts of background galaxies and requires precise control of systematic errors. Monte Carlo control loops (MCCL) is a forward modeling method designed to tackle this problem. It relies on the ultra fast image generator (UFig) to produce simulated images tuned to match the target data statistically, followed by calibrations and tolerance loops. Here, we present the first end-to-end application of this method, on the Dark Energy Survey (DES) Year 1 wide field imaging data. We simultaneously measure the shear power spectrum $C_ℓ$ and the redshift distribution $n(z)$ of the background galaxy sample. The method includes maps of the systematic sources, point spread function (PSF), an approximate Bayesian computation (ABC) inference of the simulation model parameters, a shear calibration scheme, and a fast method to estimate the covariance matrix. We find a close statistical agreement between the simulations and the DES Y1 data using an array of diagnostics. In a nontomographic setting, we derive a set of $C_ℓ$ and $n(z)$ curves that encode the cosmic shear measurement, as well as the systematic uncertainty. Following a blinding scheme, we measure the combination of $Ω_m$, $σ_8$, and intrinsic alignment amplitude $A_{IA}$, defined as $S_8D_{IA}=σ_8(Ω_m/0.3)^{0.5}D_{IA}$, where $D_{IA}=1-0.11(A_{IA}-1)$. We find $S_8D_{IA}=0.8954_{-0.039}^{+0.054}$, where systematics are at the level of roughly 60% of the statistical errors. We discuss these results in the context of earlier cosmic shear analyses of the DES Y1 data. Our findings indicate that this method and its fast runtime offer good prospects for cosmic shear measurements with future wide-field surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

A Framework for Neural Network Inference on FPGA-Centric SmartNICs

FPGA-based SmartNICs offer great potential to significantly improve the performance of high-performance computing and warehouse data processing by tightly coupling support for reconfigurable data-intensive computation with cross-node communication, thereby mitigating the von Neumann bottleneck. Existing work, however, has been generally been limited in that it assumes an accelerator model where kernels are offloaded to SmartNICs, but most control tasks are left to the CPUs. This leads to frequent waiting, inferior performance, and scaling challenges. In this work, we propose a new distributive data-centric computing framework, named FCsN, for reconfigurable SmartNIC-based systems. Through a lightweight task circulation execution model and its implementation architecture, FCsN allows the complete detaching of kernel execution, control logic, system scheduling, and network communication to the SmarNICs. This boosts performance by: (i) avoiding the control dependency with CPUs and (ii) supporting streaming kernel execution and network communication at line rate and in a very fine-grained manner. We demonstrate the efficiency and flexibility of FCsN using various types of neural network applications including graph neural networks; as these last are both irregular and data intensive they offer an especially robust demonstration. Evaluations using commonly-used neural network models and graph datasets show that a system with the support of FCsN can achieve, on average, 144 speedups over the MPI-based standard CPU baselines.

Guo, Anqi↗
DOE OSTI2024

Machine Learning Inference of Random Medium Properties

Earth materials are heterogeneous across a range of spatial scales, but the resolvability of small structures is limited by sparse data coverage, noise, bandlimitedness, and other difficulties. In practice, heterogeneities below a certain size cannot be recovered from seismic data except through statistical medium descriptions, which even then can be difficult to uniquely determine. To improve the characterization of such heterogeneities, we develop a novel supervised machine learning (ML) model that provides insight about the recoverability of statistical medium properties from elastic waveform data and succeeds despite cycle-skipping and other challenges well known from elastic waveform inversion. We demonstrate the approach using random media generated by superimposing self-affine random variations on homogeneous and layered background structures. After training on sparsely-recorded, high-frequency waveforms from hundreds of different random medium realizations, we show the ability of our ML model to recover correlation lengths and other statistical properties of interest to near-surface and crustal seismology, among other fields. For frequency passbands and spatial offsets encountered in seismology, Gaussian correlation lengths and the amplitude of the random variations relative to the background model are recovered even in challenging scenarios involving unknown medium parameters, complex crustal structures, and low signal-to-noise ratio. In comparison, von Kármán correlation lengths, which are related to larger-wavelength variations of the medium than Gaussian correlation lengths, are not as well recovered. These results provide one of the first and most systematic investigations of the recoverability of statistical properties of heterogeneities below the resolution limit of deterministic seismic tomography, and suggest practical ML strategies for high-frequency waveform seismology.

58 GEOSCIENCES↗
DOE OSTI2020

Tomographic Reconstruction of the Neutron Time-Energy Spectrum from a Dense Plasma Focus

A dense plasma focus (DPF) nuclear fusion device is an attractive pulsed neutron source in many applications, due to its relatively large neutron yield, produced in a short time duration. In order to design a DPF that generates neutrons within a specified time profile, or generates a neutron energy spectrum with specific properties, it is necessary to be able to characterize and model the results of the fusion process releasing the neutrons. The time-energy spectrum of fusion neutrons is an ideal quantity to use to validate multiphysics codes that simulate the pinch and fusion processes, because it is a quantity that requires the physics of each of the stages leading up to and ending in the fusion reaction to be simulated correctly. In particular, since DPF fusion neutrons are not monoenergetic -- and there can often be several fusion pinches creating neutrons -- a computer simulation matching high quality neutron spectrum measurements provides great confidence in the fidelity of the simulation. In order to make such a comparison, it is first necessary to have quality measurements from which to infer the spectrum. In this work we pose neutron spectroscopy as the classical tomographic inverse problem from neutron time of flight data at multiple distances, but enhanced by using an additional measurement of the time profile of the fusion pinch near its source and by using detector pairs set up in a geometry that allows for scatter background subtraction. The detector pairs enhance the quality of the time of flight measurements, and the additional constraint posed by the measured time profile allows for reconstructions discretized as finely as the time measurements and in energy as finely as 100 keV, without the problem being underdetermined. We present results from a Deuterium-fueled DPF at the U.S. Department of Energy's Nevada National Security Site and show that we can infer the time-energy spectrum from our measurements for both single and multi-pinch fusion reactions with equal fidelity.

Catenacci, Jared↗
DOE OSTI2024

MultiPEM Toolbox: User Manual [Rev. 2]

This document explains use of the Multi-Phenomenology Explosion Monitoring (Multi PEM) Toolbox, a collection of R scripts for estimating the unknown device parameters of a new event with uncertainty quantification. The methodology and application used for illustration in this user manual are fully documented in a Los Alamos National Laboratory technical report hereafter designated “WPA” for reference. Additional details on the application are found in a recent journal article. Two assessment types are available: rapid and complete. Rapid assessments are conducted in two stages, as described in Section 2. In the first stage, calibration data are used to estimate forward and error model parameters (WPA, §5.1) and (if relevant) errors-in-variables yield values for calibration sources (WPA, §3, Equation (3)). In the second stage, new event data are used to estimate the unknown new event device parameters (WPA, §5.2) with uncertainty quantification. Two options for treating the inferred first stage parameters in second stage Bayesian analysis are available: fixing them at their maximum likelihood estimate (default), or multiple imputation. Multiple imputation involves utilizing several posterior samples (imputations) of the first stage parameters as fixed values in the second stage posterior sampling of the new event device parameters. Second stage sampling is conducted across imputations in parallel to improve computational efficiency. This method produces improved uncertainty quantification of the new event device parameters compared with the default treatment of the first stage parameters, at the expense of additional computation. Complete assessments are conducted in a single stage, as described in Section 3. Calibration and (if relevant) new event data are used simultaneously to estimate all forward model, error model, and (if relevant) new event device parameters with uncertainty quantification on the latter. As the name suggests, rapid assessments generally run substantially faster than complete assessments (even with multiple imputation), because the results of first stage analysis can be stored and incorporated into estimating a relatively low-dimensional space of new event device parameters whenever relevant new event data becomes available. On the other hand, complete assessments must be run on the full set of model and device parameters with calibration and new event data every time the latter becomes available.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Deep learning-based spatio-temporal estimate of greenhouse gas emissions using satellite data

Accurate estimation of greenhouse gases (GHGs) emissions is very important for developing mitigation strategies to climate change by controlling and reducing GHG emissions. This project aims to develop multiple deep learning approaches to estimate anthropogenic greenhouse gas emissions using multiple types of satellite data. NO2 concentration is chosen as an example of GHGs to evaluate the proposed approach. Two sentinel satellites (sentinel-2 and sentinel-5P) provide multiscale observations of GHGs from 10-60m resolution (sentinel-2) to ~kilometer scale resolution (sentinel-5P). Among multiple deep learning (DL) architectures evaluated, two best DL models demonstrate that key features of spatio-temporal satellite data and additional information (e.g., observation times and/or coordinates of ground stations) can be extracted using convolutional neural networks and feed forward neural networks, respectively. In particular, irregular time series data from different NO 2 observation stations limit the flexibility of long short-term memory architecture, requiring zero-padding to fill in missing data. However, deep neural operator (DNO) architecture can stack time-series data as input, providing the flexibility of input structure without zero-padding. As a result, the DNO outperformed other deep learning architectures to account for time-varying features. Overall, temporal patterns with smooth seasonal variations were predicted very well, while frequent fluctuation patterns were not predicted well. In addition, uncertainty quantification using conformal inference method is performed to account for prediction ranges. Overall, this research will lead to a new groundwork for estimating greenhouse gas concentrations using multiple satellite data to enhance our capability of tracking the cause of climate change and developing mitigation strategies.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

FLYING SERVING: On-the-Fly Parallelism Switching for Large Language Model Serving

Production LLM serving must simultaneously deliver high throughput, low latency, and sufficient context capacity under non-stationary traffic and mixed request requirements. Data parallelism (DP) maximizes throughput by running independent replicas, while tensor parallelism (TP) reduces per-request latency and pools memory for long-context inference. However, existing serving stacks typically commit to a static parallelism configuration at deployment; adapting to bursts, priorities, or long-context requests is often disruptive and slow. We present Flying Serving, a vLLM-based system that enables online DP-TP switching without restarting engine workers. Flying Serving makes reconfiguration practical by virtualizing the state that would otherwise force data movement: (i) a zero-copy Model Weights Manager that exposes TP shard views on demand, (ii) a KV Cache Adaptor that preserves request KV state across DP/TP layouts, (iii) an eagerly initialized Communicator Pool to amortize collective setup, and (iv) a deadlock-free scheduler that coordinates safe transitions under execution skew. Across three popular LLMs and realistic serving scenarios, Flying Serving improves performance by up to 4.79 × under high load and 3.47 × under low load while supporting latency- and memory-driven requests.

Gao, Shouwei [ORNL]↗
DOE OSTI2023

Joint Analysis of the Epidemic Evolution and Human Mobility During the First Wave of COVID-19 in Spain: Retrospective Study

Background The initial wave of the COVID-19 pandemic placed a tremendous strain on health care systems worldwide. To mitigate the spread of the virus, many countries implemented stringent nonpharmaceutical interventions (NPIs), which significantly altered human behavior both before and after their enactment. Despite these efforts, a precise assessment of the impact and efficacy of these NPIs, as well as the extent of human behavioral changes, remained elusive. Objective In this study, we conducted a retrospective analysis of the initial wave of COVID-19 in Spain to better comprehend the influence of NPIs and their interaction with human behavior. Such investigations are vital for devising future mitigation strategies to combat COVID-19 and enhance epidemic preparedness more broadly. Methods We used a combination of national and regional retrospective analyses of pandemic incidence alongside large-scale mobility data to assess the impact and timing of government-implemented NPIs in combating COVID-19. Additionally, we compared these findings with a model-based inference of hospitalizations and fatalities. This model-based approach enabled us to construct counterfactual scenarios that gauged the consequences of delayed initiation of epidemic response measures. Results Our analysis demonstrated that the pre–national lockdown epidemic response, encompassing regional measures and heightened individual awareness, significantly contributed to reducing the disease burden in Spain. The mobility data indicated that people adjusted their behavior in response to the regional epidemiological situation before the nationwide lockdown was implemented. Counterfactual scenarios suggested that without this early epidemic response, there would have been an estimated 45,400 (95% CI 37,400-58,000) fatalities and 182,600 (95% CI 150,400-233,800) hospitalizations compared to the reported figures of 27,800 fatalities and 107,600 hospitalizations, respectively. Conclusions Our findings underscore the significance of self-implemented prevention measures by the population and regional NPIs before the national lockdown in Spain. The study also emphasizes the necessity for prompt and precise data quantification prior to enacting enforced measures. This highlights the critical interplay between NPIs, epidemic progression, and human behavior. This interdependence presents a challenge in predicting the impact of NPIs before they are implemented.

Public, Environmental & Occupational Health↗
DOE OSTI2026

A Bayesian inferencing framework for ultrasound wave speed measurements in metal additive manufacturing

Process-related changes during metal additive manufacturing introduce microstructural variability in the material properties of printed parts, directly affecting component reliability. Accurate estimation of these property variations with part performance are essential for quality assurance. Ultrasound testing offers a non-destructive means to estimate mechanical properties and detect defects; however, conventional analysis methods often neglect the influence of microstructural variability, limiting their effectiveness. Here, this research presents a Bayesian inference technique for quantifying wave speed uncertainty from ultrasound measurements of metal additive manufactured parts. By integrating prior ultrasound data with a Bayesian model, the proposed approach generates posterior density estimates of wave speed that systematically account for manufacturing-induced variability and uncertainty. The novelty of this research lies in applying a Bayesian framework to analyze experimental ultrasound measurements within the context of metal additive manufacturing variability. The method enhances the accuracy of wave speed estimation by 64%, defect position by 50% and increases confidence associated with wave speed variance by 30% across different porosity levels, thereby providing a robust foundation for improved decision-making and increased reliability in additively manufactured components.

Additive manufacturing↗
DOE OSTI2021

Experimental inference of neutral and impurity transport in Alcator C-Mod using high-resolution x-ray and ultra-violet spectra

We present experimental inferences of cross-field impurity transport coefficients for AlcatorC-Mod plasmas without edge-localized modes, leveraging a novel forward model for the entire Ca Kα spectrum, including satellite lines within the spectral range, to compare to high-resolution x-ray imaging crystal spectroscopy (XICS). These measurements are complemented by extreme ultra-violet (EUV) spectroscopy that constrains transport closer to the edge. Using new atomic data sets for both XICS and EUV analysis has enabled consideration of line ratios across both spectral ranges and has increased the accuracy of inferred transport coefficients. Inclusion of charge exchange between edge thermal neutrals and impurities is shown to be extremely important in C-Mod pedestals. Atomic D neutral densities from experimental D Lyα measurements at the midplane are compared to SOLPS-ITER simulations, finding good agreement. Bayesian inferences of impurity transport coefficients are presented for L-, EDA H-, and I-mode discharges, making use of the Aurora package for forward modeling and combining our spectroscopic constraints. Experimentally inferred diffusion profiles are found to match turbulent transport models at midradius within uncertainties, using both quasilinear gyro-fluid TGLF SAT-1 and nonlinear ion-scale gyrokinetic CGYRO simulations. Furthermore, significant discrepancies in convection are observed in some cases, suggesting difficulties in predictions of flat or hollow impurity profiles.

neutrals↗
DOE OSTI2025

Forecasting constraints on the high-z IGM thermal state from the Lyman-α forest flux autocorrelation function

ABSTRACT The autocorrelation function of the Lyman-$\alpha$ (Ly $\alpha$) forest flux from high-z quasars probes the small-scale structure of the intergalactic medium (IGM). The thermal state of the IGM, determined by the physics of reionization, sets the small-scale power observed in the Ly $\alpha$ forest. To explore the sensitivity of the autocorrelation function to the IGM’s thermal state, we compute the autocorrelation function from a cosmological hydrodynamical simulation with an instantaneous reionization model and 135 post-processed thermal states. Using mock data sets of 20 quasars, we forecast constraints on $T_0$ and $\gamma$, which characterize the post-processed IGM thermal state, at $5.4 \le z \le 6$. While this model simplifies the IGM’s thermal state, it serves as a key first step in assessing future observational prospects. We also perform an inference test on mocks and re-weight out posterior distributions to guarantee that they exhibit statistically correct behaviour. At $z = 5.4$, we find that an idealized data set constrains $T_0$ to 59 per cent and $\gamma$ to 16 per cent at the 1$\sigma$ equivalent confidence level. To explore more realistic, non-instantaneous reionization scenarios, we analyse four models combining temperature and ultraviolet background (UVB) fluctuations at $z = 5.8$. We find that mock data generated from a model with both temperature and UVB fluctuations can rule out a model with only temperature fluctuations at the $> 1\sigma$ level 73.9 per cent of the time.

Wolfson, Molly↗
DOE OSTI2022

Deep-learning-guided high-resolution subsurface reflectivity imaging with application to ground-penetrating radar data

Subsurface reflectivity imaging is one of the most important geophysical characterization methods for revealing subsurface structures. In many cases, accurate subsurface reflectivity imaging is challenging because of, for example, random or coherent noise in the data and sparse source-receiver observation geometry. Here, we develop a deep-learning-guided iterative imaging method to improve subsurface structure imaging. Specifically, we train a supervised neural network to infer a noise-free, high-resolution image from a noisy, low-resolution image and use this estimated image as guidance to regularize least-squares imaging. We develop a systematic method to generate high-quality synthetic training data (data-label pairs) to train the guidance neural network. The trained neural network can provide high-fidelity predictions even for field-data images that are not in the training data. We validate our new imaging method using one synthetic and two field ground-penetrating radar data examples, and find that our method can produce clean, high-resolution subsurface reflectivity images where existing single-pass and least-squares imaging methods fail due to noise and insufficient data coverage.

58 GEOSCIENCES↗
DOE OSTI2025

Direct NeTS sampling of nuclear graphite $S(α, β, T)$ in Serpent

For advanced reactor applications, Neural Thermal Scattering (NeTS) modules were developed to predict the thermal scattering law (TSL or $S(α, β, T)$) of a nuclear graphite neutron moderator. NeTS are multi-layer, feedforward artificial neural networks, which act as universal function approximators designed for TSL datasets. In this case, a 4-layer neural network with 164 neurons per layer is trained using FLASSH evaluated data in PyTorch and serialized as a torchscript dictionary to predict $S(α, β, T)$ on-the-fly. Relative, absolute and maximum percent deviations of NeTS from File 7 data generated using the FLASSH code are on the order of 0.01%, 0.1% and 1%, respectively, with low inference latencies of 0.000172 s per $S(α, β, T)$ at a given temperature. Capturing the full dimensionality of possible inelastic neutron-lattice interactions, NeTS functionality is embedded in the Serpent Monte Carlo code, where $S(α, β, T)_{NeTS}$ sampling is conducted on-the-fly and compared to ACE look-up-tables for predicting TREAT criticality. k-eff differences between sampling algorithms of 6 pcm are observed and are within the order of Monte Carlo uncertainty. Compared to discrete and continuous-energy ACE files (30 MB and 131 MB per temperature), the NeTS format is on the order of 200–300 kB for a continuous-temperature, interpolation-free representation of $S(α, β, T)$ and cross sections. NeTS-in-Serpent runtimes comparable with ACE look-up tables are achieved by scaling NeTS for high performance computing architectures with hybrid OpenMP + MPI parallelization. This work validates a novel, self-contained reactor physics framework for predictive cross sections, and demonstrates a general methodology for embedding modern machine learning libraries within existing neutronic analysis frameworks.

Nuclear Criticality Safety Program (NCSP)↗