Search NASA⌕ Search

SEARCH · Search NASA

Results for “Statistical Learning”

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 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Explainable tokamak-agnostic forecasting of fusion plasma instability via megahertz turbulent fluctuations

Scientific applications of artificial intelligence (AI) often remain limited by device-specific training and unexplained “black-box” approaches, creating fundamental barriers to cross-system generalization. This challenge is critical for nuclear fusion, where future reactors will have limited operational data for AI training. Here, we demonstrate that our neural network, trained solely on megahertz-scale turbulence measurements from one machine (DIII-D), forecasts Type-I edge localized mode (ELM) onsets in a different tokamak (KSTAR) through zero-shot weight transfer following physics-consistent preprocessing without device-specific retraining. Through an explainable AI framework combining gradient-weighted class activation mapping with physics validation, we reveal that our network can internalize physics relationships governing the ELM instabilities rather than memorizing device-specific patterns. The network perceives spatiotemporal features that correlate consistently with independently calculated instability growth rates, magnetohydrodynamic stability limits, and pedestal structure dynamics. Statistical analyses of dimensionally-reduced saliency features reveal the identical triangular features between the saliency representations, instability growth rates, and prediction probability across tokamaks, providing evidence that our forecasting system can show tokamak-agnostic generalization. This work contributes to a foundation for explainable scientific AI systems, where cross-system developments are essential for transcending traditional domain-specific constraints.

AI↗
DOE OSTI2025

TPCpp-10M: Simulated proton-proton collisions in a time projection chamber for AI foundation models

Scientific foundation models hold great promise for advancing nuclear and particle physics by improving analysis precision and accelerating discovery. Yet, progress in this field is often limited by the lack of openly available large scale datasets, as well as standardized evaluation tasks and metrics. Furthermore, the specialized knowledge and software typically required to process particle physics data pose significant barriers to interdisciplinary collaboration with the broader machine learning community. This work introduces a large, openly accessible dataset of 10 million simulated proton-proton collisions, designed to support self-supervised training of foundation models. To facilitate ease of use, the dataset is provided in a common NumPy format. In addition, it includes 70,000 labeled examples spanning three well defined downstream tasks: track finding, particle identification, and noise tagging, to enable systematic evaluation of the foundation model's adaptability. The simulated data are generated using the Pythia Monte Carlo event generator at a center of mass energy of $\sqrt{s}$ = 200 GeV and processed with Geant4 to include realistic detector conditions and signal emulation in the sPHENIX Time Projection Chamber at the Relativistic Heavy Ion Collider, located at Brookhaven National Laboratory. This dataset resource establishes a common ground for interdisciplinary research, enabling machine learning scientists and physicists alike to explore scaling behaviors, assess transferability, and accelerate progress toward foundation models in nuclear and high energy physics. The complete simulation and reconstruction chain is reproducible with the sPHENIX software stack. All data and code locations are provided under Data Accessibility.

Data Analysis, Statistics and Probability (physics↗
DOE OSTI2024

Risk-Aware Framework Development for Disruption Prediction: Alcator C-Mod and DIII-D Survival Analysis

Abstract Survival regression models can achieve longer warning times at similar receiver operating characteristic performance than previously investigated models. Survival regression models are also shown to predict the time until a disruption will occur with lower error than other predictors. Time-to-event predictions from time-series data can be obtained with a survival analysis statistical framework, and there have been many tools developed for this task which we aim to apply to disruption prediction. Using the open-source Auton-Survival package we have implemented disruption predictors with the survival regression models Cox Proportional Hazards, Deep Cox Proportional Hazards, and Deep Survival Machines. To compare with previous work, we also include predictors using a Random Forest binary classifier, and a conditional Kaplan-Meier formalism. We benchmarked the performance of these five predictors using experimental data from the Alcator C-Mod and DIII-D tokamaks by simulating alarms on each individual shot. We find that developing machine-relevant metrics to evaluate models is an important area for future work. While this study finds cases where disruptive conditions are not predicted, there are instances where the desired outcome is produced. Giving the plasma control system the expected time-to-disruption will allow it to determine the optimal actuator response in real time to minimize risk of damage to the device.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Efficient and generalizable nested Fourier-DeepONet for three-dimensional geological carbon sequestration

Geological carbon sequestration (GCS) involves injecting CO2 into subsurface geological formationsfor permanent storage. Numerical simulations could guide decisions in GCS projects by predictingCO 2 migration pathways and the pressure distribution in storage formation. However, these simula-tions are often computationally expensive due to highly coupled physics and large spatial-temporalsimulation domains. Surrogate modelling with data-driven machine learning has become a promis-ing alternative to accelerate physics-based simulations. Among these, the Fourier neural operator(FNO) has been applied to three-dimensional synthetic subsurface models. Despite its good accuracyin simulating CO 2 plume migration, it requires large computational resources in training and alsolacks generalizability. Here, to further improve performance, we have developed a nested Fourier-DeepONet by combining the expressiveness of the FNO with the modularity of a deep operatornetwork (DeepONet). This new framework is twice as efficient as a nested FNO for training and has atleast 80% lower GPU memory requirement due to its flexibility to treat temporal coordinates sepa-rately. These performance improvements are achieved without compromising prediction accuracy.In addition, the generalization and extrapolation ability of nested Fourier-DeepONet beyond thetraining range has been thoroughly evaluated. Nested Fourier-DeepONet outperformed the nestedFNO for extrapolation in time with more than 50% reduced error. It also exhibited good extrapolationaccuracy beyond the training range in terms of reservoir properties, number of wells, and injectionrate.

Lee, Jonathan E. [Department of Chemical and Envir↗
NASA NTRSDate not supplied

Spaceflight Associated Neuro-Ocular Syndrome: A Brief for Astronaut Candidates

INTRODUCTION: Spaceflight Associated Neuro-ocular Syndrome (SANS) was first described in 2011 as a clinical process involving degradation of visual acuity and structural changes of the eye when exposed to microgravity. This syndrome poses a risk to astronaut vision, safety, and mission performance, especially during extended-duration missions. The modern astronaut cohort is a diverse group of men and women including scientists, engineers, teachers, pilots, business professionals, and artists. Given their diverse backgrounds, it is imperative all crews are educated on SANS risks and potential countermeasures. This education must be comprehensive and easily understood, regardless of academic background or specific expertise. TOPIC: A PowerPoint was created to serve as a presentation and stand-alone educational aid for all astronaut candidates (ASCANS). This medium was chosen as it can be updated dynamically as new information emerges. The brief was split into sections to facilitate organized delivery of complex content in easy-to-understand components. Sections included descriptive statistics, signs and symptoms, outcomes, myths and ongoing research, countermeasures, ongoing surveillance, and unknown and future directions. More research is necessary in order to understand this syndrome, and without continued ASCAN participation in research, acquisition of data will not be possible. These efforts will become increasingly critical with deep space and exploration missions. APPLICATION: We recommend regular use of this brief for all ASCANS, and potentially veteran crewmembers, given the high risks and potentially poor outcomes associated with SANS. For ASCANS, recommendations include that this brief be provided at the start of their training, with a bi-annual refresher course for all astronauts. This ensures they are current on SANS information, given the dynamic nature of this syndrome. Additionally, this brief will be made available to all as a self-study aid, utilized as a refresher immediately prior to launch. It is anticipated that this brief will promote the health, safety, and wellbeing of professional and private crews alike, during and after long-duration spaceflight. RESOURCES: Brunstetter, T., Tarver, B., & Tsung, A. (2022, October). Spaceflight Associated Neuro-ocular Syndrome (Sans) and its Risk to Nasa Astronauts. JSC Aerospace Medicine Clerkship. Houston, TX; Lyndon B. Johnson Space Center. Laurie, S. S., Macias, B. R., Pardon, L. P., Brunstetter, T., Tarver, W. J., Gibson, C. R., Greenwald, S. H., Jasien, K. V., Mason, S., & Tsung, A. (2022). (rep.). Evidence Report: Risk of Spaceflight Associated Neuro-ocular Syndrome (SANS). Houston, TX: National Aeronautics and Space Administration. Mader, T. H., Gibson, C. R., Pass, A. F., Kramer, L. A., Lee, A. G., Fogarty, J., ... & Polk, J. D. (2011). Optic disc edema, globe flattening, choroidal folds, and hyperopic shifts observed in astronauts after long-duration space flight. Ophthalmology, 118(10), 2058-2069. LEARNING OBJECTIVES: • The audience will be able to describe current Spaceflight Associated Neuro-ocular Syndrome (SANS) findings, significance, and countermeasures. • The audience will understand the importance of astronauts’ familiarity with SANS prior to flying and the large risks associated with this syndrome. CME QUESTIONS (3 QUESTIONS, MULTIPLE CHOICE A-D): 1. What is the approximate prevalence rate of SANS in astronauts? A. 22% B. 50% C. 66% D. 80% Answer: C. 2. Which of the ocular findings are NOT associated with SANS? A. Optic disc edema B. Retinal detachment C. Globe flattening D. Chorioretinal folds Answer: B. 3. Which of the following diagnostic technologies are NOT available in-flight? A. Ocular ultrasound B. Retinal photography C. MRI Orbit D. Optical Coherence Tomography (OCT) Answer: C.

Sara Stroble Mason↗
DOE OSTI2026

Unsupervised anomaly detection in MeV ultrafast electron diffraction

MeV ultrafast electron diffraction (MUED) is a pump-probe technique used to study the dynamic structural evolution of materials. An ultrashort laser pulse triggers structural changes, which are then probed by an ultrashort relativistic electron beam. To overcome low signal-to-noise ratios, diffraction patterns are averaged over thousands of shots. However, shot-to-shot instabilities in the electron beam can distort individual patterns, introducing uncertainty. Improving MUED accuracy requires detecting and removing these anomalous patterns from large datasets. In this work, we developed a fully unsupervised methodology for the detection of anomalous diffraction patterns. Using a convolutional autoencoder, we calculate the reconstruction mean squared error of the diffraction patterns. Based on the statistical analysis of this error, we provide the user an estimation of the probability that the pattern is normal, which also allows a posterior visual inspection of the images that are difficult to classify. This method has been trained with only 100 diffraction patterns and tested on 1521 patterns, resulting in a false positive rate between 0.2% and 0.4%, with a training time of 10 s per image and a test time of about 1 s per image. Here, the proposed methodology can also be applied to other diffraction techniques in which large datasets are collected that include faulty images due to instrumental instabilities.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

Ripening of Rh Nanoparticle Catalysts in Reverse Water–Gas Shift via a Data-Driven Model Combining Physics, Theory, and Experiment

Degradation via sintering is an ongoing challenge that impedes the broad commercial success of supported metallic nanoparticle catalysts. To mitigate degradation via informed catalyst design and process operations, here we aim to disambiguate the underlying mechanisms of sintering by combining theory and experiment in a quantitative framework. While mechanistic sintering models exist, they only model a single sintering pathway, even though multiple sintering mechanisms can occur simultaneously or dominate at different stages of the process. Data-driven machine learning models have emerged as a means to represent complex processes through data regression. However, machine learning models have very large data needs and lack mechanistic insights due to their black-box encoding. To develop an interpretive model of catalyst degradation via sintering, we constructed a hybrid model combining mechanistic “physics-based” models and data-driven methods to obtain both reliable predictions and mechanistic insights regarding experimentally observed sintering phenomena. Focusing on nanoparticle sintering in the Rh–TiO 2 catalyst for the reverse water–gas shift (RWGS) reaction, the hybrid model couples a mechanistic term for Ostwald ripening with energy values calculated via density functional theory (DFT) with a parametric, data-driven discrepancy function term for unmodeled mechanisms. The hybrid model is trained using Bayesian inference with data collected from small-angle X-ray scattering (SAXS) in situ experiments wherein average nanoparticle diameter versus time was measured at three relevant operating temperatures. The calibrated hybrid model results show that an Ostwald ripening-only model parameterized with fixed DFT energies does not fully capture the time and temperature dependence of the SAXS-observed sintering kinetics, and that an additional functional contribution, or DFT energy calibration, is required to reconcile simulation and experiment. Analysis of the hybrid-model error confirms that the hybrid model outperforms both the purely mechanistic and purely data-driven alternatives in terms of expected predictive accuracy for time-evolving average particle sizes. Furthermore, the results support the hypothesis that the Ostwald ripening mechanism is less important for explaining the sintering phenomena as operating temperature increases under an assumed fixed DFT parameterization. This could be explained in one of two ways: either latent, unmodeled sintering mechanisms dominate at higher temperatures, or the DFT uncertainty increases with temperature. The proposed modeling approach directly links theory to experiments and simulations via a statistical hybrid modeling framework and can be extended to other catalytic systems to improve predictive models and mechanistic understanding.

Bayesian hybrid modeling↗
DOE OSTI2025

AI for nuclear physics: the EXCLAIM project

An overview of the recent activity of the newly funded EXCLusives with AI and Machine learning (EXCLAIM) collaboration is presented. The main goal of the collaboration is to develop a framework to implement AI and machine learning techniques in problems emerging from the phenomenology of high energy exclusive scattering processes from nucleons and nuclei, maximizing the information that can be extracted from various sets of experimental data, while implementing theoretical constraints from lattice QCD. A specific perspective embraced by EXCLAIM is to use the methods of theoretical physics to understand the working of ML, beyond its standardized applications to physics analyses which most often rely on industrially provided tools, in an automated way.

Analysis and statistical methods↗
DOE OSTI2025

Resimulation-based self-supervised learning for pretraining physics foundation models

Self-supervised learning (SSL) is at the core of training modern large machine learning models, providing a scheme for learning powerful representations that can be used in a variety of downstream tasks. However, SSL strategies must be adapted to the type of training data and downstream tasks required. We propose resimulation-based self-supervised representation learning (RS3L), a novel simulation-based SSL strategy that employs a method of resimulation to drive data augmentation for contrastive learning in the physical sciences, particularly, in fields that rely on stochastic simulators. By intervening in the middle of the simulation process and rerunning simulation components downstream of the intervention, we generate multiple realizations of an event, thus producing a set of augmentations covering all physics-driven variations available in the simulator. Using experiments from high-energy physics, we explore how this strategy may enable the development of a foundation model; we show how RS3L pretraining enables powerful performance in downstream tasks such as discrimination of a variety of objects and uncertainty mitigation. In addition to our results, we make the RS3L dataset publicly available for further studies on how to improve SSL strategies.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Tools for unbinned unfolding

Machine learning has enabled differential cross section measurements that are not discretized. Going beyond the traditional histogram-based paradigm, these unbinned unfolding methods are rapidly being integrated into experimental workflows. Here, in order to enable widespread adaptation and standardization, we develop methods, benchmarks, and software for unbinned unfolding. For methodology, we demonstrate the utility of boosted decision trees for unfolding with a relatively small number of high-level features. This complements state-of-the-art deep learning models capable of unfolding the full phase space. To benchmark unbinned unfolding methods, we develop an extension of existing dataset to include acceptance effects, a necessary challenge for real measurements. Additionally, we directly compare binned and unbinned methods using discretized inputs for the latter in order to control for the binning itself. Lastly, we have assembled two software packages for the OmniFold unbinned unfolding method that should serve as the starting point for any future analyses using this technique. One package is based on the widely-used RooUnfold framework and the other is a standalone package available through the Python Package Index (PyPI).

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Exploring Geothermal Potential of Great Basin Sub-Regions

The INnovative Geothermal Exploration through Novel Investigations Of Undiscovered Systems (INGENIOUS) project aims to discover new, economically viable hidden geothermal systems in the Great Basin region by building on previous work in play fairway analysis and machine learning. A key objective of this project is to develop an exploration workflow to reduce geothermal exploration risks for hidden geothermal systems. A single preliminary play fairway workflow was developed from the assessment of the regional INGENIOUS geological, geophysical, and geochemical datasets. This workflow provided new preliminary predictive geothermal fairway maps for the INGENIOUS study area, which encompasses most of Nevada, western Utah, southern Idaho, southeastern Oregon, and easternmost California. However, a recent study (incorporating machine learning techniques) of a portion of Nevada identified four geologic domains and determined that the relative importance of individual datasets or features as indicators of geothermal potential may differ across these domains. The INGENIOUS study area includes a much larger and more geologically diverse region; therefore, additional geologic domains or sub-regions are expected. To assess the sub-regions in the INGENIOUS study area, principal component analysis and k-means clustering were applied. Preliminary results indicate that the INGENIOUS regional data cluster into groups that relate to different geologic domains in the Great Basin region. These include domains such as the Walker Lane, extensional western Great Basin region, broad lower strain region in the eastern Great Basin of western Utah and eastern Nevada, Quaternary volcanic fields, and the area adjacent to the Snake River Plain. These clusters are assessed to determine the key geologic drivers of the identified clusters. Understanding this variability can provide key insights for the exploration and characterization of hidden geothermal systems in the Great Basin region and could indicate the need to develop multiple geothermal conceptual models and play fairway workflows for the INGENIOUS study area.

exploration↗
DOE OSTI2025

The Power of Many: An Ensemble Approach to Spectral Similarity

Quantifying the similarity between two mass spectra─a known reference mass spectrum and an unidentified sample mass spectrum─is at the heart of compound identification workflows in gas chromatography–mass spectrometry (GC-MS). The reference spectrum most like the sample is assigned as its identification (provided some quantitative similarity threshold is met, e.g., 80%) and thus accurately measuring similarity is essential. Significant research has gone toward developing metrics for this purpose, each of which has attempted to improve upon existing methods by incorporating GC-MS-specific information (e.g., peak ratios or retention times) or adopting various statistical and algorithmic frameworks. While this active development has led to a plethora of similarity metrics with demonstrated value across different contexts, the unfortunate consequence has been confusion surrounding which metric should be used as a global standard. No such metric is currently accepted as the standard method because different metrics have demonstrated optimal performance in different contexts. In this work, we propose an ensemble approach to spectral similarity scoring that combines the collective information from across existing similarity metrics to form an improved, globally representative similarity metric as a step toward establishing a global standard method. In conclusion, the resulting ensemble metrics are evaluated on over 88,000 spectra of varying complexity and demonstrate improved abilities to accurately rank the correct reference spectrum as the top-matching candidate for a sample relative to the rankings generated by individual similarity scores.

Carbohydrates↗
DOE OSTI2024

GrainGNN: A dynamic graph neural network for predicting 3D grain microstructure

We propose GrainGNN, a surrogate model for the evolution of polycrystalline grain structure under rapid solidification conditions in metal additive manufacturing. High fidelity simulations of solidification microstructures are typically performed using multicomponent partial differential equations (PDEs) with moving interfaces. The inherent randomness of the PDE initial conditions (grain seeds) necessitates ensemble simulations to predict microstructure statistics, e.g., grain size, aspect ratio, and crystallographic orientation. Here, currently such ensemble simulations are prohibitively expensive and surrogates are necessary.In GrainGNN, we use a dynamic graph to represent interface motion and topological changes due to grain coarsening. We use a reduced representation of the microstructure using hand-crafted features; we combine pattern finding and altering graph algorithms with two neural networks, a classifier (for topological changes) and a regressor (for interface motion). Both networks have an encoder-decoder architecture; the encoder has a multi-layer transformer long-short-term-memory architecture; the decoder is a single layer perceptron.We evaluate GrainGNN by comparing it to high-fidelity phase field simulations for in-distribution and out-of-distribution grain configurations for solidification under laser power bed fusion conditions. GrainGNN results in 80%–90% pointwise accuracy; and nearly identical distributions of scalar quantities of interest (QoI) between phase field and GrainGNN simulations compared using Kolmogorov-Smirnov test. GrainGNN's inference speedup (PyTorch on single x86 CPU) over a high-fidelity phase field simulation (CUDA on a single NVIDIA A100 GPU) is 150×–2000× for 100-initial grain problem. Further, using GrainGNN, we model the formation of 11,600 grains in 220 seconds on a single CPU core.

36 MATERIALS SCIENCE↗
NASA NTRS2016

The IPAC Image Subtraction and Discovery Pipeline for the Intermediate Palomar Transient Factory

We describe the near real-time transient-source discovery engine for the intermediate Palomar Transient Factory (iPTF), currently in operations at the Infrared Processing and Analysis Center (IPAC), Caltech. We coin this system the IPAC/iPTF Discovery Engine (or IDE). We review the algorithms used for PSF-matching, image subtraction, detection, photometry, and machine-learned (ML) vetting of extracted transient candidates. We also review the performance of our ML classifier. For a limiting signal-to-noise ratio of 4 in relatively unconfused regions, bogus candidates from processing artifacts and imperfect image subtractions outnumber real transients by approximately equal to 10:1. This can be considerably higher for image data with inaccurate astrometric and/or PSF-matching solutions. Despite this occasionally high contamination rate, the ML classifier is able to identify real transients with an efficiency (or completeness) of approximately equal to 97% for a maximum tolerable false-positive rate of 1% when classifying raw candidates. All subtraction-image metrics, source features, ML probability-based real-bogus scores, contextual metadata from other surveys, and possible associations with known Solar System objects are stored in a relational database for retrieval by the various science working groups. We review our efforts in mitigating false-positives and our experience in optimizing the overall system in response to the multitude of science projects underway with iPTF.

methods: analytical – methods: data analysis –↗
DOE OSTI2024

Autocalibration of the E3SM Version 2 Atmosphere Model Using a PCA-Based Surrogate for Spatial Fields

Global Climate Model tuning (calibration) is a tedious and time-consuming process, with high-dimensional input and output fields. Experts typically tune by iteratively running climate simulations with hand-picked values of tuning parameters. Many, in both the statistical and climate literature, have proposed alternative calibration methods, but most are impractical or difficult to implement. We present a practical, robust, and rigorous calibration approach on the atmosphere-only model of the Department of Energy's Energy Exascale Earth System Model (E3SM) version 2. Our approach can be summarized into two main parts: (a) the training of a surrogate that predicts E3SM output in a fraction of the time compared to running E3SM, and (b) gradient-based parameter optimization. To train the surrogate, we generate a set of designed ensemble runs that span our input parameter space and use polynomial chaos expansions on a reduced output space to fit the E3SM output. We use this surrogate in an optimization scheme to identify values of the input parameters for which our model best matches gridded spatial fields of climate observations. To validate our choice of parameters, we run E3SMv2 with the optimal parameter values and compare prediction results to expertly-tuned simulations across 45 different output fields. This flexible, robust, and automated approach is straightforward to implement, and we demonstrate that the resulting model output matches present day climate observations as well or better than the corresponding output from expert tuned parameter values, while considering high-dimensional output and operating in a fraction of the time.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Statistical analysis of HAADF-STEM images to determine the surface coverage and distribution of immobilized molecular complexes

The surface immobilization of molecular catalysts is attractive because it combines the benefits of homogeneous and heterogeneous catalysis. However, determining the surface coverage and distribution of a molecular catalyst on a solid support is often challenging, inhibiting our ability to design improved catalytic systems. Here, in this work, we demonstrate that the combination of scanning transmission electron microscopy (STEM) and image analysis of the individual positions of heavy atoms in transition metal complexes via a convolutional neural network (CNN) allows statistically robust determination of the surface coverage and distribution of immobilized molecular catalysts. These observations provide information about how changes in the functionalization conditions, attachment group, and structure of the molecular catalyst affect the surface coverage and distribution, providing insight into the chemical mechanism of surface immobilization. The method could be generally valuable for correlating the surface coverage and distribution to the activity, selectivity, and stability of a catalytic system.

HAADF-STEM↗
DOE OSTI2025

A hybrid neural architecture: Online attosecond x-ray characterization

The emergence of high-repetition-rate x-ray free-electron lasers (XFELs), such as SLAC’s LCLS-II, serves as our canonical example for autonomous controls that necessitate high-throughput diagnostics paired with streaming computational pipelines capable of single-shot analysis with extremely low latency. We present the deterministic characterization with an integrated parallelizable hybrid resolver architecture, a hybrid machine learning framework designed for fast, accurate analysis of XFEL diagnostics using angular streaking-based sinogram images. This architecture integrates convolutional neural networks and bidirectional long short-term memory models to denoise input, identify x-ray sub-spike features, and extract sub-spike relative delays with sub-30 attosecond temporal resolution. Deployed on low-latency hardware, it achieves over 10 kHz throughput with 168.3 μs inference latency, indicating scalability to 14 kHz with field-programmable gate array integration. By transforming regression tasks into classification problems and leveraging optimized error encoding, we achieve high precision with low-latency performance that is critical for real-time streaming event selection and experimental control feedback signals. This represents a key development in real-time control pipelines for next-generation autonomous science, generally, and high repetition-rate x-ray experiments in particular.

Accelerator Physics (physics.acc-ph)↗
DOE OSTI2026

Recent advances in plasma control and physics research in the Large Helical Device

The Large Helical Device (LHD), the largest superconducting helical system in the world, is equipped with advanced heating and diagnostic tools, facilitating plasma control and physics research. Data assimilation was employed for electron temperature control using a real-time Thomson scattering system and real time prediction code. A virtual LHD environment enabled visualization of escaping high-energy tritium ions and demonstrated that these ions impact the rear side of the divertor plate. Pioneering results crucial to plasma control have also been achieved. Real-time wall conditioning using Lithium granule dropping improved bulk ion energy and particle transport while simultaneously enhancing the heavy impurity transport. Progress has also been made in the investigation of turbulence-driven transport. At the confinement bifurcation, ion-scale turbulence decreased, while electron-scale turbulence increased. A change in the anisotropy of turbulent eddies was also observed at the confinement bifurcation. Coexistence of local and non-local turbulence was identified in electron-scale turbulence. Non-local turbulence exhibited the rapid spatial propagation of perturbations throughout the plasma, while local turbulence followed the temperature gradient. A transition between drift-wave turbulence and magnetohydrodynamics (MHD) turbulence was observed with the turbulence minimized at the transition condition. Machine learning analysis was employed to evaluate the temperate and density conditions of this turbulence transition. Then, real-time control of fueling and heating was applied to maintain the turbulence transition condition, improving the energy confinement enhancement factor by 20%. In addition, evidence was obtained for collisionless ion heating by energetic-ion-driven geodesic acoustic modes and MHD bursts. These achievements represent unique contributions to the development of fusion reactors.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗