Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Machine learning in physics”

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 1,333 records · Page 74
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Scalable Bayesian Physics-Informed Kolmogorov-Arnold Networks

Uncertainty quantification (UQ) plays a pivotal role in scientific machine learning, especially when surrogate models are used to approximate complex systems. Although multilayer perceptions (MLPs) are commonly employed as surrogates, they often suffer from overfitting due to their large number of parameters. Kolmogorov-Arnold networks (KANs) offer an alternative solution with fewer parameters. However, gradient-based inference methods, such as Hamiltonian Monte Carlo (HMC), may result in computational inefficiency when applied to KANs, especially for large-scale datasets, due to the high cost of back-propagation. To address these challenges, we propose a novel approach, combining the dropout Tikhonov ensemble Kalman inversion (DTEKI) with Chebyshev KANs. This gradient-free method effectively mitigates overfitting and enhances numerical stability. In addition, we incorporate the active subspace method to reduce the parameter-space dimensionality, allowing us to improve the accuracy of predictions and obtain more reliable uncertainty estimates. Extensive experiments demonstrate the efficacy of our approach in various test cases, including scenarios with large datasets and high noise levels. Our results show that the new method achieves comparable or better accuracy, much higher efficiency as well as stability compared to HMC, in addition to scalability. Moreover, by leveraging the low-dimensional parameter subspace, our method preserves prediction accuracy while substantially reducing further the computational cost.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Elsa: enhanced latent spaces for improved collider simulations

Abstract Simulations play a key role for inference in collider physics. We explore various approaches for enhancing the precision of simulations using machine learning, including interventions at the end of the simulation chain (reweighting), at the beginning of the simulation chain (pre-processing), and connections between the end and beginning (latent space refinement). To clearly illustrate our approaches, we use W + jets matrix element surrogate simulations based on normalizing flows as a prototypical example. First, weights in the data space are derived using machine learning classifiers. Then, we pull back the data-space weights to the latent space to produce unweighted examples and employ the Latent Space Refinement ( Laser ) protocol using Hamiltonian Monte Carlo. An alternative approach is an augmented normalizing flow, which allows for different dimensions in the latent and target spaces. These methods are studied for various pre-processing strategies, including a new and general method for massive particles at hadron colliders that is a tweak on the widely-used RamboOnDiet mapping. We find that modified simulations can achieve sub-percent precision across a wide range of phase space.

Nachman, Benjamin (ORCID:0000000310240932)↗
DOE OSTI2024

Rapid Electrochemical Diagnosis of Battery Health and Safety from Cells to Modules

Rapid electrochemical diagnosis of battery health and failure is critical for ensuring reliable battery performance and battery safety. Traditional battery health diagnostics such as capacity measurements and DC pulse tests are reliable and well-understood, however, these measurements of battery capacity and resistance do not capture all aspects of battery degradation. Other aspects of degradation, such as electrolyte decomposition, lithium-plating, and particle cracking are difficult to detect electrochemically but are crucial to measure to get a full picture of battery safety and flag out potential failures. In this work, lab- and field-aged commercial lithium-ion batteries and modules of various chemistries and formats are tested using a variety of traditional electrochemical characterization methods as well as using 2-minute pseudo-random DC pulse sequences at rest and during charge/discharge. The electrochemical measurements are compared to physical cell measurements, cell efficiency, drive cycle performance, physical and thermal heterogeneity, and qualitative safety metrics using statistical and machine-learning methods to discover if a comprehensive "battery health map" can be accurately identified using only rapid DC measurements.

ADVANCED PROPULSION SYSTEMS,ENERGY STORAGE↗
DOE OSTI2025

Graph-based Reversible Evaluation and Tangents Library

GRETL is a C++ library for evaluation, re-evaluation and algorithmic differentiation of functional operations on an arbitrary computational graph with limited memory usage. Similar to popular machine learning frameworks in Python, like PyTorch and JAX, it tracks and stores both operations and output data as functions are evaluated. Once this composition of functions is built up, the entire chain of operations can be back propagated to compute sensitivities of the final result with respect to any number of inputs. In contrast to most machine learning applications, memory usage becomes the bottleneck for back propagation in many physics applications, especially for time-dependent PDEs. Dynamic check pointing becomes essential. An important distinguishing feature of GRETL is its ability to limit the maximum memory usage by automatically dynamic checkpointing the data output for each graph operation (see Wang, Moin, Iaccarino, 2009). During backpropagation, parts of the graph that are no longer in memory are automatically re-evaluated from upstream checkpointed states as needed for derivative sensitivity calculations (or more precisely, for vector-Jacobian products). GRETL is particularly beneficial for applications, such as coupled multi-physics, where deriving adjoint-based sensitivities and managing checkpoint memory across modules becomes onerous. Cases which can be readily handled by the GRETL library include: different time-integration algorithms per physics (e.g., coupled predictor-corrector algorithms, IMEX, etc.), sub-cycling, asynchronous integrators, state dependent timestep sizes, iterative solvers and coupling algorithms, controller algorithms, and more.

Tupek, MichaelR [Lawrence Livermore National Labor↗
DOE OSTI2024

Interpreting and Accelerating Transformers for Jet Tagging

Attention-based transformers are ubiquitous in machine learning applications from natural language processing to computer vision. In high energy physics, one central application is to classify collimated particle showers in colliders based on the particle of origin, known as jet tagging. In this work, we study the interpretatbility and prospects for acceleration of Particle Transformer (ParT), a state-of-the-art model, leverages particle-level attention to improve jet-tagging performance. We analyzing ParT's attention maps and particle-pair correlations in the eta-phi plane, revealing intriguing features, such as a binary attention pattern that identifies critical substructure in jets. These insights enhance our understanding of the model's internal workings and learning process and hint at ways to improve its efficiency. Along these lines, we also explore low-rank attention, attention alternatives, and dynamic quantization to accelerate transformers for jet tagging. With quantization, we achieve a 50% reduction in model size and a 10% increase in inference speed without compromising accuracy. These combined efforts enhance both the performance and the interpretability of transformers in high-energy physics, opening avenues for more efficient and physics-driven model designs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Computational Materials Design for Ceramic Nuclear Waste Forms Using Machine Learning, First-Principles Calculations, and Kinetics Rate Theory

Ceramic waste forms are designed to immobilize radionuclides for permanent disposal in geological repositories. One of the principal criteria for the effective incorporation of waste elements is their compatibility with the host material. In terms of performance under environmental conditions, the resistance of the waste forms to degradation over long periods of time is a critical concern when they are exposed to natural environments. Due to their unique crystallographic features and behavior in nature environment as exemplified by their natural analogues, ceramic waste forms are capable of incorporating problematic nuclear waste elements while showing promising chemical durability in aqueous environments. Recent studies of apatite- and hollandite-structured waste forms demonstrated an approach that can predict the compositions of ceramic waste forms and their long-term dissolution rate by a combination of computational techniques including machine learning, first-principles thermodynamics calculations, and modeling using kinetic rate equations based on critical laboratory experiments. By integrating the predictions of elemental incorporation and degradation kinetics in a holistic framework, the approach could be promising for the design of advanced ceramic waste forms with optimized incorporation capacity and environmental degradation performance. Such an approach could provide a path for accelerated ceramic waste form development and performance prediction for problematic nuclear waste elements.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Foreword to special issue: Papers from the 63rd annual meeting of the APS Division of Plasma Physics, November 8–12, 2021

The 63rd annual meeting of the APS Division of Plasma Physics (DPP) was held on November 8–12, 2021 in Pittsburgh at the David Lawrence Convention Center with both a live (in person) component and a virtual component. Following guidance from an APS COVID task force, all in-person attendees were fully vaccinated and masked. More than 800 physicists attended, safely, in-person. With both virtual and on-site participants, discussions were lively, and the research presentations showed unmatched mastery in the modern observation, theory, simulation, and manipulation of plasma. The presentations included four invited review talks, 97 invited talks, four tutorials, and four presentations from this year's prize and award recipients. There were more than 1200 contributed poster presentations and 725 contributed oral presentations. Including both in-person and remote attendees, DPP 2021 had a record of 2232 participants. As a hybrid meeting, in-person presentations of all invited presentations were broadcast live and were accompanied by a Q&A discussion. Contributed oral and poster presentations were prerecorded along with options to schedule in-person discussions on demand. Five mini-conferences were held: “Gatekeeper Workshop: Creating a Diverse, Equitable, and Inclusive Pipeline,” “Collisionless Shocks in Laboratory and Space Plasmas,” “The High Repetition Rate Frontier in High-Energy-Density Physics,” “Measuring and Modeling Plasma Surface Interactions,” and “The Second Mini-Conference on Machine Learning, Data Science and Artificial Intelligence in Plasma Research.” Finally, on the day before the official start of the meeting, an afternoon “for students, by students” included lightning talks, plasma trivia, and an informal occasion to connect with other students, learn how to get the most from the DPP Annual Meeting, and share successful ways to connect with colleagues and advance their professional careers.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Online Characterization of Mixed Plastic Waste Using Machine Learning and Mid-Infrared Spectroscopy

To recycle the mixed plastic wastes (MPW), it is important to obtain the compositional information online in real time. We present a sensing framework based on a convolutional neural network (CNN) and mid-infrared spectroscopy (MIR) for the rapid and accurate characterization of MPW. The MPW samples are placed on a moving platform to mimic the industrial environment. The MIR spectra are collected at the rate of 100 Hz, and the proposed CNN architecture can reach an overall prediction accuracy close to 100%. Therefore, the proposed method paves the way toward the online MPW characterization in industrial applications where high throughput is needed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Breeding Realistic D‐Brane Models

Abstract Intersecting branes provide a useful mechanism to construct particle physics models from string theory with a wide variety of desirable characteristics. The landscape of such models can be enormous, and navigating towards regions which are most phenomenologically interesting is potentially challenging. Machine learning techniques can be used to efficiently construct large numbers of consistent and phenomenologically desirable models. In this work we phrase the problem of finding consistent intersecting D‐brane models in terms of genetic algorithms, which mimic natural selection to evolve a population collectively towards optimal solutions. For a four‐dimensional supersymmetric type IIA orientifold with intersecting D6‐branes, we demonstrate that unique, fully consistent models can be easily constructed, and, by a judicious choice of search environment and hyper‐parameters, of the found models contain the desired Standard Model gauge group factor. Having a sizable sample allows us to draw some preliminary landscape statistics of intersecting brane models both with and without the restriction of having the Standard Model gauge factor.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Chemical signature characterization with hyperspectral imagery: novel deep learning model architectures and physically-motivated data augmentation techniques

The high spectral resolution afforded by Hyperspectral Imaging (HSI) sensors is poised to bring unprecedented advancements to signature characterization applications. Thus far, much of the research in the machine learning field devoted to HSI applications has focused on a few specific tasks like land-use land-cover classification. In land classification tasks, spatial information is very important, and model architectures are often designed to leverage spatial contexts. However, it is unclear how well these spatially-tuned models will translate to tasks where spectral information is critical, like the detection and characterization of chemicals. In this work, we compare spectral models (inputs are 1D spectra) and spatial-spectral models (inputs are 3D cubes) in the context of predicting chemical concentration maps. We find that spatial-spectral models perform the best, though we find a wide range in performance across the different architectures tested. Additionally, we find that model performance is impacted by the availability of training data, particularly in scenarios where the training data doesn't fully capture the true variance of real-world conditions. We find that data augmentation can help mitigate sparse coverage of observed parameter space (e.g., seasonal or geographic variability in ground cover), and present augmentation strategies that are tailored to hyperspectral data.

• Artificial intelligence (AI) / machine learning ↗