SEARCH · Engineering Papers

Results for “learning efficiency”

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.

SourceResearch connectionFrom yearThrough year

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

At least 1,027 records · Page 57

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Probabilistic Nanomagnetic Memories for Uncertain and Robust Machine Learning

This project evaluated the use of emerging spintronic memory devices for robust and efficient variational inference schemes. Variational inference (VI) schemes, which constrain the distribution for each weight to be a Gaussian distribution with a mean and standard deviation, are a tractable method for calculating posterior distributions of weights in a Bayesian neural network such that this neural network can also be trained using the powerful backpropagation algorithm. Our project focuses on domain-wall magnetic tunnel junctions (DW-MTJs), a powerful multi-functional spintronic synapse design that can achieve low power switching while also opening the pathway towards repeatable, analog operation using fabricated notches. Our initial efforts to employ DW-MTJs as an all-in-one stochastic synapse with both a mean and standard deviation didn’t end up meeting the quality metrics for hardware-friendly VI. In the future, new device stacks and methods for expressive anisotropy modification may make this idea still possible. However, as a fall back that immediately satisfies our requirements, we invented and detailed how the combination of a DW-MTJ synapse encoding the mean and a probabilistic Bayes-MTJ device, programmed via a ferroelectric or ionically modifiable layer, can robustly and expressively implement VI. This design includes a physics-informed small circuit model, that was scaled up to perform and demonstrate rigorous uncertainty quantification applications, up to and including small convolutional networks on a grayscale image classification task, and larger (Residual) networks implementing multi-channel image classification. Lastly, as these results and ideas all depend upon the idea of an inference application where weights (spintronic memory states) remain non-volatile, the retention of these synapses for the notched case was further interrogated. These investigations revealed and emphasized the importance of both notch geometry and anisotropy modification in order to further enhance the endurance of written spintronic states. In the near future, these results will be mapped to effective predictions for room temperature and elevated operation DW-MTJ memory retention, and experimentally verified when devices become available.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

DOE OSTI2022

ThunderSecure: deploying real-time intrusion detection for 100G research networks by leveraging stream-based features and one-class classification network

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks (40/100G) across distributed sites for processing, storage, and analysis. Like general purpose networks, research networks experience intrusions. However, monitoring anomalies in such high-speed network traffics is challenging given current cyber-infrastructure. Moreover, traditional network intrusion detection systems (NIDS) are signature based. However, anomaly patterns are difficult to define and that rulesets are often not updated frequently enough to reflect the changes of attack behaviors. We present ThunderSecure, a high-throughput, unsupervised learning-based intrusions detection system for 100G research networks. ThunderSecure implements an efficient packet processing and detection pipeline using multi-cores and GPUs. It extracts statistical and temporal features from real-time network data streams and feeds them to a one-class anomaly detection network. A baseline of normal distribution will be created based on the training observation. Testing traffic deviated from the learned profile will be marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets mirrored from two 100G network connections at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. Results show that ThunderSecure can recognize science data traffic captured long after the training and made nearly certain detection on the segment of the streams where anomalous flows were injected.

100G research network↗

DOE OSTI2026

Hierarchical Testing of a Hybrid Machine Learning‐Physics Global Atmosphere Model

Machine learning (ML)-based models have demonstrated high skill and computational efficiency, often outperforming conventional physics-based models in weather and subseasonal predictions. While prior studies have assessed their fidelity in capturing synoptic-scale atmospheric dynamics, their performance across timescales and under out-of-distribution forcing, such as +3K or +4K uniform-warming forcings, and the sources of biases remain elusive, to establish the model's reliability for Earth science. Here, we design three sets of experiments targeting synoptic-scale phenomena, interannual variability, and out-of-distribution uniform-warming forcings. We evaluate the Neural General Circulation Model (NeuralGCM), a hybrid model integrating a dynamical core with ML-based component, against observations and physics-based Earth system models (ESMs). At the synoptic scale, NeuralGCM captures the evolution and propagation of extratropical cyclones with performance comparable to ESMs. At the interannual scale, when forced by El Niño-Southern Oscillation sea surface temperature (SST) anomalies, NeuralGCM successfully reproduces associated teleconnection patterns but exhibits deficiencies in capturing nonlinear response. Under out-of-distribution uniform-warming forcings, NeuralGCM simulates similar responses in global-average temperature and precipitation and reproduces large-scale tropospheric circulation features similar to those in ESMs. Notable weaknesses include overestimating the tracks and spatial extent of extratropical cyclones, biases in the teleconnected wave train triggered by tropical SST anomalies, and differences in upper-level warming and stratospheric circulation responses to SST warming compared to physics-based ESMs. The causes of these weaknesses were explored. Despite the noted weaknesses, NeuralGCM reproduces responses across experiments reasonably and performs comparably to ESMs. By integrating a dynamical core with ML, NeuralGCM shows potential for developing ML-based ESMs.

global warming↗

DOE OSTI2023

Learning local equivariant representations for large-scale atomistic dynamics

Abstract A simultaneously accurate and computationally efficient parametrization of the potential energy surface of molecules and materials is a long-standing goal in the natural sciences. While atom-centered message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited the accessible length-scales. Local methods, conversely, scale to large simulations but have suffered from inferior accuracy. This work introduces Allegro, a strictly local equivariant deep neural network interatomic potential architecture that simultaneously exhibits excellent accuracy and scalability. Allegro represents a many-body potential using iterated tensor products of learned equivariant representations without atom-centered message passing. Allegro obtains improvements over state-of-the-art methods on QM9 and revMD17. A single tensor product layer outperforms existing deep MPNNs and transformers on QM9. Furthermore, Allegro displays remarkable generalization to out-of-distribution data. Molecular simulations using Allegro recover structural and kinetic properties of an amorphous electrolyte in excellent agreement with ab-initio simulations. Finally, we demonstrate parallelization with a simulation of 100 million atoms.

74 ATOMIC AND MOLECULAR PHYSICS↗

DOE OSTI2015

Towards computational design of zeolite catalysts for CO 2 reduction

Carbon dioxide, an energy waste by-product with significant environmental consequences can be utilized and converted into useful chemical products such as formic acid, formaldehyde, methanol or methane, but more energy and cost efficient catalytic processes are required. Here we develop the methodology for the intelligent selection of porous zeolites for dual-adsorption of hydrogen and carbon dioxide as templates for preparing the optimal catalytic environment for carbon dioxide reduction. Useful zeolite catalysts were computationally screened from over 300 thousand zeolite structures using a combination of molecular simulation and machine-learning techniques. Several of the top candidates were very promising energy-efficient templates for catalysis with the potential to perform at 50% above conventional reactors. As a result, it is also found that an optimal cavity size of around 6 Å is required to maximize the change in entropy–enthalpy upon adsorption with a maximum void space >30% to boost product formation per volume of material.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2026

Benchmarking the performance of uncertainty quantification methods for neural network-based interatomic potentials

Machine-learned interatomic potentials (ML-IAPs) continue to gain popularity as accurate, computationally efficient replacements for traditional, physics-based interatomic potentials and expensive ab initio methods. Uncertainty quantification (UQ) of ML-IAPs is a growing area of research as UQ is critical in many applications of IAPs, such as developing curated datasets, active learning-based data augmentation, self-improving models, and estimating the uncertainty of molecular dynamics simulations. In this paper, we construct and benchmark a series of different neural network potentials (NNPs) with varying network architectures to determine the performance of these models with respect to both the mean and uncertainty calibration error. Each NNP method is specifically designed to predict either epistemic or aleatoric uncertainty with particular focus on the differences in behavior between the epistemic and aleatoric uncertainty estimates. We benchmark these methods using multiple datasets common in the ML-IAP literature. The results show that the aleatoric uncertainty from single-shot model architectures is a competitive alternative to ensemble-based epistemic uncertainty predictions in regions of sufficient data-density. However, in regions where the representative data is sparse, aleatoric uncertainty models tend to overpredict and epistemic methods tend to underpredict the actual model error. We conclude that the type of UQ is crucial when discussing performance of probabilistic model results as different methods have different performance characteristics depending on the regime in which they are evaluated. Therefore, the type of UQ method should be carefully evaluated against both the data characteristics and requirements for the intended application.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Development of the Mathematics of Learning Curve Models for Evaluating Small Modular Reactor Economics

small modular reactors (SMRs)↗

DOE OSTI2014

Development of the Mathematics of Learning Curve Models for Evaluating Small Modular Reactor Economics

The cost of nuclear power is a straightforward yet complicated topic. It is straightforward in that the cost of nuclear power is a function of the cost to build the nuclear power plant, the cost to operate and maintain it, and the cost to provide fuel for it. It is complicated in that some of those costs are not necessarily known, introducing uncertainty into the analysis. For large light water reactor (LWR)-based nuclear power plants, the uncertainty is mainly contained within the cost of construction. The typical costs of operations and maintenance (O&M), as well as fuel, are well known based on the current fleet of LWRs. However, the last currently operating reactor to come online was Watts Bar 1 in May 1996; thus, the expected construction costs for gigawatt (GW)-class reactors in the United States are based on information nearly two decades old. Extrapolating construction, O&M, and fuel costs from GW-class LWRs to LWR-based small modular reactors (SMRs) introduces even more complication. The per-installed-kilowatt construction costs for SMRs are likely to be higher than those for the GW-class reactors based on the property of the economy of scale. Generally speaking, the economy of scale is the tendency for overall costs to increase slower than the overall production capacity. For power plants, this means that doubling the power production capacity would be expected to cost less than twice as much. Applying this property in the opposite direction, halving the power production capacity would be expected to cost more than half as much. This can potentially make the SMRs less competitive in the electricity market against the GW-class reactors, as well as against other power sources such as natural gas and subsidized renewables. One factor that can potentially aid the SMRs in achieving economic competitiveness is an economy of numbers, as opposed to the economy of scale, associated with learning curves. The basic concept of the learning curve is that the more a new process is repeated, the more efficient the process can be made. Assuming that efficiency directly relates to cost means that the more a new process is repeated successfully and efficiently, the less costly the process can be made. This factor ties directly into the factory fabrication and modularization aspect of the SMR paradigm—manufacturing serial, standardized, identical components for use in nuclear power plants can allow the SMR industry to use the learning curves to predict and optimize deployment costs. Thus, SMR deployment cost estimates can be modeled using learning curve theory to provide some insight into how and when SMRs would be competitive with both the GW-class reactors and other mutual competitors. However, assuming static learning curve behavior for all aspects of SMR construction is likely an invalid approach—not all components and systems in SMRs will be subject to learning, and those that are subject to learning may not learn at the same rate. A learning curvebased model must be capable of tracking these unique behaviors. At the same time, the behavior of the aggregated and integrated SMR deployment is a more useful metric for discussion and comparison—ultimately, the most useful learning curve information is that which describes power plant deployment as a whole. Therefore, the goal of SMR component and systems learning modeling includes a method to produce an overall power plant learning model. This report describes the mathematics of learning curves that can be used for estimating unit cost in future production for individual components and aggregated systems. All necessary assumptions are provided to support the derivations presented. Future work will demonstrate the implementation of this methodology and provide example calculations of SMR cost

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2026

A Step-by-Step Protocol from METASPACE to Biological Interpretation

Mass spectrometry imaging (MSI) represents an exceptional tool for exploring complex biological systems spatially at the molecular level. However, due to its multidimensional nature and large-scale data output, it presents considerable challenges when it comes to extracting meaningful biological insights. Recent advancements, such as the METASPACE platform, have enabled researchers to efficiently process, annotate, and interpret MSI datasets by leveraging machine learning and cloud-based infrastructure. In this tutorial, we present a detailed and user-friendly R-pipeline designed to help METASPACE users navigate untargeted metabolomic annotations and transform them into practical insights about their biological systems. By combining METASPACE annotations with rapid R-based screening, this workflow not only streamlined the analytical process but also enhanced the understanding of spatial molecular distribution, especially for complex systems. Here, this easy-to-follow approach has the potential for applications in diagnostics, drug discovery, environmental and ecological processes, and more. We envision this pipeline to be particularly useful for newcomers to the field of MSI and

Moreno Pedraza, Abigail↗

DOE OSTI2023

Simultaneous CTEQ-TEA extraction of PDFs and SMEFT parameters from jet and $t\bar{t}$ data

Recasting phenomenological Lagrangians in terms of SM effective field theory (SMEFT) provides a valuable means of connecting potential BSM physics at momenta well above the electroweak scale to experimental signatures at lower energies. In this work we jointly fit the Wilson coefficients of SMEFT operators as well as the PDFs in an extension of the CT18 global analysis framework, obtaining self-consistent constraints to possible BSM physics effects. Global fits are boosted with machine-learning techniques in the form of neural networks to ensure efficient scans of the full PDF+SMEFT parameter space. We focus on several operators relevant for top-quark pair and jet production at hadron colliders and obtain constraints on the Wilson coefficients with Lagrange Multiplier scans. We find mild correlations between the extracted Wilson coefficients, PDFs, and other QCD parameters, and see indications that these correlations may become more prominent in future analyses based on data of higher precision. This work serves as a new platform for joint analyses of SM and BSM physics based on the CTEQ-TEA framework.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Influence of extreme temperature conditions on CO 2 direct air capture using amino-acid solutions

Geological features play a pivotal role in determining the feasibility of deploying CO₂ direct air capture (DAC) technologies, primarily because they influence the availability of cost-effective energy sources, such as natural gas and geothermal energy, and also due to the potential for CO₂ sequestration. Many regions face challenges due to variable weather conditions including seasonal temperature fluctuations, high or low humidity, and sub-ambient temperatures. These extremes can reduce DAC performance or even lead to catastrophic events. Aqueous solvents considered for DAC systems are particularly vulnerable to seasonal variations in colder climates, where the solvent may underperform or freeze. It is therefore essential to investigate the CO₂ capture efficiency of aqueous solvents across a broad range of environmental temperatures, spanning sub-zero to hot conditions (>30 °C). In this study, DAC operation is examined using a high-flux solvent–air crossflow contactor under two major weather scenarios: (i) cold conditions below 0 °C and (ii) hot conditions above 30 °C. A parametric study is conducted to investigate the contactor performance regarding CO₂ removal efficiency, uptake capacity, and reaction kinetics versus temperature when the air velocity through the contactor exceeds 1 m/s. The efficacy of the contactor is systematically investigated using various anti-freeze amino-acid solvent formulations. A mass-transfer mechanistic model is developed to assess the process performance over a wide temperature range and propose scalable design guidelines. Machine learning is also employed to identify key parameters affecting the CO₂ capture efficiency. It is shown that air velocity and temperature are the primary factors influencing CO₂ uptake. Based on performance data obtained under subfreezing temperatures, a technoeconomic analysis is conducted to evaluate the feasibility of using aqueous solvents in seasonal cold regions. In conclusion, the findings of this study provide valuable insights into siting considerations for deploying solvent-based DAC, thereby contributing to the advancement of sustainable carbon removal solutions.

Air–liquid contactor↗

DOE OSTI2025

A comparison of surrogate constitutive models for viscoplastic creep simulation of HT-9 steel

Mechanistic microstructure-informed constitutive models for the mechanical response of polycrystals are a cornerstone of computational materials science. However, as these models become increasingly more complex – often involving coupled differential equations describing the effect of specific deformation modes – their associated computational costs can become prohibitive, particularly in optimization or uncertainty quantification tasks that require numerous model evaluations. To address this challenge, surrogate constitutive models that balance accuracy and computational efficiency are highly desirable. Data-driven surrogate models, that learn the constitutive relation directly from data, have emerged as a promising solution. In this work, we develop two local surrogate models for the viscoplastic response of a steel: a piecewise response surface method and a mixture of experts model. These surrogates are designed to adapt to complex material behavior, which may vary with material parameters or operating conditions. The surrogate constitutive models are applied to creep simulations of HT-9 steel, an alloy of considerable interest to the nuclear energy sector due to its high tolerance to radiation damage, using training data generated from viscoplastic self-consistent (VPSC) simulations. In conclusion, we define a set of test metrics to numerically assess the accuracy of our surrogate models for predicting viscoplastic material behavior, and show that the mixture of experts model outperforms the piecewise response surface method in terms of accuracy.

36 MATERIALS SCIENCE↗

DOE OSTI2018

Constructing High-Dimensional Neural Network Potential Energy Surfaces for Gas–Surface Scattering and Reactions

While the ab initio molecular dynamics (AIMD) approach to gas–surface interaction has been instrumental in exploring important issues such as energy transfer and reactivity, it is only amenable to short-time events and a limited number of trajectories because of the on-the-fly nature of the density functional theory (DFT) calculations. In this work, we report a high-dimensional global reactive potential energy surface (PES) constructed with high fidelity from judiciously placed DFT points, using a machine learning method; and it is orders-of-magnitude more efficient than AIMD in dynamical calculations and can be employed in various simulations without performing additional electronic structure calculations. Importantly, the surface atoms are included in such a PES, which provides a unique platform for studying energy transfer and scattering/reaction of the impinging molecule on the solid surface on an equal footing.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

Data-driven studies of magnetic two-dimensional materials

We use a data-driven approach to study the magnetic and thermodynamic properties of van der Waals (vdW) layered materials. We investigate monolayers of the form A 2 B 2 X 6 , based on the known material Cr 2 Ge 2 Te 6 , using density functional theory (DFT) calculations and machine learning methods to determine their magnetic properties, such as magnetic order and magnetic moment. We also examine formation energies and use them as a proxy for chemical stability. We show that machine learning tools, combined with DFT calculations, can provide a computationally efficient means to predict properties of such two-dimensional (2D) magnetic materials. Our data analytics approach provides insights into the microscopic origins of magnetic ordering in these systems. For instance, we find that the X site strongly affects the magnetic coupling between neighboring A sites, which drives the magnetic ordering. Our approach opens new ways for rapid discovery of chemically stable vdW materials that exhibit magnetic behavior.

42 ENGINEERING↗

DOE OSTI2023

On fast simulation of dynamical system with neural vector enhanced numerical solver

The large-scale simulation of dynamical systems is critical in numerous scientific and engineering disciplines. However, traditional numerical solvers are limited by the choice of step sizes when estimating integration, resulting in a trade-off between accuracy and computational efficiency. To address this challenge, we introduce a deep learning-based corrector called Neural Vector (NeurVec), which can compensate for integration errors and enable larger time step sizes in simulations. Our extensive experiments on a variety of complex dynamical system benchmarks demonstrate that NeurVec exhibits remarkable generalization capability on a continuous phase space, even when trained using limited and discrete data. NeurVec significantly accelerates traditional solvers, achieving speeds tens to hundreds of times faster while maintaining high levels of accuracy and stability. Moreover, NeurVec’s simple-yet-effective design, combined with its ease of implementation, has the potential to establish a new paradigm for fast-solving differential equations based on deep learning.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Active Learning-driven Quantitative Synthesis-Structure-Property Relations for Improving Performance and Revealing Active Sites of Nitrogen-Doped Carbon for the Hydrogen Evolution Reaction

While quantitative structure-properties relations (QSPRs) have been developed successfully in multiple fields, catalyst synthesis affects structure and in turn performance, making simple QSPRs inadequate. Furthermore, catalysts often have multiple active sites preventing one from obtaining insights into structure-property relations. Here, we develop a data-driven quantitative synthesis-structure-property relation (QS2PRs) methodology to elucidate correlations between catalyst synthesis conditions, structural properties as well as observed performance and to provide fundamental insights into active sites and a systematic way to optimize practical catalysts. Here, we demonstrate the approach to the synthesis of nitrogen-doped catalysts (NDC) made via pyrolysis for the performance of the electrochemical hydrogen evolution reaction (HER), quantified by the onset potential and the current density. We determine crystallinity, nitrogen species type and fraction, surface area, and pore structure of the NDC’s using XRD, XPS, and BET characterization. We demonstrated that an active learning-based optimization combined with various elementary machine learning tools (regression, principal component analysis, partial least squares) can efficiently identify optimum pyrolysis conditions to tune structural characteristics and performance with concomitant savings in materials and experimental time. Unlike previous reports on the importance of pyridinic or graphitic nitrogen, we discover that the electrochemical performance is not driven by a single catalyst property; rather, it arises from a multivariate influence of nitrogen dopants, pore structure and disorder in the NDC materials. Identification of active sites can help mechanistic understanding and further catalyst improvement.

42 ENGINEERING↗

DOE OSTI2021

Superconducting neural networks with disordered Josephson junction array synaptic networks and leaky integrate-and-fire loop neurons

Fully coupled randomly disordered recurrent superconducting networks with additional open-ended channels for inputs and outputs are considered the basis to introduce a new architecture to neuromorphic computing in this work. Various building blocks of such a network are designed around disordered array synaptic networks using superconducting devices and circuits as an example, while emphasizing that a similar architectural approach may be compatible with several other materials and devices. A multiply coupled (interconnected) disordered array of superconducting loops containing Josephson junctions [equivalent to superconducting quantum interference devices (SQUIDs)] forms the aforementioned collective synaptic network that forms a fully recurrent network together with compatible neuron-like elements and feedback loops, enabling unsupervised learning. This approach aims to take advantage of superior power efficiency, propagation speed, and synchronizability of a small world or a random network over an ordered/regular network. Additionally, it offers a significant factor of increase in scalability. Here, a compatible leaky integrate-and-fire neuron made of superconducting loops with Josephson junctions is presented, along with circuit components for feedback loops as needed to complete the recurrent network. Several of these individual disordered array neural networks can further be coupled together in a similarly disordered way to form a hierarchical architecture of recurrent neural networks that is often suggested as similar to a biological brain.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2024

Evidence for B + → K + ν ν ¯ decays

We search for the rare decay B + → K + ν ν ¯ in a 362 fb − 1 sample of electron-positron collisions at the ϒ ( 4 S ) resonance collected with the Belle II detector at the SuperKEKB collider. We use the inclusive properties of the accompanying B meson in ϒ ( 4 S ) → B B ¯ events to suppress background from other decays of the signal B candidate and light-quark pair production. We validate the measurement with an auxiliary analysis based on a conventional hadronic reconstruction of the accompanying B meson. For background suppression, we exploit distinct signal features using machine learning methods tuned with simulated data. The signal-reconstruction efficiency and background suppression are validated through various control channels. The branching fraction is extracted in a maximum likelihood fit. Our inclusive and hadronic analyses yield consistent results for the B + → K + ν ν ¯ branching fraction of [ 2.7 ± 0.5 ( stat ) ± 0.5 ( syst ) ] × 10 − 5 and [ 1.1 − 0.8 + 0.9 ( stat ) − 0.5 + 0.8 ( syst ) ] × 10 − 5 , respectively. Combining the results, we determine the branching fraction of the decay B + → K + ν ν ¯ to be [ 2.3 ± 0.5 ( stat ) − 0.4 + 0.5 ( syst ) ] × 10 − 5 , providing the first evidence for this decay at 3.5 standard deviations. The combined result is 2.7 standard deviations above the standard model expectation. Published by the American Physical Society 2024

Astronomy & Astrophysics↗