Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Training Time”

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 1,513 records · Page 84
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Safeguards-Informed Hybrid Imagery Dataset [Poster]

Deep Learning computer vision models require many thousands of properly labelled images for training, which is especially challenging for safeguards and nonproliferation, given that safeguards-relevant images are typically rare due to the sensitivity and limited availability of the technologies. Creating relevant images through real-world staging is costly and limiting in scope. Expert-labeling is expensive, time consuming, and error prone. We aim to develop a data set of both realworld and synthetic images that are relevant to the nuclear safeguards domain that can be used to support multiple data science research questions. In the process of developing this data, we aim to develop a novel workflow to validate synthetic images using machine learning explainability methods, testing among multiple computer vision algorithms, and iterative synthetic data rendering. We will deliver one million images – both real-world and synthetically rendered – of two types uranium storage and transportation containers with labelled ground truth and associated adversarial examples.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Deep Learning-based Surrogate Model for Efficient Reservoir Simulation in Large-scale Geological Carbon Storage: Application in IBDP Dataset

This project introduces an advanced deep learning (DL)-based surrogate modeling approach to enhance the efficiency and accuracy of large-scale geological carbon storage (GCS) simulations. Using the Illinois Basin Decatur Project (IBDP) dataset as training data, the study employs a residual U-Net architecture to predict critical state variables such as pressure and CO₂ saturation, as well as CO₂ plume migration. By incorporating key geological parameters (e.g., porosity, permeability, and rock facies) and physics-informed inputs like the diffusive time of flight and time step, the DL model effectively reduces computational complexity while maintaining robust physical constraints. Compared to traditional simulators like Eclipse, the DL model achieves remarkable accuracy, with a root mean square error (RMSE) of 1.57 psi for pressure and 0.007 for saturation, and dramatically reduces computational time from hours to just 69.9 seconds for 50-step simulations. These results demonstrate the potential of innovative DL methodologies to improve the predictivity and operational efficiency of GCS simulations, providing a reliable foundation for decision-making in CCS operations. Supported by the SMART initiative, this project underscores the success of leveraging computational innovations to advance CCS technologies.

advanced deep learning↗
DOE OSTI2026

Hofmann Stability Charts Revisited for PIP-II: From Classical Theory to Assumption-Free and ML-Driven Maps

The Hofmann stability chart remains a standard for visualizing parametric resonances in space-charge–dominated linacs, but its use typically relies on non-oscillatory Vlasov dispersion relations with simplifying assumptions (continuous focusing, KV phase space, linear optics, limited transverse–longitudinal coupling). We revisit the chart for the PIP-II linac along three tracks. (1) We reproduce the conventional maps in the (νz/νx, νx/ν0x) plane for relevant εz/εx, providing a validated reference. (2) We remove key assumptions by deriving stability surfaces directly from multi-particle tracking with realistic lattice discreteness, RF defocusing, solenoid/quad optics, and bunched-beam dynamics; local tunes and early-time growth rates are estimated from envelope oscillations and projected to the same coordinates. These assumption-reduced maps recover the canonical stopbands while revealing shifts and broadenings driven by tune modulation, non-KV distributions, and transverse–longitudinal coupling at PIP-II intensities. (3) We train a compact machine-learning surrogate that emulates the growth surface from zero-current optics, tune depression, emittance ratio, bunching factor, and selected lattice descriptors, enabling rapid scans and online working-point selection. We compare the three representations on representative PIP-II sections and discuss implications for commissioning guard bands, resonance avoidance, and routine operations.

Pathak, Abhishek [Fermilab] (ORCID:000000021704208↗
DOE OSTI2023

Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning

Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.

Maldonado Puente, Bryan↗
DOE OSTI2023

ICRF wave propagation and absorption modelling via machine learning

A surrogate model of the wave absorption in the ion cyclotron range of frequencies is presented. The model is trained to capture the physics of 1D electron and ion power absorption profiles for both the high harmonic fast wave scheme in NSTX, and the minority heating scheme in WEST. The surrogate models, based on both the random forest regressor and the multilayer perceptron algorithms, reduce inference time of 1D power absorption profiles from 1-5 minutes required by TORIC to ∼50 µs with high accuracy (i.e. R2 = 0.71−0.96).

Sánchez-Villar↗
DOE OSTI2025

$\mathrm{SageNet}$: Fast Neural Network Emulation of the Stiff-amplified Gravitational Waves from Inflation

Accurate modeling of the inflationary gravitational waves (GWs) requires time-consuming, iterative numerical integrations of differential equations to take into account their backreaction on the expansion history. To improve computational efficiency while preserving accuracy, we present the Stiff-amplified Gravitational-wave Emulator Network (SageNet), a deep learning framework designed to replace conventional numerical solvers (code available at https://github.com/YifangLuo/SageNet). SageNet employs a long short-term memory architecture to emulate the present-day energy density spectrum of the inflationary GWs with possible stiff amplification, Ω GW (f). Trained on a data set of 25,689 numerically generated solutions, SageNet allows accurate reconstructions of Ω GW (f) and generalizes well to a wide range of cosmological parameters; 90.9% of the test emulations with randomly distributed parameters exhibit errors of under 4%. In addition, SageNet demonstrates its ability to learn and reproduce the artificial, adaptive sampling patterns in numerical calculations, which implement denser sampling of frequencies around changes in spectral indices in Ω GW (f). The dual capability of learning both physical and artificial features of the numerical GW spectra establishes SageNet as a robust alternative to exact numerical methods. Finally, our benchmark tests show that SageNet reduces the computation time from tens of seconds to milliseconds, achieving a speedup of ∼10 4 times over standard CPU-based numerical solvers with the potential for further acceleration on GPU hardware. These capabilities make SageNet a powerful tool for accelerating Bayesian inference procedures for extended cosmological models. In a broad sense, the SageNet framework offers a fast, accurate, and generalizable solution to modeling cosmological observables whose theoretical predictions demand costly differential equation solvers.

Astronomy data modeling↗
DOE OSTI2024

Multi-Level Structural Damage Characterization Using Sparse Acoustic Sensor Networks and Knowledge Transferred Deep Learning

Standard structural health monitoring techniques face well-known difficulties for comprehensive defect diagnosis in real-world structures that have structural, material, or geometric complexity. This motivates the exploration of machine-learning-based structural health monitoring methods in complex structures. However, creating sufficient training data sets with various defects is an ongoing challenge for data-driven machine (deep) learning algorithms. The ability to transfer the knowledge of a trained neural network from one component to another or to other sections of the same component would drastically reduce the required training data set. Also, it would facilitate computationally inexpensive machine learning based inspection systems. In this work, a machine-learning-based multi-level damage characterization is demonstrated with the ability to transfer trained knowledge within the sparse sensor network. A novel network spatial assistance and an adaptive convolution technique are proposed for efficient knowledge transfer within the deep learning algorithm. Proposed structural health monitoring method is experimentally evaluated on an aluminum plate with artificially induced defects. It was observed that the method improves the performance of knowledge transferred damage characterization by 50% during localization and 24% during severity assessment. Further, experiments using time windows with and without multiple edge reflections are studied. Results reveal that multiply scattered waves contain rich and deterministic defect signatures that can be mined using deep learning neural networks, improving the accuracy of both identification and quantification. In the case of a fixed sensor network, using multiply scattered waves shows 100% prediction accuracy at all levels of damage characterization.

36 MATERIALS SCIENCE↗
DOE OSTI2024

PAIReD jet: A multi-pronged resonance tagging strategy across all Lorentz boosts

We propose a new approach of jet-based event reconstruction that aims to optimally exploit correlations between the products of a hadronic multi-pronged decay across all Lorentz boost regimes. The new approach utilizes clustered small-radius jets as seeds to define unconventional jets, referred to as PAIReD jets. The constituents of these jets are subsequently used as inputs to machine learning-based algorithms to identify the flavor content of the jet. We demonstrate that this approach achieves higher efficiencies in the reconstruction of signal events containing heavy-flavor jets compared to other event reconstruction strategies at all Lorentz boost regimes. Classifiers trained on PAIReD jets also have significantly better background rejections compared to those based on traditional event reconstruction approaches using small-radius jets at low Lorentz boost regimes. The combined effect of a higher signal reconstruction efficiency and better classification performance results in a two to four times stronger rejection of light-flavor jets compared to conventional strategies at low Lorentz-boosts, and rejection rates similar to classifiers based on large-radius multi-pronged jets at high Lorentz-boost regimes.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Reliable extrapolation of deep neural operators informed by physics or sparse observations

Deep neural operators can learn nonlinear mappings between infinite-dimensional function spaces via deep neural networks. As promising surrogate solvers of partial differential equations (PDEs) for real-time prediction, deep neural operators such as deep operator networks (DeepONets) provide a new simulation paradigm in science and engineering. Pure data-driven neural operators and deep learning models, in general, are usually limited to interpolation scenarios, where new predictions utilize inputs within the support of the training set. However, in the inference stage of real-world applications, the input may lie outside the support, i.e., extrapolation is required, which may result to large errors and unavoidable failure of deep learning models. Here, we address this challenge of extrapolation for deep neural operators. First, we systematically investigate the extrapolation behavior of DeepONets by quantifying the extrapolation complexity, via the 2-Wasserstein distance between two function spaces and propose a new strategy of bias–variance trade-off for extrapolation with respect to model capacity. Subsequently, we develop a complete workflow, including extrapolation determination, and we propose five reliable learning methods that guarantee a safe prediction under extrapolation by requiring additional information—the governing PDEs of the system or sparse new observations. The proposed methods are based on either fine-tuning a pre-trained DeepONet or multifidelity learning. We demonstrate the effectiveness of the proposed framework for various types of parametric PDEs. Furthermore, our systematic comparisons provide practical guidelines for selecting a proper extrapolation method depending on the available information, desired accuracy, and required inference speed.

42 ENGINEERING↗
DOE OSTI2021

Automated discovery of a robust interatomic potential for aluminum

Machine learning, trained on quantum mechanics (QM) calculations, is a powerful tool for modeling potential energy surfaces. A critical factor is the quality and diversity of the training dataset. Here we present a highly automated approach to dataset construction and demonstrate the method by building a potential for elemental aluminum (ANI-Al). In our active learning scheme, the ML potential under development is used to drive non-equilibrium molecular dynamics simulations with time-varying applied temperatures. Whenever a configuration is reached for which the ML uncertainty is large, new QM data is collected. The ML model is periodically retrained on all available QM data. The final ANI-Al potential makes very accurate predictions of radial distribution function in melt, liquid-solid coexistence curve, and crystal properties such as defect energies and barriers. We perform a 1.3M atom shock simulation and show that ANI-Al force predictions shine in their agreement with new reference DFT calculations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Modeling gas–shell mixing in ICF with separated reactants

Mixing between fuel and shell materials in ICF implosions can affect implosion dynamics and even prevent ignition. We use data from a series of separated reactant experiments on the National Ignition Facility to calibrate and test the predictive power of gas–shell mix models. Two models are used to estimate fuel–shell mix: a Reynolds-averaged turbulence model and molecular diffusion. Minor uncertainties in capsule manufacture, experimental conditions, and values for mix model parameters produce significant variation in simulation results. Using input/output pairs from 1D simulations, we train Gaussian process surrogate models to predict experimental quantities of interest. The surrogates are used to construct posteriors for mix model parameters by marginalizing over uncertainties in capsule manufacture and experimental conditions. Mix models are calibrated with a subset of experimental data (neutron yields, ion temperature, and bang time) and tested using the remaining data. In general, both the diffusion and turbulence model correctly predict experimental DT and TT neutron yields. Despite having more free parameters, the turbulence model underpredicts ion temperature at high convergence ratio. Furthermore, the simpler diffusion model correctly predicts these temperatures, suggesting nonhydrodynamic gas–shell mix. The computational model consistently overpredicts DD neutron yield, indicating possible shortcomings outside of the mix model.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Fast inference of Boosted Decision Trees in FPGAs for particle physics

We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Machine learning of hidden variables in multiscale fluid simulation

Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ‘hidden variables’ that are non-local in space and time. The behavior of the ‘hidden variables’ in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or particle-in-cell modeling.

42 ENGINEERING↗
DOE OSTI2026

REV-INR: Regularized Evidential Implicit Neural Representation for Uncertainty-Aware Volume Visualization

Applications of Implicit Neural Representations (INRs) have emerged as a promising deep learning approach for compactly representing large volumetric datasets. These models can act as surrogates for volume data, enabling efficient storage and on-demand reconstruction via model predictions. However, conventional deterministic INRs only provide value predictions without insights into the model’s prediction uncertainty or the impact of inherent noisiness in the data. This limitation can lead to unreliable data interpretation and visualization due to prediction inaccuracies in the reconstructed volume. Identifying erroneous results extracted from model-predicted data may be infeasible, as raw data may be unavailable due to its large size. To address this challenge, we introduce REV-INR, Regularized Evidential Implicit Neural Representation, which learns to predict data values accurately along with the associated coordinate-level data uncertainty and model uncertainty using only a single forward pass of the trained REV-INR during inference. By comprehensively comparing and contrasting REV-INR with existing well-established deep uncertainty estimation methods, we show that REV-INR achieves the best volume reconstruction quality with robust data (aleatoric) and model (epistemic) uncertainty estimates using the fastest inference time. Consequently, we demonstrate that REV-INR facilitates assessment of the reliability and trustworthiness of the extracted isosurfaces and volume visualization results, enabling analyses to be solely driven by model-predicted data.

Saklani, Shanu [Indian Institute of Technology, Ka↗
DOE OSTI2023

Cyberguardians and STEM Warriors (Final Technical Report (FTR))

In the past decade, solar power, with and without energy storage has become the fastest growing source of energy generation in the world. In the U.S., solar employment more than doubled from 105,145 jobs in 2011 to 255,037 jobs in 2021, four times faster than the U.S. job growth rate overall. These factors, combined with technology advancements, creates a skills gap that puts tremendous stress on society to deliver the workers to fill the open job requisitions. The Cyberguardians and STEM Warriors project (Cyberguardians) was designed to address the trained-worker shortage in the energy industry in three ways: 1) by developing educational curriculum that addresses DER technology changes; 2) by delivering curriculum to prospective workers, including military veterans and their families, via universities, community colleges, and vocational training outlets; and 3) introducing individuals who have completed training to employers that can hire them. Cyberguardians exceeded its curriculum goals by producing 27 academic units of university-accredited material (12 total courses) covering energy fundamentals, smart inverters, Distributed Energy Resource (DER) data communication, cybersecurity, standardization, certification, data analytics, and IEEE 1547 standard topics. The North American Board of Certified Energy Practitioners (NABCEP) also accredited the material for use in their credential program. Seven instructors were recruited and trained, and six academic institutions (University of California San Diego, State University of New York, North Carolina State University, Harper Community College, Green Village Academy, and the SunSpec Alliance) were enlisted, meeting program goals. All course material was published under the Creative Commons license and made available royalty free, thus providing a long-lasting public benefit. The program’s outreach program vastly exceeded program goals and incorporated the efforts of 13 outreach partners (11 of which are veteran focused), an advisory board representing 15 companies, webinars and 10’s of thousands of email messages sent to prospective students and hiring managers. Despite these efforts, the global pandemic depressed anticipated program participation by about a third. Still, a total of 396 students enrolled and 289 completed the courses and were accredited. The job applicant task achieved similar results (111 realized vs a 174 goal) but reported job placement was weaker at (9 realized vs. a 51 goal). The Cyberguardians program fills a critical void for cost-effective, royalty-free curriculum and training pertaining to DER technologies and cybersecurity that prospective energy workers must possess to be effective in the 21 st century. On this basis alone, the investment of taxpayer funds will pay dividends for years to come.

14 SOLAR ENERGY↗
DOE OSTI2023

Using Eye-Tracking to Quantify Reverse Engineering Expertise

Software reverse engineering (RE) requires analysts to closely read and make decisions about code. Little is known about what makes an analyst successful, making it difficult to train new analysts or design tools to augment existing ones. The goal of this project was to quantify the eye movement behaviors supporting RE and code comprehension more generally. We applied eye-tracking methods from the language comprehension literature to understand where analysts direct their attention over time when completing tasks (e.g., function identification, bug detection). Across three studies, we manipulated aspects of code hypothesized to impact comprehension (e.g., variable name meaningfulness, code complexity) and presentation methods (e.g., line-by-line, free viewing, gaze-contingent moving window) to understand effects on accuracy and gaze patterns. Results showed clear benefits of meaningful variable names, and effects of expertise on global and line-specific viewing patterns. Findings could inspire empirically-supported tool or analytic adaptations that help to reduce analyst workload.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Genome dependent Cas9/gRNA search time underlies sequence dependent gRNA activity

Abstract CRISPR-Cas9 is a powerful DNA editing tool. A gRNA directs Cas9 to cleave any DNA sequence with a PAM. However, some gRNA sequences mediate cleavage at higher efficiencies than others. To understand this, numerous studies have screened large gRNA libraries and developed algorithms to predict gRNA sequence dependent activity. These algorithms do not predict other datasets as well as their training dataset and do not predict well between species. Here, to better understand these discrepancies, we retrospectively examine sequence features that impact gRNA activity in 44 published data sets. We find strong evidence that gRNA sequence dependent activity is largely influenced by the ability of the Cas9/gRNA complex to find the target site rather than activity at the target site and that this drives sequence dependent differences in gRNA activity between different species. This understanding will help guide future work to understand Cas9 activity as well as efforts to identify optimal gRNAs and improve Cas9 variants.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

ML-Based Power System Stability Assessment Considering Network Topology Changes: WECC 20,000+ Bus System Case Study

Modern power grids are fast-changing and thus require real-time monitoring and online stability assessment. With the rapid development of machine learning (ML) techniques, using data-driven models to provide fast and accurate estimations of power system stability marginal information, such as frequency nadir for frequency stability and critical clearing time (CCT) for transient stability, have become possible. However, despite the numerous research on ML-based methods for frequency nadir and CCT prediction, there is limited work on the impact of different network topology changes. Furthermore, most previous studies only focused on small or synthetic systems, and there is a lack of research on actual large power system models. In this paper, the above issues are addressed by studying the actual U.S. Western Electricity Coordinating Council (WECC) system model with more than 20,000 buses. Massive simulations are conducted in PowerWorld Simulator to study the impact of various topology change scenarios on both frequency stability and transient stability. System operating information is extracted from the success dispatch cases of various network topologies to generate a comprehensive dataset for ML-based models. Two ML methods, random forest (RF) and multilayer perceptron (MLP) neural network, are trained and tested for both frequency nadir prediction and CCT prediction. Test results have proven the models are capable of online stability assessment for large power networks such as the WECC system with sufficient accuracy.

critical clearing time↗