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 1,639 records · Page 91
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Integrating PCTRAN with AI-Driven Host-Intrusion Detection and Secured Container Systems for Advanced Malware Analysis (Summer Internship Report)

This study presents a solution for enhancing the security of the Personal Computer Transient Analyzer (PCTRAN) PC-based Nuclear Power Plant Simulator by integrating the software with an artificial intelligence (AI)-driven host-intrusion detection system (HIDS), in addition to a secured container system, for malware analysis. PCTRAN is a Windows XP-based software package that has the ability to simulate a variety of accident and transient conditions for nuclear power plants (NPPs). It offers a high-resolution replica of the Nuclear Steam Supply System (NSSS) and displays the status of important parameters allowing for operator interaction. By including AI-driven HIDS for the NSSS, the framework can identify security threats in real-time, ensuring the integrity of the nuclear simulation environment. Additionally, the secured container system offers the ability to isolate and analyze malware, preventing potential threats from affecting core systems. The integration process involves extensive testing and validation in order to ensure accuracy, reliability, and compliance with security policies. This framework sets a new precedent for secure simulation and training in NPP operations, and offers insight for future advancements in cybersecurity.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2021

Artificial Judgement Assistance from teXt (AJAX): Applying Open Domain Question Answering to Nuclear Non-proliferation Analysis

Nuclear non-proliferation analysis is complex and subjective, as the data is sparse, and examples are rare and diverse. While analysing non-proliferation data, it is often desired that the findings be completely auditable such that any claim or assertion can be sourced directly to the reference material from which it was derived. Currently this is accomplished by analysts thoroughly documenting underlying assumptions and clearly referencing details to source documents. This is a labour-intensive and time-consuming process that can be difficult to scale with geometrically increasing quantities of data. In this work, we describe an approach to leverage bi-directional language models for nuclear non-proliferation analysis. It has been shown recently that these models not only capture language syntax but also some of the relational knowledge present in the training data. We have devised a unique Salt and Pepper strategy for testing the knowledge present in the language models, while also introducing auditability function in our pipeline. We demonstrate that fine-tuning the bi-directional language models on domain specific corpus improves their ability to answer domain-specific factoid questions. Our hope is that the results presented in this paper will further the natural language processing (NLP) field by introducing the ability to audit the answers provided by the language models to bring forward the source of said knowledge.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2023

Dilated causal convolutional neural networks for forecasting zone airflow to estimate short-term energy consumption

Here this paper investigates the use of dilated causal convolutional neural networks for fine- grained temporal forecasting of building zone states. Specifically, we build and evaluate models using a small set of exogenous features (e.g., external temperature) to autoregressively predict zone airflow setpoints every minute for a 24-hour prediction window. We carefully explore the trade-off between generality and specificity in these models, training and evaluating them based on zone, zone type, month, season, and combinations thereof. When evaluated for a commercial office building in Eastern Washington with 16 zones served by variable air volume air handling units, we find that the highest performance comes from a zone-specific, season-agnostic approach; with it, we obtain an R 2 of 0.704 (averaged over zones) and an average normalized root mean square error (nRMSE) of 0.111. In contrast, the most general model (trained across all zones and seasons) yields an R 2 of only 0.416 and a nRMSE of 0.168, while a baseline zone-specific reduced order model obtains 0.443 R 2 and 0.159 nRMSE. We also report on factors affecting airflow forecasting performance, on the ability of models trained on a specific zone to generalize to other zones, and on the capability of those models trained on a specific month to generalize to other months.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

Processed Lab Data for Neural Network-Based Shear Stress Level Prediction

Machine learning can be used to predict fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. The files are extracted features and labels from lab data (experiment p4679). The features are extracted with a non-overlapping window from the original acoustic data. The first column is the time of the window. The second and third columns are the mean and the variance of the acoustic data in this window, respectively. The 4th-11th column is the the power spectrum density ranging from low to high frequency. And the last column is the corresponding label (shear stress level). The name of the file means which driving velocity the sequence is generated from. Data were generated from laboratory friction experiments conducted with a biaxial shear apparatus. Experiments were conducted in the double direct shear configuration in which two fault zones are sheared between three rigid forcing blocks. Our samples consisted of two 5-mm-thick layers of simulated fault gouge with a nominal contact area of 10 by 10 cm^2. Gouge material consisted of soda-lime glass beads with initial particle size between 105 and 149 micrometers. Prior to shearing, we impose a constant fault normal stress of 2 MPa using a servo-controlled load-feedback mechanism and allow the sample to compact. Once the sample has reached a constant layer thickness, the central block is driven down at constant rate of 10 micrometers per second. In tandem, we collect an AE signal continuously at 4 MHz from a piezoceramic sensor embedded in a steel forcing block about 22 mm from the gouge layer The data from this experiment can be used with the deep learning algorithm to train it for future fault property prediction.

15 GEOTHERMAL ENERGY↗
DOE OSTI2022

Initial Findings from US Department of Energy’s Better Plants Virtual in-Plant Training on 50001 Ready

Manufacturing facilities use about 35% of the domestic energy in the United States every year. Implementing an effective energy management system (EnMS) is one of the most important approaches to improve energy efficiency. However, the implementation of EnMS is low for many countries (including the US) and even for energy-intensive sectors. The reasons for the low implementation rate of energy management systems had been investigated by multiple researchers, but very few studies have focused on the barriers and challenges of implementing ISO 50001-based energy management systems. To contribute to this understudied area, this paper discusses the implementation and outcomes of the first Better Plants 50001 Ready Virtual In-plant Training. This paper first provides an overview of 50001 Ready and the 50001 Ready Navigator Tool. Then, it provides details on this training event and its outcomes. Finally, it discusses findings from the responses to 40 live polling questions about the status of the 25 tasks of the 50001 Ready Navigator for participating companies, key components of the participating manufacturing companies’ energy management systems, and challenges and barriers that these companies are facing. The findings suggest that although many companies understood the importance of an effective energy management system, about half of them do not understand the required resources for building energy management systems, and most of them have only just begun establishing these systems and need more assistance and resources in multiple areas. More specifically, more assistance is necessary for the following: (1) improving corporate management’s understanding of the time and resources needed to build an EnMS as well as the benefits; (2) creating linear regression models for more accurate energy performance tracking; (3) understanding energy use, collecting and analyzing energy performance data; (4) optimizing equipment operational controls, and creating action plans.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗
DOE OSTI2023

Temperature outweighs light and flow as the predominant driver of dissolved oxygen in US rivers

The concentration of dissolved oxygen (DO), an important measure of water quality and river metabolism, varies tremendously in time and space. Riverine DO is commonly perceived as regulated by interacting and competing drivers (light, temperature and flow) that define rivers’ climate. Its continental-scale drivers, however, have remained elusive, partly due to the scarcity and spatio-temporal inconsistency of water quality data. Here we show, via a deep learning model (long short-term memory) trained using data from 580 rivers, that temperature predominantly drives daily DO dynamics in the contiguous United States. Light comes a close second, whereas flow imparts minimal influence. This work showcases the promise of using deep learning models for data filling that enables large-scale systematic analysis of patterns and drivers. Results show fairly accurate prediction of DO by temperature alone, and declining DO in warming rivers, which has important implications for water security and ecosystem health in the future climate.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Adversarial methods to reduce simulation bias in neutrino interaction event filtering at liquid argon time projection chambers

For current and future neutrino oscillation experiments using large liquid argon time projection chambers (LAr-TPCs), a key challenge is identifying neutrino interactions from the pervading cosmic-ray background. Rejection of such background is often possible using traditional cut-based selections, but this typically requires the prior use of computationally expensive reconstruction algorithms. This work demonstrates an alternative approach of using a 3D submanifold sparse convolutional network trained on low-level information from the scintillation light signal of interactions inside LAr-TPCs. This technique is applied to example simulations from ICARUS, the far detector of the short baseline neutrino program at Fermilab. The results of the network, show that cosmic background is reduced by up to 76.3% whilst neutrino interaction selection efficiency remains over 98.9%. We further present a way to mitigate potential biases from imperfect input simulations by applying domain adversarial neural networks (DANNs), for which modified simulated samples are introduced to imitate real data and a small portion of them are used for adversarial training. A series of mock-data studies are performed and demonstrate the effectiveness of using DANNs to mitigate biases, showing neutrino interaction selection efficiency performances significantly better than that achieved without the adversarial training.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

BitGNN: Unlocking the Performance Potential of Binary Graph Neural Networks on GPUs

Graph Neural Networks (GNNs) have shown compelling results in many graph-based learning tasks. They are, however, time-consuming. Recent work has shown a promising direction in improving GNN speed and shrinking the size — network binarization, which binarizes network values and operations. Prior work, however, mainly focused on algorithm designs, leaving it open on how to fully materialize the performance potential. This work fills the gap by proposing techniques to best map binary GNNs and their computations to fit the nature of bit manipulations, optimizations and algorithms to maximize BSpMM kernel efficiency, and solutions to other factors influencing the end-to-end time on GPUs. Results on real-world graphs show that the proposed techniques outperform state of-the-art binary GNN implementations by 21-67× with little accuracy loss.

Chen, Jou-An↗
DOE OSTI2021

Deep Learning with Quantized Neural Networks for Gravitational-wave Forecasting of Eccentric Compact Binary Coalescence

We present the first application of deep learning forecasting for binary neutron stars, neutron star–black hole systems, and binary black hole mergers that span an eccentricity range e ≤ 0.9. In this work, we train neural networks that describe these astrophysical populations, and then test their performance by injecting simulated eccentric signals in advanced Laser Interferometer Gravitational-Wave Observatory (LIGO) noise available at the Gravitational Wave Open Science Center to (1) quantify how fast neural networks identify these signals before the binary components merge; (2) quantify how accurately neural networks estimate the time to merger once gravitational waves are identified; and (3) estimate the time-dependent sky localization of these events from early detection to merger. Our findings show that deep learning can identify eccentric signals from a few seconds (for binary black holes) up to tens of seconds (for binary neutron stars) prior to merger. A quantized version of our neural networks achieves 4× reduction in model size, and up to 2.5× inference speedup. These novel algorithms may be used to facilitate time-sensitive multimessenger astrophysics observations of compact binaries in dense stellar environments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Using discrete Bayesian networks for diagnosing and isolating cross-level faults in HVAC systems

Fault detection and diagnosis (FDD) technologies are critical to ensure satisfactory building performance, such as reducing energy wastes and negative impacts on occupant comfort and productivity. Existing FDD technologies mainly focus on component-level FDD solutions, which could lead to mis-diagnosis of cross-level faults in heating, ventilating, and air-conditioning (HVAC) systems. Cross-level faults are those faults that occur in one component or subsystem, but cause operational abnormalities in other components or subsystems, and result in a building level performance degradation. How to effectively diagnose the root cause of a cross-level fault is the focus of this study. Here, this paper presents a novel discrete Bayesian Network (DisBN)-based method for diagnosing cross-level faults in an HVAC system commonly used in commercial buildings. A two-level DisBN structure model is developed in this study. The parameters used in the DisBN model are obtained either from expert knowledge or through machine-learning strategies from normal system operation data. Meanwhile, the probability parameters are discretized to incorporate the uncertainties associated with typical expert knowledge. Thus, the developed DisBN method addresses the challenges many other BN based FDD methods face, i.e., the lack of fault data for BN parameter training. The developed DisBN represents causal relationships between a fault and its cross-level system impacts (i.e., fault symptoms or fault indicators) by considering how fault impacts propagate across different levels in an HVAC system. A weather and schedule information-based Pattern Matching (WPM) method is employed to automatically create WPM baseline data sets for each incoming real time snapshot data from the building systems. Consequently, BN inference and real-time diagnostics are achieved by comparing incoming snapshot data and the WPM baseline data set. The proposed method is evaluated using experimental fault data collected in a campus building. Fault diagnosis results demonstrate that the WPM-DisBN method is effective at locating the root causes of cross-level faults in an HVAC system.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Machine learning inversion of interatomic force constants from single-crystal inelastic neutron scattering

Atomic vibrations govern many macroscopic properties of materials, but experiments to comprehensively probe them remain challenging. Inelastic neutron scattering (INS) is a powerful technique to map phonon dispersions in crystals, especially when leveraging modern time-of-flight (ToF) spectrometers with large detectors. However, efficiently and robustly extracting interatomic force constants (FCs) parameterizing phonon dynamics from experimental spectra remains a bottleneck due to the complexity and high dimensionality of ToF INS datasets. Here, we present a machine learning approach for the direct inversion of FCs from single-crystal INS measurements. The framework leverages synthetic training data generated using universal machine-learned force fields and an efficient physics-based forward model. We benchmark two neural architectures–one emphasizing structured latent representation learning and the other direct, supervised spectral regression–across simulated datasets for two materials under idealized and noisy conditions. The latent-representation model is subsequently applied to experimental single-crystal INS data on germanium. The model is shown to reproduce FCs derived from both first-principles simulations and from iterative optimization, and furthermore achieves reliable inference even from sparse, single-orientation measurements representing short data acquisitions. Analysis of the learned latent space reveals semantically continuous and physically interpretable encodings that support strong cross-domain generalization. By bridging theoretical and experimental domains, we establish a path toward rapid inversion of experimental spectra and data-driven interpretation of temperature-dependent lattice dynamics.

42 ENGINEERING↗
DOE OSTI2022

Fast emulation of quantum three-body scattering

Here, we develop a class of emulators for solving quantum three-body scattering problems. They are based on combining the variational method for scattering observables and the recently proposed eigenvector continuation concept. The emulators are first trained by the exact scattering solutions of the governing Hamiltonian at a small number of points in its parameter space, and then employed to make interpolations and extrapolations in that space. Through a schematic nuclear-physics model with finite-range two and three-body interactions, we demonstrate the emulators to be extremely accurate and efficient. The computing time for emulation is on the scale of milliseconds (on a laptop), with relative errors ranging from 10 –13 to 10 –4 depending on the case. The emulators also require little memory. We argue that these emulators can be generalized to even more challenging scattering problems. Furthermore, this general strategy may be applicable for building the same type of emulators in other fields, wherever variational methods can be developed for evaluating physical models.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Novel Data Driven Noise Emulation Framework using Deep Neural Network for Generating Synthetic PMU Measurements

Sensors play a critical role in supporting day-to-day grid operations and they are essential to operator’s decision-making process. Furthermore, sensors and sensor behaviors need to be emulated with grid simulations to perform modeling studies and to design cutting edge power systems applications. Ensuring the accurate behavior of these applications requires accurate emulation of sensors and pertinent signals. However, most grid simulators and modeling tools assume either zero error scenarios or simplistic noise models that may not always correlate to real-world sensors. To address the above issue, this work presents an initial study on the noise characteristics of phasor measurement units (PMUs), along with models for recreating their unique noise signatures. The proposed methods (both analytical and machine-learning-based) provide a substantial increase in a sensor’s model fidelity, a feature that can be leveraged by an end-user application to yield more accurate system representations. The proposed methods were then applied to micro PMU data from the EPFL microgrid campus to extract sensor noise profiles. This data was used to train a deep learning model, which was tested to emulate the noise characteristics present in actual signals. Based on the observed results and the employed data-driven methodology, the proposed methods may be adapted to replicate the behavior of other grid sensors and power new applications capable of detecting sensor degradation and eventual device failures in near real-time.

PMU, noise emulation, synthetic measurements, deep↗
DOE OSTI2024

Flexible silicon photonic architecture for accelerating distributed deep learning

The increasing size and complexity of deep learning (DL) models have led to the wide adoption of distributed training methods in datacenters (DCs) and high-performance computing (HPC) systems. However, communication among distributed computing units (CUs) has emerged as a major bottleneck in the training process. In this study, we propose Flex-SiPAC, a flexible silicon photonic accelerated compute cluster designed to accelerate multi-tenant distributed DL training workloads. Flex-SiPAC takes a co-design approach that combines a silicon photonic hardware platform with a tailored collective algorithm, optimized to leverage the unique physical properties of the architecture. The hardware platform integrates a novel wavelength-reconfigurable transceiver design and a micro-resonator-based wavelength-reconfigurable switch, enabling the system to achieve flexible bandwidth steering in the wavelength domain. The collective algorithm is designed to support reconfigurable topologies, enabling efficient all-reduce communications that are commonly used in DL training. The feasibility of the Flex-SiPAC architecture is demonstrated through two testbed experiments. First, an optical testbed experiment demonstrates the flexible routing of wavelengths by shuffling an array of input wavelengths using a custom-designed spatial-wavelength selective switch. Second, a four-GPU testbed running two DL workloads shows a 23% improvement in job completion time compared to a similarly sized leaf-spine topology. We further evaluate Flex-SiPAC using large-scale simulations, which show that Flex-SiPAC is able to reduce the communication time by 26% to 29% compared to state-of-the-art compute clusters under representative collective operations.

Wu, Zhenguo (ORCID:0000000322847985)↗
DOE OSTI2025

PickerXL, A Large Deep Learning Model to Measure Arrival Times from Noisy Seismic Signals

Precisely measuring seismic arrival times is a labor-intensive task but is critical for both earthquake monitoring and subsurface imaging. Recently published deep learning models have demonstrated superior performance compared to traditional automatic approaches for picking arrival times. Although existing deep learning models have shown promising results, further advancements are necessary as their performance is not yet satisfactory especially when applied to new regions and station networks. Increasing model size has led to improved performance in other machine learning applications. Here, we aimed to investigate whether enlarging deep learning models can increase performance on accepted benchmarks. We trained three models of varying sizes, small (1X), medium (4X), and large (16X), using globally distributed local and regional earthquake signals and background noise waveforms from a benchmark dataset, Stanford Earthquake Dataset. Our results indicate that the largest model (PickerXL) outperforms both the smaller models and Seisbench implementation of the PhaseNet model, which has the same number of parameters as our small model. The PickerXL model’s enhanced capacity to extract complex patterns from seismograms contributes to its superior arrival picking abilities compared to the smaller model.

Chai, Chengping [Oak Ridge National Laboratory (OR↗
DOE OSTI2023

Attack Surface Analysis of the Digital Twins interface with Advanced Sensor and Instrumentation Interfaces: Cyber Threat Assessment and Attack Demonstration for Digital Twins in Advanced Reactor Architectures

A digital twin is a virtual representation of a physical system or object using real-time data that can predict and analyze how the system or object performs. This relatively new technology can be applied to the field of nuclear power generation, to aid in the design and development of new nuclear power plants and reduce operation costs using predictive maintenance and other data analytical methods. While there are already companies utilizing simulation software to train operators and technicians in the nuclear industry, some are now transitioning to utilizing their existing technology, software, and methods to develop digital twin solutions for the next generation of nuclear power plants, offering their services to utilities and government organizations around the world.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

In‐operando thermal transport characterization of moving particle bed heat exchanger

The project aimed to develop modulated photothermal radiometry (MPR) for conducting in-operando measurements of thermal transport in moving particle bed heat exchangers (MPBE) located at Sandia National Laboratory (SNL). The plan was to initially conduct lab-scale measurements on a 1-meter-long particle channel at UCSD, followed by in- operando measurements at SNL on a 20 kWth MPBE prototype built for this project, and if time permitting, potential final deployment on SNL’s 1 MWth G3P3 MPBE. The first two goals were successfully achieved but due to changes with commissioning timeline of the G3P3 MPBE, in-operando measurements on the G3P3 MPBE could not be conducted. The measurement campaigns have resulted in some promising results and the MPR tool has been handed over to trained SNL personnel for future use. MPR measurements on UCSD’s 1-meter-long particle channel showed a peculiar length dependence of the particle bed thermal conductivity and near-wall thermal resistance. The bed thermal conductivity increased, and the near-wall thermal resistance decreased, with increasing distance from the channel inlet, both eventually reaching a plateau. This was found to be due to hydrostatic pressure screening by the channel walls, known as the “Janssen effect”, and is the first experimental study on its role in granular heat transfer. A 20 kWth small-scale MPBE prototype, the first of its kind with 3 mm particle channels and 2 mm sCO2 plates, was designed and manufactured by Vacuum Process Engineering (VPE) for this project. This MPBE prototype was installed in SNL’s HEX test stand and the MPR tool was integrated into the test stand for in-operando measurements. Several modifications have been made to the MPR setup for facile installation and easy operation at SNL. Measurements with Carbo HSP40/70 particles were conducted with various particle and sCO2 temperatures and flow rates. Further, apart from regular MPBE operation, fault scenarios such as plate warping, particle stagnation, and choking were also detected and MPR showed excellent sensitivity to the detection of such conditions. Through the course of this project, MPR has evolved from a sensitive laboratory technique to a rugged industrial tool with various applications in high temperature measurements and diagnostics.

14 SOLAR ENERGY↗