Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Online Training”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

End-to-end online performance data capture and analysis for scientific workflows

With the increased prevalence of employing workflows for scientific computing and a push towards exascale computing, it has become paramount that we are able to analyze characteristics of scientific applications to better understand their impact on the underlying infrastructure and vice-versa. Such analysis can help drive the design, development, and optimization of these next generation systems and solutions. Here, we present the architecture, integrated with existing well-established and newly developed tools, to collect online performance statistics of workflow executions from various, heterogeneous sources and publish them in a distributed database (Elasticsearch). Using this architecture, we are able to correlate online workflow performance data, with data from the underlying infrastructure, and present them in a useful and intuitive way via an online dashboard. We have validated our approach by executing two classes of real-world workflows, both under normal and anomalous conditions. The first is an I/O-intensive genome analysis workflow; the second, a CPU- and memory-intensive material science workflow. Based on the data collected in Elasticsearch, we are able to demonstrate that we can correctly identify anomalies that we injected. The resulting end-to-end data collection of workflow performance data is an important resource of training data for automated machine learning analysis.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2011

ARC+(Registered Trademark) and ARC PC Welding Simulators: Teach Welders with Virtual Interactive 3D Technologies

123 Certification Inc., a Montreal based company, has developed an innovative hands-on welding simulator solution to help build the welding workforce in the most simple way. The solution lies in virtual reality technology, which has been fully tested since the early 90's. President and founder of 123 Certification Inc., Mr. Claude Choquet Ing. Msc. IWE. acts as a bridge between the welding and the programming world. Working in these fields for more than 20 years. he has filed 12 patents world-wide for a gesture control platform with leading edge hardware related to simulation. In the summer of 2006. Mr Choquet was proud to be invited to the annual IIW International Weld ing Congress in Quebec City to launch the ARC+ welding simulator. A 100% virtual reality system and web based training center was developed to simulate multi process. multi-materiaL multi-position and multi pass welding. The simulator is intended to train welding students and apprentices in schools or industries. The welding simulator is composed of a real welding e[eetrode holder (SMAW-GTAW) and gun (GMAW-FCAW). a head mounted display (HMD), a 6 degrees of freedom tracking system for interaction between the user's hands and head. as well as external audio speakers. Both guns and HMD are interacting online and simultaneously. The welding simulation is based on the law of physics and empirical results from detailed analysis of a series of welding tests based on industrial applications tested over the last 20 years. The simulation runs in real-time, using a local logic network to determine the quality and shape of the created weld. These results are based on the orientation distance. and speed of the welding torch and depth of penetration. The welding process and resulting weld bc.1d are displayed in a virtual environment with screenplay interactive training modules. For review. weld quality and recorded process values can be displayed and diagnosed after welding. To help in the le.tming process, a learning curve for each student and each Virtual Welding Class'" can be plotted, for an instructor's review or a required third party evaluation.

Choquet, Claude↗
DOE OSTI2025

Myco-Ed: Mycological curriculum for education and discovery

Fungi are important and hyperdiverse organisms, yet chronically understudied. Most fungal clades have no reference genomes, impeding our understanding of their ecosystem functions and use as solutions in health and biotechnology. Also, opportunities for training in fungal biology and genomics are lacking, creating a bottleneck that hinders the recruitment and cultivation of a talented future mycological workforce. To address these issues, we developed Myco-Ed, an educational program offering training and scientific contributions through genome sequencing and analysis. Myco-Ed empowers students to pursue careers in fungal biology while improving fungal resources. Myco-Ed has been piloted at 12 institutions (15 classrooms) ranging from online e-Campuses to R1 universities, resulting in hundreds of fungal observations and many new high-quality reference genomes.

Branco, Sara↗
DOE OSTI2024

Unpinning the Lynchpin: Development of a system to introduce redundancy at Department of Energy laboratories

The United States Department of Energy (DOE) laboratory system employs over 78,000 people and had a budget of $8.1 billion for Fiscal Year 2023 (FY23). This diverse workforce includes employees at all levels from operational support staff to senior level management. Together, the labs take on incredibly complex and unique technical challenges in an attempt to address fundamental National and global needs. These technical challenges often require specialized training such that only a single person is truly capable of completing specific tasks. With the modern competitive environment, and an aging workforce, the risk of technical capability loss is highly probable with potentially severe impact. An online survey of Lawrence Livermore National Laboratories (LLNL) employees, a typical DOE lab, found that of responders with greater than 10 years of experience, 100% of respondents felt burnt out, 40% felt they were not paid fairly, and 50% believe that not enough is done by leadership to retain them as an employee. Couple that to an aging workforce, and the risk of losing institutional knowledge due to single point of failure, or lynchpin, is significant. There is a need for a system which reduces the risk of operational and technical losses due to lynchpin employees leaving the organization. A mitigation of this risk would lead to the successful transfer of knowledge from one generation to the next, improving the rate of development and minimizing general risk of mistakes.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2026

Real-Time Detection of Charge Jumps in Superconducting Qubits with a Convolutional Neural Network

Ionizing radiation from cosmic rays and gammas can induce discontinuous jumps in the environmental charge of superconducting qubits (charge jumps), causing correlated errors that challenge fault-tolerant quantum computing while simultaneously providing a detection signature for quantum sensing applications. Current detection methods operate offline, introducing latency incompatible with in-the-loop qubit control. In this paper, an online detector of charge jumps for superconducting qubits, based on a dilated causal convolutional neural network (DCCNN) designed for in-the-loop deployment on the Quantum Instrumentation Control Kit (QICK) platform, is presented. The network is trained on synthetic Ramsey tomography scans generated from qubit templates measured at the Northwestern Experimental Underground Site (NEXUS) at Fermilab, and translated to FPGA firmware via hls4ml with ap_fixed$\langle 16,6 \rangle$ quantization, reaching a per-inference latency of $6.19 μ$s on the Zynq UltraScale+ RFSoC ZCU216. At this operating point the DCCNN matches the detection efficiency of the established offline $χ^2$ algorithm ($0.843 \pm 0.022$ vs. $0.866 \pm 0.020$ on $|Δq| \in [0.1, 0.5] e$ at matched false-positive rate), while requiring no per-qubit hyperparameter tuning. This shifts charge-jump detection from a post-hoc diagnostic to a control-loop primitive, enabling adaptive protocols that respond to radiation-induced events in situ, with applications to quantum-computing error mitigation and to the use of superconducting qubits as particle detectors.

Gaytan-Villarreal, Daniel [Carnegie Mellon U.]↗
DOE OSTI2022

Contextual Active Online Model Selection with Expert Advice

How can we collect the most useful labels to learn a model selection policy, when presented with arbitrary heterogeneous data streams? In this paper, we formulate this task as a contextual active model selection problem, where at each round the learner receives an unlabeled data point along with a context. The goal is to output the best model for any given context without obtaining an excessive amount of labels. In particular, we focus on the task of selecting pre-trained classifiers, and propose a contextual active model selection algorithm (CAMS), which relies on a novel uncertainty sampling query criterion defined on a given policy class for adaptive model selection. In comparison to prior art, our algorithm does not assume a globally optimal model. We provide rigorous theoretical analysis for the regret and query complexity under both adversarial and stochastic settings. Our experiments on several benchmark classification datasets demonstrate the algorithm’s effectiveness in terms of both regret and query complexity. Notably, to achieve the same accuracy, CAMS incurs less than 10% of the label cost when compared to the best online model selection baselines on CIFAR10.

Liu, Xuefeng↗
DOE OSTI2021

Emerging Frameworks for Advancing Scientific Workflows Research, Development, and Education

Lightning talks of the Workflows in Support of Large-Scale Science (WORKS) workshop are a venue where the workflow community (researchers, developers, and users) can discuss work in progress, emerging technologies and frameworks, and training and education materials. This paper summarizes the WORKS 2021 lightning talks, which cover four broad topics: (i) libEnsemble, a Python library to coordinate the concurrent evaluation of dynamic ensembles of calculations; (ii) Edu WRENCH, a set of online pedagogic modules that provides simulation-driven hands-on activity in the browser; (iii) VisDict, an envisioned visual dictionary framework that will translate terms, jargon, and concepts between research domains and workflow providers; and (iv) Pegasus Kickstart, a lightweight tool for capturing workflow tasks' performance, including performance metrics from Nvidia GPUs.

Casanova, Henri↗
DOE OSTI2025

Performance Evaluation of Vertical Federated Machine Learning Against Adversarial Threats on Wide-Area Control System: Preprint

Federated machine learning (FL) is gaining significant popularity to develop cybersecurity solutions in power grids because of its advanced capability to support decentralized data handing at local devices, its privacy preservation, and its low-bandwidth requirement. However, the evolving adversarial machine learning (AML) threats raise significant concerns for the cybersecurity of FL architectures. The FL-based split neural network (SplitNN) achieves high performance through the decentralized training of local neural network models while preserving data privacy across multiple entities. In this paper, we propose a methodology for evaluating the performance of a vertical FLbased anomaly detector against different types of AML attacks, including denial-of-service attacks, adversarial data injection attacks, and replay attacks on the trained local models deployed in the grid network. For a case study, we consider the modified IEEE 13-bus system, and we develop SplitNN-based binary and multiclass classification models to detect, locate, and identify different types of data integrity attacks on the volt-watt control with two pooling layers: maximum pooling and AvgPool. Our experimental results, computed through performance metrics, reveal that the severity of these AML attacks varies with the integrated pooling mechanism, the type of classification model, and the nature of the cyberattack. Further, the AML attacks negatively impacted the prediction time per sample for the pretrained SplitNN during the online testing.

adversarial threats↗
DOE OSTI2019

Formulation and Comparison of Two Real-Time Predictive Gear Shift Algorithms for Connected/Automated Heavy-Duty Vehicles

Our work investigates the problem of predictive gear scheduling for fuel consumption minimization in connected/automated heavy trucks. The literature highlights the fuel economy benefits of such predictive scheduling, but there is a need to optimize such scheduling online, in real time. To address this need, we begin by using dynamic programming (DP) to schedule gear shifting offline, in a manner that achieves a globally optimal Pareto tradeoff between the conflicting objectives of minimizing fuel consumption and shift frequency. The computational cost of DP is unfavorable for online implementation, but we present two algorithms addressing this issue. Both algorithms rely on the fact that in the Pareto limit where fuel consumption minimization is the sole objective, DP furnishes a simple static shift map . Our first algorithm trains a recurrent neural network to prune the shift schedule generated by this map. The second algorithm performs this pruning in a direct manner tailored to reduce the schedule's rain flow count. We simulate these algorithms for different drive cycles. Both algorithms achieve a reasonable tradeoff between fuel consumption and gear shift frequency. Yet, the rain flow count algorithm is both more effective in approaching the DP-based Pareto front and more computationally efficient.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2022

Real Time Predictive and Adaptive Hybrid Powertrain Control Development via Neuroevolution

The real-time application of powertrain-based predictive energy management (PrEM) brings the prospect of additional energy savings for hybrid powertrains. Torque split optimal control methodologies have been a focus in the automotive industry and academia for many years. Their real-time application in modern vehicles is, however, still lagging behind. While conventional exact and non-exact optimal control techniques such as Dynamic Programming and Model Predictive Control have been demonstrated, they suffer from the curse of dimensionality and quickly display limitations with high system complexity and highly stochastic environment operation. This paper demonstrates that Neuroevolution associated drive cycle classification algorithms can infer optimal control strategies for any system complexity and environment, hence streamlining and speeding up the control development process. Neuroevolution also circumvents the integration of low fidelity online plant models, further avoiding prohibitive embedded computing requirements and fidelity loss. This brings the prospect of optimal control to complex multi-physics system applications. The methodology presented here covers the development of the drive cycles used to train and validate the neurocontrollers and classifiers, as well as the application of the Neuroevolution process.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2021

National and Regional Initiatives to Promote Energy Efficiency and Renewable Energy Through State Energy Offices

The National Association of State Energy Officials (NASEO) worked with the U.S. Department of Energy’s (DOE) Office of Energy Efficiency and Renewable Energy (EERE), Weatherization and Intergovernmental Programs Office (WIP) over a ten year period to provide technical assistance, research and analyses, and enhanced coordination between DOE and the State and Territory Energy Offices. Over the life of the agreement, NASEO worked with WIP and the states to provide statewide strategic energy plan analyses and recommendations, including customized technical assistance to the states; peer to peer financing assistance via NASEO’s Financing Committee and supporting activities; training on core energy policies and programs, including dialogues and resources to support energy-air coordination, a training for new State Energy Office Directors, peer to peer exchange via a rural energy taskforce, and support to states to enhance buildings efficiency, home energy labeling, and technology deployment opportunities; support for State Energy Program metrics; and regional coordination via regional coordinators and peer exchange opportunities both in-person and online. Over the life of the project, the position and role of State Energy Offices within state government has changed, with now 80 percent of State Energy Office Directors serving as their Governor’s energy advisor, or reporting directly to their Governor’s energy advisor. This shift in stature has made it ever more crucial for State Energy Offices to receive timely and relevant technical assistance across a range of energy issue areas. NASEO, in collaboration with WIP, was able to deliver this technical assistance, and energy efficiency and renewable energy deployment has accelerated across the country. The State Energy Office Directors and their staff continue to engage in NASEO’s Committees – many of which were supported through this agreement – and have provided formal and informal feedback on the value of the programs supported through this award (e.g., reporting via survey increased understanding of their roles and technical assistance offerings provided by WIP and NASEO following the New Director Trainings). Moreover, resources developed through this award (e.g., Comprehensive State Energy Planning Guidelines, State Energy Loan Fund Database, Rural Data Resources for State Energy Planning and Programs, The Value of Adding home Energy Score to Low-Income Energy Efficiency Programs, etc.) have been cited by states as instrumental to their understanding of specific energy issue areas and in many cases led directly to enhanced program design within a state (e.g. Iowa citing their review of NASEO’s Comprehensive State Energy Planning Guidelines as a necessary first step in their planning process, later following may of the steps outlined in the guidelines). The priorities of the states and federal government have evolved over the last decade, with an increasing focus on climate mitigation and adaptation, equity impacts and considerations, energy security and resilience, and enhanced energy efficiency and renewable energy technology deployment, but the roots of these new priorities are based in the state and federal policies and programs, and research and analyses, that were supported in part through this award and other complementary initiatives. NASEO looks forward to continuing to support the states, and collaborate and coordinate with DOE, as we build on this important foundation in the years ahead.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

A Deep Learning Approach for In-Network Synchrophasor Missing Data Recovery Using Programmable Network Switches

Phasor measurement unit (PMU) networks deliver accurate and timely measurements, which is essential for managing today’s electric power systems. To ensure data quality and enhance the cyber-resilience of PMU networks against malicious attacks and data errors, this study presents an online PMU missing data recovery scheme by leveraging P4 programmable switches. The data plane incorporates a customized PMU protocol parser that abstracts the necessary payload data for recovery. Recovery processes are executed in the control plane using a pre-trained machine learning model. Both traditional and advanced ML models, such as transformer and TimeGPT, are explicitly employed for data prediction. This approach ensures rapid and precise data recovery. Performance evaluations focus on recovery speed and accuracy, using a real dataset from a campus microgrid. With 20% missing PMU data, the mean absolute percentage error for voltage magnitude is 0.0384%, and the phase angle error discrepancy is approximately 0.4064%.

Phasor Measurement Unit, Machine Learning, Program↗
NASA NTRS1967

The analysis of Gemini 7 EEG data by three different computer oriented methods

The problem in assessment of electroencephalograms has been the subject element in visual evaluations of such records as well as the need for extensive training (2 or 3 years) before suitable candidate is capable to perform this task. There have been variety of computer type analyses of EEG during the past ten years or more, and we are still searching for a reliable, inexpensive, online data processing procedure for this task. We are aware our Soviet colleagues are routinely using computer assessment of EEGs, using relatively inexpensive hardware. In our report we endeavor to show how four different assessment procedures (visual assessment, zero crossing technique, smoothing and peak counting technique, and Weibull Statistic) may be utilized, and indicate the advantages and disadvantages peculiar to each. We wish to point out that in the Gemini 7 EEG analog records supplied to the Henry Ford Hospital, Channel II (mid-line central to mid-line occipital electrodes) could not be assessed reliably after approximately 26 hours of flight. Channel I (left central to left occipital electrodes) provided the more assessable EEG up to approximately 54 hours of flight. These channels were compared during the first 26 hours of flight, and no significant difference was found by our various assessment procedures. We, therefore, have limited this report to the findings on the analyses of Channel I during its approximate 54 hours of flight.

Proctor, L. D.↗
NASA NTRS2010

Frequency Stabilization of DFB Laser Diodes at 1572 nm for Spaceborne Lidar Measurements of CO2

We report a fiber-based, pulsed laser seeder system that rapidly switches among 6 wavelengths across atmospheric carbon dioxide (CO2) absorption line near 1572.3 nm for measurements of global CO2 mixing ratios to 1-ppmv precision. One master DFB laser diode has been frequency-locked to the CO2 line center using a frequency modulation technique, suppressing its peak-to-peak frequency drifts to 0.3 MHz at 0.8 sec averaging time over 72 hours. Four online DFB laser diodes have been offset-locked to the master laser using phase locked loops, with virtually the same sub-MHz absolute accuracy. The 6 lasers were externally modulated and then combined to produce the measurement pulse train.

Numata, Kenji↗
DOE OSTI2022

First Plant Cell Atlas symposium report

The Plant Cell Atlas (PCA) community hosted a virtual symposium on December 9 and 10, 2021 on single cell and spatial omics technologies. The conference gathered almost 500 academic, industry, and government leaders to identify the needs and directions of the PCA community and to explore how establishing a data synthesis center would address these needs and accelerate progress. This report details the presentations and discussions focused on the possibility of a data synthesis center for a PCA and the expected impacts of such a center on advancing science and technology globally. Community discussions focused on topics such as data analysis tools and annotation standards; computational expertise and cyber-infrastructure; modes of community organization and engagement; methods for ensuring a broad reach in the PCA community; recruitment, training, and nurturing of new talent; and the overall impact of the PCA initiative. These targeted discussions facilitated dialogue among the participants to gauge whether PCA might be a vehicle for formulating a data synthesis center. The conversations also explored how online tools can be leveraged to help broaden the reach of the PCA (i.e., online contests, virtual networking, and social media stakeholder engagement) and decrease costs of conducting research (e.g., virtual REU opportunities). Major recommendations for the future of the PCA included establishing standards, creating dashboards for easy and intuitive access to data, and engaging with a broad community of stakeholders. The discussions also identified the following as being essential to the PCA's success: identifying homologous cell-type markers and their biocuration, publishing datasets and computational pipelines, utilizing online tools for communication (such as Slack), and user-friendly data visualization and data sharing. In conclusion, the development of a data synthesis center will help the PCA community achieve these goals by providing a centralized repository for existing and new data, a platform for sharing tools, and new analytical approaches through collaborative, multidisciplinary efforts. A data synthesis center will help the PCA reach milestones, such as community-supported data evaluation metrics, accelerating plant research necessary for human and environmental health.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2025

An Efficient GPU-Accelerated Multi-Source Global Fit Pipeline for LISA Data Analysis

The large-scale analysis task of deciphering gravitational wave signals in the LISA data stream will be difficult, requiring a large amount of computational resources and extensive development of computational methods. Its high dimensionality, multiple model types, and complicated noise profile require a global fit to all parameters and input models simultaneously. In this work, we detail our global fit algorithm, called “Erebor,” designed to accomplish this challenging task. It is capable of analysing current state-of-the-art datasets and then growing into the future as more pieces of the pipeline are completed and added. We describe our pipeline strategy, the algorithmic setup, and the results from our analysis of the LDC2A Sangria dataset, which contains Massive Black Hole Binaries, compact Galactic Binaries, and a parameterized noise spectrum whose parameters are unknown to the user. The Erebor algorithm includes three unique and very useful contributions: GPU acceleration for enhanced computational efficiency; ensemble MCMC sampling with multiple MCMC walkers per temperature for better mixing and parallelized sample creation; and special online updates to reversible-jump (or trans-dimensional) sampling distributions to ensure sampler mixing and accurate initial estimates for detectable sources in the data. We recover posterior distributions for all 15 (6) of the injected MBHBs in the LDC2A training (hidden) dataset. We catalog ∼12000 Galactic Binaries (∼8000 as high confidence detections) for both the training and hidden datasets. All of the sources and their posterior distributions are provided in publicly available catalogs.

LISA global fit↗
DOE OSTI2020

Machine Learning-Based PV Reserve Determination Strategy for Frequency Control on the WECC System: Preprint

Frequency control from Photovoltaic (PV) plants has great potential to address the frequency response challenge of the power system with high renewable penetration. However, using model-based approaches to determine the optimal PV headroom reserve requires significant online computation and is intractable for an interconnection level system. This paper proposes a machine learning based strategy, that is suitable for real-time operation, to determine the optimal PV reserve for frequency control. The proposed machine learning algorithm is trained and tested on 1,987 offline simulations of a 60% renewable penetration Western Electricity Coordinating Council (WECC) system. Furthermore, the proposed reserve determination strategy is applied on a realistic one-day operation profile of the WECC system and demonstrates over 40% PV headroom saving compared to a conservative approach. It is evident that the proposed strategy can efficiently and effectively determine the optimal PV frequency control reserve for realistic interconnection systems.

41 EE - Solar Energy Technologies Office (EE-4S)↗
DOE OSTI2024

Harnessing large language models’ zero-shot and few-shot learning capabilities for regulatory research

Abstract Large language models (LLMs) are sophisticated AI-driven models trained on vast sources of natural language data. They are adept at generating responses that closely mimic human conversational patterns. One of the most notable examples is OpenAI's ChatGPT, which has been extensively used across diverse sectors. Despite their flexibility, a significant challenge arises as most users must transmit their data to the servers of companies operating these models. Utilizing ChatGPT or similar models online may inadvertently expose sensitive information to the risk of data breaches. Therefore, implementing LLMs that are open source and smaller in scale within a secure local network becomes a crucial step for organizations where ensuring data privacy and protection has the highest priority, such as regulatory agencies. As a feasibility evaluation, we implemented a series of open-source LLMs within a regulatory agency’s local network and assessed their performance on specific tasks involving extracting relevant clinical pharmacology information from regulatory drug labels. Our research shows that some models work well in the context of few- or zero-shot learning, achieving performance comparable, or even better than, neural network models that needed thousands of training samples. One of the models was selected to address a real-world issue of finding intrinsic factors that affect drugs' clinical exposure without any training or fine-tuning. In a dataset of over 700 000 sentences, the model showed a 78.5% accuracy rate. Our work pointed to the possibility of implementing open-source LLMs within a secure local network and using these models to perform various natural language processing tasks when large numbers of training examples are unavailable.

Biochemistry & Molecular Biology↗