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 1,117 records · Page 62
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Multivariate Time Series Anomaly Detection with Few Positive Samples

Given the scarcity of anomalies in real-world applications, the majority of literature has been focusing on modeling normality. The learned representations enable anomaly detection as the normality model is trained to capture certain key underlying data regularities under normal circumstances. In practical settings, particularly industrial time series anomaly detection, we often encounter situations where a large amount of normal operation data is available along with a small number of anomaly events collected over time. This practical situation calls for methodologies to leverage these small number of anomaly events to create a better anomaly detector. In this paper, we introduce two methodologies to address the needs of this practical situation and compared them with recently developed state of the art techniques. Our proposed methods anchor on representative learning of normal operation with autoregressive (AR) model along with loss components to encourage representations that separate normal versus few positive examples. We applied the proposed methods to two industrial anomaly detection datasets and demonstrated effective performance in comparison with approaches from literature. Our study also points out additional challenges with adopting such methods in practical applications.

Xue, Feng↗
DOE OSTI2020

Implementing Mixed Reality Tools to Support Mission Delivery at Hanford - 20520

Mission Support Alliance's (MSA) Public Works and Information Systems organizations have been working with Microsoft Corporation to develop and implement mixed reality operational solutions using their HoloLens technology. This collaborative effort is intended to drive innovative solutions and significantly improve MSA's efficiency in performing work at the Hanford Site. HoloLens, a mixed reality tool, is a commercial off-the-shelf technology that combines a head-mounted viewing screen with applications to help people and organizations learn, communicate, and collaborate more effectively through the use of mixed reality. https://microsoft.com/en-us/hololens MSA's initial pilot successfully demonstrated its ability to use mixed reality utilizing HoloLens' advanced features in a variety of ways, illustrating the potential efficiency gains previously stated as goals of this pilot. These achievements include the development of proprietary technology that allows synchronization between a HoloLens device and a mobile device or alternate global positioning system (GPS) device. This synchronization enables the HoloLens device to access real time GPS location data anywhere on the Hanford Site. This real time location data, combined with MSA's improved Hanford Geographical Information System (GIS) data, enables the mixed reality pilot application to identify underground utility systems as well as related utility attributes. This ability will aid utility workers in several areas, including excavation activities, the placement of large cranes, identifying interactions and conflicts with future underground utility placements and, eventually, recognizing and de-conflicting tank waste transfer system valve alignments. The HoloLens mixed reality application is also able to provide full or small scale holographic images of a facility's digital twin. This feature will allow MSA's Water and Sewer Utilities team to train and validate procedures associated with the new water treatment facility virtually, before it is constructed. This reduces the amount of time it takes to complete the training and procedure validation, and accelerates the overall construction schedule. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2021

Three-dimensional coherent X-ray diffraction imaging via deep convolutional neural networks

Abstract As a critical component of coherent X-ray diffraction imaging (CDI), phase retrieval has been extensively applied in X-ray structural science to recover the 3D morphological information inside measured particles. Despite meeting all the oversampling requirements of Sayre and Shannon, current phase retrieval approaches still have trouble achieving a unique inversion of experimental data in the presence of noise. Here, we propose to overcome this limitation by incorporating a 3D Machine Learning (ML) model combining (optional) supervised learning with transfer learning. The trained ML model can rapidly provide an immediate result with high accuracy which could benefit real-time experiments, and the predicted result can be further refined with transfer learning. More significantly, the proposed ML model can be used without any prior training to learn the missing phases of an image based on minimization of an appropriate ‘loss function’ alone. We demonstrate significantly improved performance with experimental Bragg CDI data over traditional iterative phase retrieval algorithms.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS1988

Software engineering and Ada (Trademark) training: An implementation model for NASA

The choice of Ada for software engineering for projects such as the Space Station has resulted in government and industrial groups considering training programs that help workers become familiar with both a software culture and the intricacies of a new computer language. The questions of how much time it takes to learn software engineering with Ada, how much an organization should invest in such training, and how the training should be structured are considered. Software engineering is an emerging, dynamic discipline. It is defined by the author as the establishment and application of sound engineering environments, tools, methods, models, principles, and concepts combined with appropriate standards, guidelines, and practices to support computing which is correct, modifiable, reliable and safe, efficient, and understandable throughout the life cycle of the application. Neither the training programs needed, nor the content of such programs, have been well established. This study addresses the requirements for training for NASA personnel and recommends an implementation plan. A curriculum and a means of delivery are recommended. It is further suggested that a knowledgeable programmer may be able to learn Ada in 5 days, but that it takes 6 to 9 months to evolve into a software engineer who uses the language correctly and effectively. The curriculum and implementation plan can be adapted for each NASA Center according to the needs dictated by each project.

Legrand, Sue↗
DOE OSTI2022

Robust Event Classification Using Imperfect Real-world PMU Data

Here, this paper studies robust event classification using imperfect real-world phasor measurement unit (PMU) data. By analyzing the real-world PMU data, we find it is challenging to directly use this dataset for event classifiers due to the low data quality observed in PMU measurements and event logs. To address these challenges, we develop a novel machine learning framework for training robust event classifiers, which consists of three main steps: data preprocessing, fine-grained event data extraction, and feature engineering. Specifically, the data preprocessing step addresses the data quality issues of PMU measurements (e.g., bad data and missing data); in the fine-grained event data extraction step, a model-free event detection method is developed to accurately localize the events from the inaccurate event timestamps in the event logs; and the feature engineering step constructs the event features based on the patterns of different event types, in order to improve the performance and the interpretability of the event classifiers. Based on the proposed framework, we develop a workflow for event classification using the real-world PMU data streaming into the system in real time. Using the proposed framework, robust event classifiers can be efficiently trained based on many off-the-shelf lightweight machine learning models. Numerical experiments using the real-world dataset from the Western Interconnection of the U.S power transmission grid show that the event classifiers trained under the proposed framework can achieve high classification accuracy while being robust against low-quality data.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
NASA NTRS1985

Actions for productivity improvement in crew training

Improvement of the productivity of astronaut crew instructors in the Space Shuttle program and beyond is proposed. It is suggested that instructor certification plans should be established to shorten the time required for trainers to develop their skills and improve their ability to convey those skills. Members of the training cadre should be thoroughly cross trained in their task. This provides better understanding of the overall task and greater flexibility in instructor utilization. Improved facility access will give instructors the benefit of practical application experience. Former crews should be integrated into the training of upcoming crews to bridge some of the gap between simulated conditions and the real world. The information contained in lengthy and complex training manuals can be presented more clearly and efficiently as computer lessons. The illustration, animation and interactive capabilities of the computer combine an effective means of explanation.

Miller, G. E.↗
NASA NTRS2005

Using EMG to anticipate head motion for virtual-environment applications

In virtual environment (VE) applications, where virtual objects are presented in a see-through head-mounted display, virtual images must be continuously stabilized in space in response to user's head motion. Time delays in head-motion compensation cause virtual objects to "swim" around instead of being stable in space which results in misalignment errors when overlaying virtual and real objects. Visual update delays are a critical technical obstacle for implementing head-mounted displays in applications such as battlefield simulation/training, telerobotics, and telemedicine. Head motion is currently measurable by a head-mounted 6-degrees-of-freedom inertial measurement unit. However, even given this information, overall VE-system latencies cannot be reduced under about 25 ms. We present a novel approach to eliminating latencies, which is premised on the fact that myoelectric signals from a muscle precede its exertion of force, thereby limb or head acceleration. We thus suggest utilizing neck-muscles' myoelectric signals to anticipate head motion. We trained a neural network to map such signals onto equivalent time-advanced inertial outputs. The resulting network can achieve time advances of up to 70 ms.

Clinical Trial↗
NASA NTRSDate not supplied

EVA Planning: Using Neutral Buoyancy Laboratory (NBL) Training to Predict in-Flight Energy Expenditure

Metabolic rate (“met rate”) is the amount of energy expended over a period of time and is influenced by many factors including body composition, level of physical activity, resting metabolic rate, sex, age, and food intake. Met rate is measured during Extravehicular Activity (EVA) training at the Neutral Buoyancy Laboratory (NBL) and during in-flight EVAs through indirect calorimetry, calculating energy expenditure from respiratory measurements of O 2 consumption and/or CO 2 production. During Extravehicular Activity (EVA) planning, metabolic cost is important to consider and is used to inform EVA duration based on spacesuit consumables associated with life support systems. Currently, NBL and previous ISS EVA met rate data for specified crewmembers are utilized to predict in-flight EVA metabolic costs based on a proposed EVA timeline. Timeline data collected during training is used to relate met rates to specific EVA activities, which are in turn assigned to more generalized EVA task categories, categorizing by both task type and restraint type. EVA task categories include EVA Setup/Cleanup, Worksite Setup/Cleanup, Cable Routing, Bolts, Fluid Connectors, Electrical Connectors, R&R Work, Miscellaneous Work, Incapacitated Crew Rescue (being rescued or performing), Assisted Crew Rescue (being assisted or performing), and Translation. Restraint types consist of Free-Float, Body Restraint Tether (BRT), Articulating Portable Foot Restraint (APFR), and Space Station Remote Manipulator System (SSRMS). From a crewmember’s historical data, individualized 10th, 50th and 90th percentile met rate estimates are generated for each task category and used to estimate the proposed EVA timeline metabolic cost. In-flight metabolic data (“As-Executed”) from recent ISS US EVAs 85-88 (totaling eight EVA crewmember met rates) was compared with their predicted metabolic cost (“As-Planned”) to evaluate the accuracy of the current met rate estimation method. Across the four EVAs, As-Executed Cumulative EVA Total Metabolic Cost (M = 5893.78 BTU, SD = 816.60) was not significantly different compared to As-Planned Cumulative EVA Total Metabolic Cost (M = 6106.60 BTU, SD = 826.62; t(7) = 0.751 , p = .477). Though not a significant difference, generally, As-Planned total estimates were slightly higher than As-Executed total metabolic cost. Relative Error for Cumulative EVA Total Metabolic Cost ranged from -33% to 14.7%, depending on the crewmember and EVA. When comparing As-Planned to A-Executed EVA task categories for Bolts, Electrical Connectors, EVA Cleanup, EVA Setup, Miscellaneous Work, Translation, Worksite Cleanup, and Worksite Setup during these EVAs, no significant differences were observed, however, there was a significant difference in As-Planned (M = 783.71 BTU, SD = 408.88) compared to As-Executed (M = 608.87 BTU, SD = 425.29) metabolic cost for the task category of Repair-and-Replace (R&R) Work (t(17) = 3.21 , p = .005). Looking closer within the R&R Work task category, As-Executed R&R Work with Free-Float restraint type (M = 706.17 BTU, SD = 352.18) was significantly less than As-Planned R&R Work with Free-Float restraint type (M= 887.96 BTU, SD = 381.42; t(12) = 2.59, p < .024). As-Executed R&R Work with SSRMS Restraint type (M = 355.89 BTU, SD = 534.65) was not significantly different from As-Planned values (M = 512.69 BTU, SD = 383.34; t(4) = 1.88, p = 0.132). These findings suggest that the energy expended performing R&R Work (Free-Float) is lower in flight than predicted. Accurate predictions of the metabolic cost of EVA are essential for planning and executing successful ISS EVAs. Overall, the current met rate prediction method is similar to actual in-flight values, slightly erring on the side of overestimation. Future work includes analysis of more historical in-flight EVA data to increase the power of the analysis, evaluating the NBL-ISS met rate conversion factor between NBL and ISS tasks, as well as exploring methods of substitution when crewmembers are missing prior task category data.

Lauren Cox↗
NASA NTRS2018

Remaining Flying Time Prediction Implementing Battery Prognostics Framework for Electric UAV's

In this paper the problem of building trust in the online safety prediction of an fixed wing small electric unmanned aerial vehicles (e-UAV) for remaining flying time is addressed. A series of flight tests are described to verify the performance of the remaining flying time prediction algorithm. The estimate of remaining flying time is used to activate an alarm when the predicted remaining time falls below a threshold of two minutes. This updates the pilot to transition to the landing sequence of the flight profile. A second alarm is activated when the battery state of charge (SOC) falls below a specified safety limit threshold. This SOC threshold is the point at which the battery energy reserve would no longer safely support enough aborted landing attempts. During the test flights, the motor system is operated with the same predefined timed airspeed profile for each test. To test the robustness of the developed prediction algorithm, partial tests were performed with and remaining were performed without a simulated power train fault. To simulate a partial power train fault in the e-UAV the pilot engages a resistor bank at a specified time during the test flight. The flying time prediction system is agnostic of the pilot's activation of the fault and must adapt to the vehicle's state. The time at which the limit threshold on battery SOC is reached, it is then used to measure the accuracy of the remaining flying time predictions. This is demonstrated through comparing results from two battery models being developed. Accuracy requirements for the alarms are considered and the results discussed.

Kulkarni, Chetan↗
DOE OSTI2024

First Search for Dark-Trident Processes Using the MicroBooNE Detector

This thesis presents a first search for dark-trident scattering in a neutrino beam using a data set corresponding to $7.2 \times 10^{20}$ protons on target taken with the MicroBooNE detector at Fermilab. Proton interactions in the neutrino target at the Main Injector produce $\pi^0$ and $\eta$ mesons, which could decay into dark matter (DM) particles via a dark photon $A^{\prime}$. A convolutional neural network is trained to identify interactions of the DM particles in the liquid-argon time projection chamber (LArTPC) exploiting its image-like reconstruction capability. In the absence of a DM signal, limits at the $90\%$ confidence level on the squared kinematic mixing parameter $\varepsilon^2$ as a function of the dark-photon mass in the range $10\le M_{A^\prime}\le 400$~MeV are provided. The limits cover previously unconstrained parameter space for the production of fermion or scalar DM particles $\chi$ for two benchmark models with mass ratios $M_{\chi}/M_{A^\prime}=0.6$ and $2$ and for dark fine-structure constants $0.1\le\alpha_D\le 1$.Date of Award1 Aug 2024Original languageEnglishAwarding InstitutionThe University of ManchesterSupervisorStefan Soldner-Rembold (Supervisor) & Justin Evans (Supervisor)

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS2000

Mid-Infrared Spectroscopy of Persistent Leonid Trains

The first infrared spectroscopy in the 3-13 micron region has been obtained of several persistent Leonid meteor trains with two different instrument types, one at a desert ground-based site and the other on-board a high-flying aircraft. The spectra exhibit common structures assigned to enhanced emissions of warm CH4, CO2, CO and H2O which may originate from heated trace air compounds or materials created in the wake of the meteor. This is the first time that any of these molecules has been observed in the spectra of persistent trains. Hence, the mid-IR observations offer a new perspective on the physical processes that occur in the path of the meteor at some time after the meteor itself has passed by. Continuum emission is observed also, but its origin has not yet been established. No 10 micron dust emission feature has been observed.

Russell, Ray W.↗
NASA NTRS1998

Effectiveness of an Expert System for Astronaut Assistance on a Sleep Experiment

Principal Investigator-in-a-Box ([PI]) is an expert system designed to train and assist astronauts with the performance of an experiment outside their field of expertise, particularly when contact with the Principal Investigators on the ground is limited or impossible. In the current case, [PI] was designed to assist with the calibration and troubleshooting procedures of the Neurolab Sleep and Respiration Experiment during the pre-sleep period of no ground contact. It displays physiological signals in real time during the pre-sleep instrumentation period, and alerts the astronauts when a poor signal quality is detected. Results of the first study indicated a beneficial effect of [PI] and training in reducing anomaly detection time and the number of undetected anomalies. For the in-flight performance, excluding the saturated signals, the expert system had an 84.2% detection accuracy, and the questionnaires filled out by the astronauts showed positive crew reactions to the expert system.

Heher, Dennis↗
NASA NTRS2015

Software Development: 3D Animations and Creating User Interfaces for Realistic Simulations

My fall 2015 semester was spent at the Lyndon B. Johnson Space Center working in the Integrated Graphics, Operations, and Analysis Laboratory (IGOAL). My first project was to create a video animation that could tell the story of OMICS. OMICS is a term being used in the field of biomedical science to describe the collective technologies that study biological systems, such as what makes up a cell and how it functions with other systems. In the IGOAL I used a large 23 inch Wacom monitor to draw storyboards, graphics, and line art animations. I used Blender as the 3D environment to sculpt, shape, cut or modify the several scenes and models for the video. A challenge creating this video was to take a term used in biomedical science and describe it in such a way that an 8th grade student can understand. I used a line art style because it would visually set the tone for what we thought was an educational style. In order to get a handle on the perspective and overall feel for the animation without overloading my workspace, I split up the 2 minute animation into several scenes. I used Blender's python scripting capabilities which allowed for the addition of plugins to add or modify tools. The scripts can also directly interact with the objects to create naturalistic patterns or movements. After collecting the rendered scenes, I used Blender's built-in video editing workspace to output the animation. My second project was to write software that emulates a physical system's interface. The interface was to simulate a boat, ROV, and winch system. Simulations are a time and cost effective way to test complicated data and provide training for operators without having to use expensive hardware. We created the virtual controls with 3-D Blender models and 2-D graphics, and then add functionality in C# using the Unity game engine. The Unity engine provides several essential behaviors of a simulator, such as the start and update functions. A framework for Unity, which was developed in the lab, provided a way to place the different widgets on the virtual console dock and have them resize correctly based on the window dimensions.. My task in this project was to create the controls and visualizations for the data coming in from the simulator for the boat portion of the project. I wrote a class for each control window to handle the functionality of that widget. I implemented 11 widgets that make up the ship portion of the simulator. The members of the lab were each masters of their craft and I'm glad I had the opportunity to learn from them. I learned to plan strategically so I could finish this project on time. I allotted time for storyboarding, development, and refinement. In regards to animating I learned to use modifiers like lattice, boolean and build deformers. I also learned how to animate with drivers, how to use the dope sheet, and how to use the graph editor. In coding I learned to limit the chances for bugs by privatizing functions that should be exclusive to their class. I learned how to use the GIT repository to commit, stash and pull the latest build. I learned a bit of everything because I had the chance to see the entire application development process from the artwork, to the implementation.

Gordillo, Orlando Enrique↗
DOE OSTI2024

Transfer learning for metamaterial design and simulation

Abstract We demonstrate transfer learning as a tool to improve the efficacy of training deep learning models based on residual neural networks (ResNets). Specifically, we examine its use for study of multi-scale electrically large metasurface arrays under open boundary conditions in electromagnetic metamaterials. Our aim is to assess the efficiency of transfer learning across a range of problem domains that vary in their resemblance to the original base problem for which the ResNet model was initially trained. We use a quasi-analytical discrete dipole approximation (DDA) method to simulate electrically large metasurface arrays to obtain ground truth data for training and testing of our deep neural network. Our approach can save significant time for examining novel metasurface designs by harnessing the power of transfer learning, as it effectively mitigates the pervasive data bottleneck issue commonly encountered in deep learning. We demonstrate that for the best case when the transfer task is sufficiently similar to the target task, a new task can be effectively trained using only a few data points yet still achieve a test mean absolute relative error of 3 % with a pre-trained neural network, realizing data reduction by a factor of 1000.

Peng, Rixi↗
DOE OSTI2024

Time-lapse seismic inversion for CO 2 saturation with SeisCO2Net: An application to Frio-II site

Seismic monitoring of geological CO 2 storage (GCS) involves highly nonlinear seismic inversion and petrophysical inversion, making it challenging to estimate CO 2 volume efficiently and detect possible early CO 2 leakages. Deep learning (DL) using convolutional neural networks (CNNs) has shown promise in solving highly nonlinear seismic inversion problems. However, direct estimation of CO 2 plume extent/saturation from time-lapse seismic gathers using DL is still underexplored, with no reported field applications to date. The investigation of field data is primarily hindered by scarcity of field data for neural network training. Other obstacles include highly nonlinear seismic-petrophysics inverse relationship, and presence of noise in field seismic data. We introduce SeisCO2Net, a deep CNN that predicts CO 2 saturation maps directly from time-lapse full waveform shot gathers. For training, we use site-specific geological information, fluid flow physics, rock physics, and seismic modeling to generate synthetic datasets that closely resemble the CO 2 storage site. Synthetic tests show promising results, inspiring us to apply SeisCO2Net's trained weights on field data collected at Frio-II GCS site by leveraging transfer learning principles. As reference, we compare SeisCO2Net's predicted CO 2 saturation maps with results obtained from physics-based inversion. Our analyses show both methods display similar CO 2 plume shapes, reasonable CO 2 plume characteristics, and comparable saturation values. Our results suggest pre-training CNNs on physics-informed synthetic datasets and then applying the learned weights to field data is a viable approach to estimating field CO 2 saturation. This method effectively addresses the scarcity of field training data, thus encouraging the feasibility of long-term GCS monitoring.

58 GEOSCIENCES↗
NASA NTRS2010

Blended Training for Combat Medics

Bleeding from extremity wounds is the number one cause of preventable death on the battlefield and current research stresses the importance of training in preparing every Soldier to use tourniquets. HapMed is designed to provide tourniquet application training to combat medics and Soldiers using a blended training solution encompassing information, demonstration, practice, and feedback. The system combines an instrumented manikin arm, PDA, and computer. The manikin arm provides several training options including stand-alone, hands-on skills training in which soldiers can experience the actual torque required to staunch bleeding from an extremity wound and be timed on tourniquet application. This is more realistic than using a block of wood to act as a limb, which is often how training is conducted today. Combining the manikin arm with the PDA allows instructors to provide scenario based training. In a classroom or field setting, an instructor can specify wound variables such as location, casualty size, and whether the wound is a tough bleed. The PDA also allows more detailed feedback to be provided. Finally, combining the manikin arm with game-based technologies, the third component, provides opportunities to build knowledge and to practice battlefield decision making. Not only do soldiers learn how to apply a tourniquet, but when to apply a tourniquet in combat. The purpose of the paper is to describe the learning science underlying the design of HapMed, illustrate the training system and ways it is being expanded to encompass other critical life-saving tasks, and report on feedback received from instructors and trainees at military training and simulation centers.

Fowlkes, Jennifer↗
DOE OSTI2024

Multimodal sensor fusion for real-time standoff estimation in directed energy deposition

In Laser Powder-based Direct Energy Deposition (LP-DED) systems, achieving consistency, precision and quality of produced parts requires tight control over printing parameters. One of the critical parameters is the standoff distance. Maintaining an optimal standoff height is crucial for achieving correct laser power density and powder catchment efficiency, as both laser and powder streams are focused at this distance. Here, this study introduces a novel approach using multimodal sensor fusion to predict standoff height in real-time. The proposed system integrates two low-profile, cost-effective sensors: an RGB coaxial camera and a high frequency and high dynamic range microphone. By utilizing a simple fully connected neural network, trained on a limited dataset, data fusion of these sensors allowed for the real-time prediction of the standoff height. The results demonstrate high resolution and accuracy of the predictions across multiple geometries and a wide range of standoff heights. This approach offers a simple, and cost-effective solution for real-time standoff height monitoring and lays the groundwork for future integration into commercial LP-DED systems.

42 ENGINEERING↗
NASA NTRS1975

Methods of body orientation in space in the absence of support under weightless conditions

The experience accumulated in training subjects in methods of body orientation in space indicates the necessity of clear planning of the training process. After theoretical familiarization with the principles of body orientation in space and reviewing training films, practical mastery of the body orientation methods begins with working out of the individual elements on the Zhukovskiy stool. Then, the correctness and sequence of movements are carefully mastered in water, and the motor skills are then reinforced under time deficit conditions, on the vaulting bars, trampolines, and, in the concluding stage of training, the methods of orienting the body in space in weightlessness are worked out in laboratory-aircraft, with and without the spacesuit and with and without a load.

Yeremin, A. V.↗