Search NASA⌕ Search

SEARCH · Search NASA

Results for “task”

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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
NASA NTRS2017

Workload and Performance in Air Traffic Control: Exploring the Influence of Levels of Automation and Variation in Task Demand

In air traffic control, task demand and workload have important implications for the safety and efficiency of air traffic. Task demand is dynamic, however, research on demand transitions and associated controller perception and performance is limited. In addition, there is a comparatively restricted understanding of the influence of task demand transitions on workload and performance, in association with automation. This study used an air traffic control simulation to investigate the influence of task demand transitions and two conditions of varying automation, on workload and efficiency-related performance. Findings showed that a both the direction of the task demand variation, and the amount of automation, influenced the relationship between workload and performance. Further research is needed to enhance understanding of demand transition and workload history effects on operator experience and performance, in both air traffic control and other safety-critical domains.

air traffic control↗
NASA NTRS2018

Multi-Operator Multi-UAV (MOMU) Control: Exploring the Influence of Sensor Tools and Playbook Task Delegation

New concepts of operations for Unmanned Aerial Vehicles (UAVs) will require a change from the current 2:1 operator to vehicle crew configuration. One particular control paradigm, largely driven by logistics, manpower, and training burdens, as well as the desire to force multiply, involves a single operator simultaneously managing multiple UAVs. This mode of operations has shown to significantly increase cognitive workload and decrease situation awareness, as operators are required to simultaneously attend to multiple sources of information. One potential way to mitigate potential drawbacks of multi-vehicle control by a single operator is to migrate to a multi-operator multi-UAV (MOMU) crew configuration, whereby M operators control N (> M) vehicles. This type of crew configuration can be organized in several ways to dynamically manage cognitive workload, match operator qualifications and skills to mission requirements, increase utilization of available assets, and thereby achieve maximum force multiplication. The present experiment examined task performance in a simulated MOMU environment and evaluated the potential benefits of sensor management aids ("Tools") as well as integrated sensor and flight automation ("Plays") compared to a fully manual condition ("Manual"). Tools support the operator by facilitating rapid understanding and management of sensor information, while the Plays support the operator by offloading/automating subtasks. Six pairs of participants were recruited for this study and tasked with sharing a pool of UAVs in order to conduct reconnaissance, surveillance, and target acquisition (RSTA) missions in adjacent Areas of Operation (AOs). Participants were given four tasks to accomplish, in order of priority: 1) prosecute High Value Targets (HVTs); 2) identify / track targets (military vehicles); 3) identify / mark civilian vehicles; and 4) respond to chat messages. Performance on the mission tasks was measured in terms of accuracy and reaction time. A composite mission score was also calculated using a payoff matrix that weighted each task according to priority. The results indicate that Playbook demonstrated better performance overall with higher accuracy rates and the highest composite score compared to Tools and Manual. The implications of these results to supporting future MOMU concepts of operations is discussed.

Playbook↗
NASA NTRS2016

Task Demand Variation in Air Traffic Control: Implications for Workload, Fatigue, and Performance

In air traffic control, task demand and workload have important implications for the safety and efficiency of air traffic, and remain dominant considerations. Within air traffic control, task demand is dynamic. However, research on demand transitions and subsequent controller perception and performance is limited. This research uses an air traffic control simulation to investigate the effect of task demand transitions, and the direction of those transitions, on workload and fatigue and one efficiency performance measure. Results indicate that a change in task demand appears to affect both workload and fatigue ratings, although not necessarily performance. In addition, participants' workload and fatigue ratings in equivalent task demand periods appear to change depending on the demand period preceding the time of the current ratings. Further research is needed to enhance understanding of demand transition and workload history effects on operator experience and performance, in both air traffic control and other safety-critical domains.

workload history↗
NASA NTRS2016

Task Demand Variation in Air Traffic Control: Implications for Controller Workload, Fatigue, and Performance

In air traffic control, task demand and workload have important implications for the safety and efficiency of air traffic, and remain dominant considerations. Within air traffic control, task demand is dynamic. However, research on demand transitions and subsequent controller perception and performance is limited. This research uses an air traffic control simulation to investigate the effect of task demand transitions, and the direction of those transitions, on workload and fatigue and one efficiency performance measure. Results indicate that a change in task demand appears to affect both workload and fatigue ratings, although not necessarily performance. In addition, participants workload and fatigue ratings in equivalent task demand periods appear to change depending on the demand period preceding the time of the current ratings. Further research is needed to enhance understanding of demand transition and workload history effects on operator experience and performance, in both air traffic control and other safety-critical domains.

workload history↗
NASA NTRS2021

Fast Near-Optimal Heterogeneous Task Allocation via Flow Decomposition

Multi-robot systems are uniquely well-suited to perform complex tasks such as patrolling and tracking, infor- mation gathering, and pick-up and delivery problems, offering significantly higher performance than single-robot systems. A fundamental building block in most multi-robot systems is dynamic task allocation: assigning robots to tasks (e.g., patrolling an area, or servicing a transportation request) as they appear based on the robots’ states to maximize reward. In many practical situations, the allocation must account for potentially heteroge- neous capabilities (e.g., availability of appropriate sensors or actuators) to ensure the feasibility of execution, and exploit predictive information concerning the likelihood of future tasks to promote a higher reward over a long time horizon. To this end, we present an efficient algorithm for predictive heterogeneous task- allocation achieving an approximation factor of at least 1/2 of the optimal reward. Our approach demonstrates that the problem can be decomposed into several homogeneous subproblems that can be solved efficiently using min-cost flow. Through simulation experiments, we show that our algorithm is faster by several orders of magnitude than a MILP-based approach.

Pavone, Marco↗
NASA NTRSDate not supplied

Assessing the Relationships Between Sensorimotor Biomarkers and Post-Landing Functional Task Performance

Spaceflight drives adaptive changes in healthy individuals appropriate for sensorimotor function in a microgravity environment. These changes are maladaptive for return to earth's gravity. The inter-individual variability of sensorimotor decrements is striking, although poorly understood. The goal of this study is to identify a set of behavioral, neuroimaging and genetic measures that can be used to predict early post-flight performance on a set of sensorimotor tasks. Astronauts are recruited who previously participated in sensorimotor field tests and/or posturography soon after long-duration spaceflight. Behavioral tests include assessments of sensory dependency and adaptability. Visual dependency involves treadmill walking while viewing a moving virtual visual scene. Vestibular thresholds are measured while seated during lateral translations. Proprioception dependency is measured during one-legged stance on a horizontal air-bearing surface. Ground assessment of adaptability is performed (1) during treadmill walking with a virtual linear hallway and a moving walking surface, and (2) during multiple trials of navigating an obstacle course while wearing reversing prisms. The neuroimaging tests will characterize individual differences in regional brain volumes (using Structural MRI) and white matter microstructure (using Diffusion Tensor Imaging) to serve as potential predictors of adaptive capacity. The genetic tests will utilize saliva samples to examine variations in four genes chosen because of their ability to differentiate sensorimotor adaptation ability in a normative population, including Catechol-O-methyltransferase (COMT), Dopamine Receptor D2 (DRD2), Brain-derived neurotrophic factor (BDNF) and the α2-adrenergic receptor. Twenty-one ISS crewmembers have been tested to date, including 6 from this past year after testing resumed post-COVID. This cohort includes 9 first-time fliers, 4F, and mission durations lasting 182 ± 32 days, mean ± std. We are utilizing a combination of three post-flight functional task outcomes: tandem walk, recovery from fall and dynamic posturography. There is considerable variability among the post-flight performance outcomes for the 21 participants to date. Based on a partial sample using an ordinal scale survey, 80% indicated their ability to perform functional tasks were more impacted postflight relative to inflight with 50% indicating they needed to restrict movements for a longer period postflight relative to inflight. While there is a strong association within tests obtained at different R+0 timepoints, by R+24 hr performance on one post-flight test does not necessarily correlate with performance on other post-flight tests. There are apparent relationships between individual measures and specific post-flight outcome measures; however, additional data is needed to draw conclusions. Preliminary statistical analysis indicates combining biomarkers will increase predictive power and this will be explored with future analyses. Our preliminary findings underscore the importance of a comprehensive post-flight test battery including different types of tasks with varying sensory feedback. We expect that understanding the relationships between these sensorimotor biomarkers and post-flight functional task performance will improve both our understanding of the individual variability and our strategy to optimize sensorimotor countermeasures.

S J Wood↗
NASA NTRS2024

Assessing the Relationships Between Sensorimotor Biomarkers and Post-Landing Functional Task Performance

Spaceflight drives adaptive changes in healthy individuals appropriate for sensorimotor function in a microgravity environment. These changes are maladaptive for return to earth's gravity. The inter-individual variability of sensorimotor decrements is striking, although poorly understood. The goal of this study is to identify a set of behavioral, neuroimaging and genetic measures that can be used to predict early post-flight performance on a set of sensorimotor tasks. Astronauts are recruited who previously participated in sensorimotor field tests and/or posturography soon after long-duration spaceflight. Behavioral tests include assessments of sensory dependency and adaptability. Visual dependency involves treadmill walking while viewing a moving virtual visual scene. Vestibular perceptual thresholds are measured while seated during lateral translations. Proprioception dependency is measured during one-legged stance on a horizontal air-bearing surface. Ground assessment of adaptability is performed(1) during treadmill walking with a virtual linear hallway and a moving walking surface, and (2) during multiple trials of navigating an obstacle course while wearing reversing prisms(adaptive Functional Mobility Test, aFMT). The neuroimaging tests will characterize individual differences in regional brain volumes (using Structural MRI) and white matter microstructure(using Diffusion Tensor Imaging) to serve as potential predictors of adaptive capacity. The genetic tests will utilize saliva samples to examine variations in four genes chosen because of their ability to differentiate sensorimotor adaptation ability in a normative population, including Catechol-O-methyltransferase (COMT), Dopamine Receptor D2 (DRD2), Brain-derived neurotrophic factor (BDNF) and the α2-adrenergic receptor. Twenty-seven ISS crewmembers have been tested to date, including 6 from this past year. This cohort includes 10first-time fliers, 6F, and mission durations lasting 178.6 ± 30.5 days, mean ± std. We are utilizing a combination of three post-flight functional task outcomes: tandem walk, recovery from fall and dynamic posturography. There is considerable variability among the post-flight performance outcomes for the 27participants to date. Based on a partial sample using an ordinal scale survey, 70% indicated their ability to perform functional tasks were more impacted postflight relative to inflight with 50% indicating they needed to restrict movements for a longer period postflight relative to inflight. While there is a strong association within tests obtained at different R+0 timepoints, by R+24 hr performance on one post-flight test does not necessarily correlate with performance on other post-flight tests. There are apparent relationships between individual measures and specific post-flight outcome measures, e.g., the cumulative time to complete the aFMT is significantly correlated to pre-to-post-flight changes in tandem walk (rho= 0.64, p = 0.001). Preliminary statistical analysis indicates combining biomarkers will increase predictive power and this will be explored with future analyses. Our preliminary findings underscore the importance of a comprehensive post-flight test battery including different types of tasks with varying sensory feedback. We expect that understanding the relationships between these sensorimotor biomarkers and post-flight functional task performance will improve both our understanding of the individual variability and our strategy to optimize sensorimotor countermeasures

S J Wood↗
NASA NTRSDate not supplied

Assessing the Relationships Between Sensorimotor Biomarkers and Post-Landing Functional Task Performance

Spaceflight drives adaptive changes in healthy individuals appropriate for sensorimotor function in a microgravity environment. These changes are maladaptive for return to earth's gravity. The inter-individual variability of sensorimotor decrements is striking, although poorly understood. The goal of this study was to identify a set of behavioral, neuroimaging and genetic measures that can be used to predict early post-flight performance on a set of sensorimotor tasks. Astronauts were recruited who previously participated in sensorimotor field tests and/or posturography soon after long-duration spaceflight. Behavioral tests included assessments of sensory dependency and adaptability. Visual dependency involved treadmill walking while viewing a moving virtual visual scene. Vestibular perceptual thresholds were measured while seated during lateral translations. Proprioception dependency was measured during one-legged stance on a horizontal air-bearing surface. Ground assessment of adaptability was performed (1) during treadmill walking with a virtual linear hallway and a moving walking surface, and (2) during multiple trials of navigating an obstacle course while wearing reversing prisms (adaptive Functional Mobility Test, aFMT). The neuroimaging tests characterized individual differences in regional brain volumes (using Structural MRI) and white matter microstructure (using Diffusion Tensor Imaging) to serve as potential predictors of adaptive capacity. The genetic tests utilized saliva samples to examine variations in four genes chosen because of their ability to differentiate sensorimotor adaptation ability in a normative population, including Catechol-O-methyltransferase (COMT), Dopamine Receptor D2 (DRD2), Brain-derived neurotrophic factor (BDNF) and the α2-adrenergic receptor. Thirty ISS crewmembers were recruited, including 12 first-time fliers, 6F, and mission durations lasting 185.5 ± 45.5 days, mean ± std. We utilized a combination of three post-flight functional task outcomes: tandem walk, recovery from fall and dynamic posturography. There was considerable variability among all post-flight performance outcomes. Based on an ordinal scale survey, 72% indicated their ability to perform functional tasks were more impacted postflight relative to inflight with 50% indicating they needed to restrict movements for a longer period postflight relative to inflight. While there is a strong association within tests obtained at different R+0 timepoints, by R+24 hr performance on one post-flight test does not necessarily correlate with performance on other post-flight tests. Preliminary statistical analysis indicates combining biomarkers will increase predictive power and this will be explored with future analyses. Our preliminary findings underscore the importance of a comprehensive post-flight test battery including different types of tasks with varying sensory feedback. Understanding the relationships between these sensorimotor biomarkers and post-flight functional task performance improve both our understanding of the individual variability and our strategy to optimize sensorimotor countermeasures.

biomarkers↗
DOE OSTI2015

Materials for Advanced Ultrasupercritical Steam Turbines Task 4: Cast Superalloy Development

The Steam Turbine critical stationary structural components are high integrity Large Shell and Valve Casing heavy section Castings, containing high temperature steam under high pressures. Hence to support the development of advanced materials technology for use in an AUSC steam turbine capable of operating with steam conditions of 760°C (1400°F) and 35 Mpa (5000 psia), Casting alloy selection and evaluation of mechanical, metallurgical properties and castability with robust manufacturing methods are mandated. Alloy down select from Phase 1 based on producability criteria and creep rupture properties tested by NETL-Albany and ORNL directed the consortium to investigate cast properties of Haynes 282 and Haynes 263. The goals of Task 4 in Phase 2 are to understand a broader range of mechanical properties, the impact of manufacturing variables on those properties. Scale up the size of heats to production levels to facilitate the understanding of the impact of heat and component weight, on metallurgical and mechanical behavior. GE Power & Water Materials and Processes Engineering for the Phase 2, Task 4.0 Castings work, systematically designed and executed casting material property evaluation, multiple test programs. Starting from 15 lbs. cylinder castings to world’s first 17,000 lbs. poured weight, heavy section large steam turbine partial valve Haynes 282 super alloy casting. This has demonstrated scalability of the material for steam Turbine applications. Activities under Task 4.0, Investigated and characterized various mechanical properties of Cast Haynes 282 and Cast Nimonic 263. The development stages involved were: 1) Small Cast Evaluation: 4 inch diam. Haynes 282 and Nimonic 263 Cylinders. This provided effects of liquidus super heat range and first baseline mechanical data on cast versions of conventional vacuum re-melted and forged Ni based super alloys. 2) Step block castings of 300 lbs. and 600 lbs. Haynes 282 from 2 foundry heats were evaluated which demonstrated the importance of proper heat treat cycles for Homogenization, and Solutionizing parameters selection and implementation. 3) Step blocks casting of Nimonic 263: Carried out casting solidification simulation analysis, NDT inspection methods evaluation, detailed test matrix for Chemical, Tensile, LCF, stress rupture, CVN impact, hardness and J1C Fracture toughness section sensitivity data and were reported. 4) Centrifugal Casting of Haynes 282, weighing 1400 lbs. with hybrid mold (half Graphite and half Chromite sand) mold assembly was cast using compressor casing production tooling. This test provided Mold cooling rates influence on centrifugally cast microstructure and mechanical properties. Graphite mold section out performs sand mold across all temperatures for 0.2% YS; %Elongation, %RA, UTS at 1400°F. Both Stress-LMP and conditional Fracture toughness plots data were in the scatter band of the wrought alloy. 5) Fundamental Studies on Cooling rates and SDAS test program. Evaluated the influence of 6 mold materials Silica, Chromite, Alumina, Silica with Indirect Chills, Zircon and Graphite on casting solidification cooling rates. Actual Casting cooling rates through Liquidus to Solidus phase transition were measured with 3 different locations based thermocouples placed in each mold. Compared with solidification simulation cooling rates and measurement of SDAS, microstructure features were reported. The test results provided engineered casting potential methods, applicable for heavy section Haynes 282 castings for optimal properties, with foundry process methods and tools. 6) Large casting of Haynes 282 Drawings and Engineering FEM models and supplemental requirements with applicable specifications were provided to suppliers for the steam turbine proto type feature valve casing casting. Molding, melting and casting pouring completed per approved Manufacturing Process Plan during 2014 Q4. The partial valve casing was successfully cast after casting methods were validated with solidification simulation analysis and the casting met NDT inspection and acceptance criteria. Heat treated and sectioned to extract trepan samples at different locations comparing with cast on coupons test data. Material properties requisite for design, such as tensile, creep/rupture, LCF, Fracture Toughness, Charpy V-notch chemical analysis testing were carried out. The test results will be presented in the final report. The typical Haynes 282 large size Steam Turbine production casting from Order to Delivery foundry schedule with the activity break up is shown in Figures 107 and 108. • From Purchase Order placement to Casting pouring ~ 26 weeks. 1. Sales and commercial review 3 2. Engineering Drawings/models review 4 3. Pattern and core box manufacturing 6 4. Casting process engineering review 4 5. FEM and solidification simulation analysis 4 6. Gating & Feeder Attachments, Ceramic tiling 2 7. Molding and coremaking production scheduling 6 8. Melting planning and schedule 3 9. Pouring, cooling and shake out 2 • From Pouring to casting Delivery ~ 29 weeks 10. Shot blast and riser cutting, gates removal 3 11. Homogenizing , solutionizing HT furnace prep 4 12. Grinding, Fettling 2 13. Aging HT Cycle, cooling 2 14. VT and LPT NDT inspections 2 15. Radiographic inspection 4 16. Mechanical testing, Chemical analysis test certs 4 17. Casting weld repair upgrades and Aging PWHT 4 18. NDT after weld repairs and casting upgrades 3 19. Casting Final Inspection and test certifications 3 20. Package and delivery 2 Hence the Total Lead time from P.O to Casting delivery is approximately 55 weeks. The Task 4.2 and Task 4.3 activities and reporting completed.

Thangirala, Mani↗
DOE OSTI2017

Agile Task Tracking Tool

The work was created to provide a tool for the purpose of improving the management of tasks associated with Agile projects. Agile projects are typically completed in an iterative manner with many short duration tasks being performed as part of iterations. These iterations are generally referred to as sprints. The objective of this work is to create a single tool that enables sprint teams to manage all of their tasks in multiple sprints and automatically produce all standard sprint performance charts with minimum effort. The format of the printed work is designed to mimic a standard Kanban board. The work is developed as a single Excel file with worksheets capable of managing up to five concurrent sprints and up to one hundred tasks. It also includes a summary worksheet providing performance information from all active sprints. There are many commercial project management systems typically designed with features desired by larger organizations with many resources managing multiple programs and projects. The audience for this work is the small organizations and Agile project teams desiring an inexpensive, simple, user-friendly, task management tool. This work uses standard readily available software, Excel, requiring minimum data entry and automatically creating summary charts and performance data. It is formatted to print out and resemble standard flip charts and provide the visuals associated with this type of work.

Duke, Roger T.↗
DOE OSTI2018

System, method, and computer program product for warming a cache for a task launch

A system, method, and computer program product for warming a cache for a task launch is described. The method includes the steps of receiving a task data structure that defines a processing task, extracting information stored in a cache warming field of the task data structure, and, prior to executing the processing task, generating a cache warming instruction that is configured to load one or more entries of a cache storage with data fetched from a memory.

Ricketts, Scott↗
DOE OSTI2021

Overview of the 2021 SDP 3C Citation Context Classification Shared Task

This paper provides an overview of the 2021 3C Citation Context Classification shared task. The second edition of the shared task was organised as part of the 2nd Workshop on Scholarly Document Processing (SDP 2021). The task is composed of two subtasks: classifying citations based on their (Subtask A) purpose and (Subtask B) influence. As in the previous year, both tasks were hosted on Kaggle and used a portion of the new ACT dataset. A total of 22 teams participated in Subtask A, and 19 teams competed in Subtask B. All the participated systems were ranked based on their achieved macro f-score. The highest scores of 0.26973 and 0.60025 were reported for sub-task A and B, respectively.

Kunnath, Suchetha N.↗
DOE OSTI2016

Wind power forecasting: IEA Wind Task 36 & future research issues

Here, this paper presents the new International Energy Agency Wind Task 36 on Forecasting, and invites to collaborate within the group. Wind power forecasts have been used operatively for over 20 years. Despite this fact, there are still several possibilities to improve the forecasts, both from the weather prediction side and from the usage of the forecasts. The new International Energy Agency (IEA) Task on Forecasting for Wind Energy tries to organise international collaboration, among national meteorological centres with an interest and/or large projects on wind forecast improvements (NOAA, DWD, MetOffice, met.no, DMI,...), operational forecaster and forecast users. The Task is divided in three work packages: Firstly, a collaboration on the improvement of the scientific basis for the wind predictions themselves. This includes numerical weather prediction model physics, but also widely distributed information on accessible datasets. Secondly, we will be aiming at an international pre-standard (an IEA Recommended Practice) on benchmarking and comparing wind power forecasts, including probabilistic forecasts. This WP will also organise benchmarks, in cooperation with the IEA Task WakeBench. Thirdly, we will be engaging end users aiming at dissemination of the best practice in the usage of wind power predictions. As first results, an overview of current issues for research in short-term forecasting of wind power is presented.

17 WIND ENERGY↗
DOE OSTI2022

Multi-task graph neural networks for simultaneous prediction of global and atomic properties in ferromagnetic systems *

Abstract We introduce a multi-tasking graph convolutional neural network, HydraGNN, to simultaneously predict both global and atomic physical properties and demonstrate with ferromagnetic materials. We train HydraGNN on an open-source ab initio density functional theory (DFT) dataset for iron-platinum with a fixed body centered tetragonal lattice structure and fixed volume to simultaneously predict the mixing enthalpy (a global feature of the system), the atomic charge transfer, and the atomic magnetic moment across configurations that span the entire compositional range. By taking advantage of underlying physical correlations between material properties, multi-task learning (MTL) with HydraGNN provides effective training even with modest amounts of data. Moreover, this is achieved with just one architecture instead of three, as required by single-task learning (STL). The first convolutional layers of the HydraGNN architecture are shared by all learning tasks and extract features common to all material properties. The following layers discriminate the features of the different properties, the results of which are fed to the separate heads of the final layer to produce predictions. Numerical results show that HydraGNN effectively captures the relation between the configurational entropy and the material properties over the entire compositional range. Overall, the accuracy of simultaneous MTL predictions is comparable to the accuracy of the STL predictions. In addition, the computational cost of training HydraGNN for MTL is much lower than the original DFT calculations and also lower than training separate STL models for each property.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2019

Extracting SIMD Parallelism from Recursive Task-Parallel Programs

The pursuit of computational efficiency has led to the proliferation of throughput-oriented hardware, from GPUs to increasingly wide vector units on commodity processors and accelerators. This hardware is designed to efficiently execute data-parallel computations in a vectorized manner. However, many algorithms are more naturally expressed as divide-and-conquer, recursive, task-parallel computations. In the absence of data parallelism, it seems that such algorithms are not well suited to throughput-oriented architectures. This paper presents a set of novel code transformations that expose the data parallelism latent in recursive, task- parallel programs. These transformations facilitate straightforward vectorization of task-parallel programs on commodity hardware. We also present scheduling policies that maintain high utilization of vector resources while limiting space usage. Across several task-parallel benchmarks, we demonstrate both efficient vector resource utilization and substantial speedup on chips using Intel’s SSE4.2 vector units, as well as accelerators using Intel’s AVX512 units. We then show through rigorous sampling that, in practice, our vectorization techniques are effective for a much larger class of programs.

Recursive Programs, task parallelism, vectorizatio↗
DOE OSTI2023

Multi-Task Learning of Scanning Electron Microscopy and Synthetic Thermal Tomography Images for Detection of Defects in Additively Manufactured Metals

One of the key challenges in laser powder bed fusion (LPBF) additive manufacturing of metals is the appearance of microscopic pores in 3D-printed metallic structures. Quality control in LPBF can be accomplished with non-destructive imaging of the actual 3D-printed structures. Thermal tomography (TT) is a promising non-contact, non-destructive imaging method, which allows for the visualization of subsurface defects in arbitrary-sized metallic structures. However, because imaging is based on heat diffusion, TT images suffer from blurring, which increases with depth. We have been investigating the enhancement of TT imaging capability using machine learning. In this work, we introduce a novel multi-task learning (MTL) approach, which simultaneously performs the classification of synthetic TT images, and segmentation of experimental scanning electron microscopy (SEM) images. Synthetic TT images are obtained from computer simulations of metallic structures with subsurface elliptical-shaped defects, while experimental SEM images are obtained from imaging of LPBF-printed stainless-steel coupons. MTL network is implemented as a shared U-net encoder between the classification and the segmentation tasks. Results of this study show that the MTL network performs better in both the classification of synthetic TT images and the segmentation of SEM images tasks, as compared to the conventional approach when the individual tasks are performed independently of each other.

36 MATERIALS SCIENCE↗
DOE OSTI2017

International Energy Agency Ocean Energy Systems Task 10 Wave Energy Converter Modeling Verification and Validation: Preprint

This is the first joint reference paper for the Ocean Energy Systems (OES) Task 10 Wave Energy Converter modeling verification and validation group. The group is established under the OES Energy Technology Network program under the International Energy Agency. OES was founded in 2001 and Task 10 was proposed by Bob Thresher (National Renewable Energy Laboratory) in 2015 and approved by the OES Executive Committee EXCO in 2016. The kickoff workshop took place in September 2016, wherein the initial baseline task was defined. Experience from similar offshore wind validation/verification projects (OC3-OC5 conducted within the International Energy Agency Wind Task 30) [1], [2] showed that a simple test case would help the initial cooperation to present results in a comparable way. A heaving sphere was chosen as the first test case. The team of project participants simulated different numerical experiments, such as heave decay tests and regular and irregular wave cases. The simulation results are presented and discussed in this paper.

16 TIDAL AND WAVE POWER↗
DOE OSTI2021

TBAA20: Task-Based Algorithms and Applications

The new challenges posed by Exascale system architectures have resulted in difficulty achieving a desired scalability using traditional distributed ­memory runtimes. Task­-based programming models show promise in addressing these challenges, providing application developers with a productive and performant approach to programming on next generation systems. Empirical studies show that task-based models can overcome load ­balancing issues that are inherent to traditional distributed ­memory runtimes, and that task-­based runtimes perform comparably to those systems when balanced. This panel is designed to explore the advantages of task-­based programming models on modern and future HPC systems from an industry, university, and national lab perspective. It aims at gathering application experts and proponents of these models to present concrete and practical examples of using task­-based runtimes to overcome the challenges posed by Exascale system architectures. This report describes the objectives, activities, and outcomes of the panel TBAA: Task­-Based Algorithms and Applications which was held at the International Conference for High Performance Computing, Networking, Storage, and Analysis (SC 20) on November 18, 2020.

97 MATHEMATICS AND COMPUTING↗