Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “task work”

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 1,189 records · Page 66

EDRIXS: An open source toolkit for simulating spectra of resonant inelastic x-ray scattering

Resonant inelastic x-ray scattering (RIXS) has become a very powerful experimental technique to probe a broad range of intrinsic elementary excitations, for example, from low energy phonons and (bi-)magnons to high energy d-d, charge-transfer and plasmon excitations in strongly correlated electronic systems. Due to the complexity of the RIXS cross-section and strong core-hole effects, theoretical simulation of the experimental RIXS spectra is still a difficult task which hampers the understanding of RIXS spectra and the development of the RIXS technique. In this work, we present an open source toolkit (dubbed EDRIXS) to facilitate the simulations of RIXS spectra of strongly correlated materials based on exact diagonalization (ED) of certain model Hamiltonians. The model Hamiltonian can be from a single atom, small cluster or Anderson impurity model, with model parameters from density functional theory plus Wannier90 or dynamical mean-field theory calculations. The spectra of x-ray absorption spectroscopy (XAS) and RIXS are then calculated using Krylov subspace techniques. This toolkit contains highly efficient ED, XAS and RIXS solvers written in modern Fortran 90 language and a convenient Python library used to prepare inputs and set up calculations. We first give a short introduction to RIXS spectroscopy, and then we discuss the implementation details of this toolkit. Lastly, we show three examples to demonstrate its usage.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

Hejre, we introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of physics described by general nonlinear partial differential equations. In this work, we present our developments in the context of solving two main classes of problems: data-driven solution and data-driven discovery of partial differential equations. Depending on the nature and arrangement of the available data, we devise two distinct types of algorithms, namely continuous time and discrete time models. The first type of models forms a new family of data-efficient spatio-temporal function approximators, while the latter type allows the use of arbitrarily accurate implicit Runge–Kutta time stepping schemes with unlimited number of stages. Moreover, the effectiveness of the proposed framework is demonstrated through a collection of classical problems in fluids, quantum mechanics, reaction–diffusion systems, and the propagation of nonlinear shallow-water waves.

97 MATHEMATICS AND COMPUTING↗

Proactive Process-Level Live Migration and Back Migration in HPC Environments

As the number of nodes in high-performance computing environments keeps increasing, faults are becoming common place. Reactive fault tolerance (FT) often does not scale due to massive I/O requirements and relies on manual job resubmission. This work complements reactive with proactive FT at the process level. Through health monitoring, a subset of node failures can be anticipated when one's health deteriorates. A novel process-level live migration mechanism supports continued execution of applications during much of process migration. This scheme is integrated into an MPI execution environment to transparently sustain health-inflicted node failures, which eradicates the need to restart and requeue MPI jobs. Experiments indicate that 1-6.5 s of prior warning are required to successfully trigger live process migration while similar operating system virtualization mechanisms require 13-24 s. This self-healing approach complements reactive FT by nearly cutting the number of checkpoints in half when 70% of the faults are handled proactively. The work also provides a novel back migration approach to eliminate load imbalance or bottlenecks caused by migrated tasks. Experiments indicate the larger the amount of outstanding execution, the higher the benefit due to back migration.

99 GENERAL AND MISCELLANEOUS↗

Efficient bifurcation and tabulation of multi-dimensional combustion manifolds using deep mixture of experts: An a priori study

This work describes and validates an approach for autonomously bifurcating turbulent combustion man-ifolds to divide regression tasks amongst specialized artificial neural networks (ANNs). This approach relies on the mixture of experts (MoE) framework, where each neural network is trained to be specialized in a given portion of the input space. The assignment of different input regions to the experts is determined by a gating network, which is a neural network classifier. In some previous studies, it has been demonstrated that bifurcation of a complex combustion manifold and fitting different ANNs for each part leads to better fits or faster inference speeds. However, the manner of bifurcation in these studies was based on heuristic approaches or clustering techniques. In contrast, the proposed technique enables automatic bifurcation using non-linear planes in high-dimensional turbulent combustion manifolds that are often associated with complex behavior due to different dominating physics in various zones. Furthermore, the proposed concept is validated using 4-dimensional (4D) and 5D flamelet tables, showing that the errors obtained with a given network size, or conversely the network size required to achieve a given accuracy, is considerably reduced. The effect of the number of experts on inference speed is also investigated, showing that by increasing the number of experts from 1 to 8, the inference time can be approximately reduced by a factor of two. Moreover, it is shown that the MoE approach divides the input manifold in a physically intuitive manner, suggesting that the MoE framework can elucidate high-dimensional datasets in a physically meaningful way.

42 ENGINEERING↗

The agile alert system for gamma-ray transients

In recent years, a new generation of space missions has offered great opportunities for discovery in high-energy astrophysics. In this article we focus on the scientific operations of the Gamma-Ray Imaging Detector (GRID) on board the AGILE space mission. AGILE-GRID, sensitive in the energy range of 30 MeV-30 GeV, has detected many γ-ray transients of both galactic and extragalactic origin. This work presents the AGILE innovative approach to fast γ-ray transient detection, which is a challenging task and a crucial part of the AGILE scientific program. The goals are to describe (1) the AGILE Gamma-Ray Alert System, (2) a new algorithm for blind search identification of transients within a short processing time, (3) the AGILE procedure for γ-ray transient alert management, and (4) the likelihood of ratio tests that are necessary to evaluate the post-trial statistical significance of the results. Special algorithms and an optimized sequence of tasks are necessary to reach our goal. Data are automatically analyzed at every orbital downlink by an alert pipeline operating on different timescales. As proper flux thresholds are exceeded, alerts are automatically generated and sent as SMS messages to cellular telephones, via e-mail, and via push notifications from an application for smartphones and tablets. These alerts are crosschecked with the results of two pipelines, and a manual analysis is performed. Being a small scientific-class mission, AGILE is characterized by optimization of both scientific analysis and ground-segment resources. The system is capable of generating alerts within two to three hours of a data downlink, an unprecedented reaction time in γ-ray astrophysics.

79 ASTRONOMY AND ASTROPHYSICS↗

Uncertainty quantification of mass models using ensemble Bayesian model averaging

Developments in the description of the masses of atomic nuclei have led to various nuclear mass models that provide predictions for masses across the whole chart of nuclides. These mass models play an important role in understanding the synthesis of heavy elements in the rapid neutron capture ( r ) process. However, it is still a challenging task to estimate the size of uncertainty associated with the predictions of each mass model. In this work, a method called ensemble Bayesian model averaging (EBMA) is introduced to quantify the uncertainty of one-neutron separation energies (S 1 n ) which are directly relevant in the calculations of r -process observables. Here, this Bayesian method provides a natural way to perform model averaging, selection, and uncertainty quantification, by combining the mass models as a mixture of normal distributions whose parameters are optimized against the experimental data, employing the Markov chain Monte Carlo method using the no-u-turn sampler. The EBMA model optimized with all the experimental S 1 n from the AME2003 nuclides are shown to provide reliable uncertainty estimates when tested with the new data in the AME2020.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Artificial Neural Network-based State Estimation for Low Observable, Unbalanced Microgrids for Microgrid Building Blocks

The microgrid building blocks (MBB) were proposed as microgrid components with combined sub-components with power conversion, communication, and microgrid control capability, or a subset of such sub-components. This work addresses the microgrid controller, present in an MBB, which requires accurate state estimation to perform its tasks, including for monitoring, power flow (dispatch), fault detection, etc. In this paper, an artificial neural network (ANN)-based framework for state estimation is proposed for an MBB, especially for unbalanced and low observable microgrids. To overcome the challenge of low observability in unbalanced systems, a concept of extended adjacent matrix is introduced to reduce the required number of measurements for state estimation. Addressing the challenges, a feed forward neural network (FNN) is utilized to enhance estimation accuracy and reliability with the reduced number of measurements. The proposed state estimation is validated through extensive simulations on a microgrid, which was achieved from the modified IEEE 34-bus distribution test feeder with multiple distributed energy resources (DERs) and demonstrated superior performance in estimation accuracy and low observability.

Choi, Jongchan↗

Scaling Uintah on the Aurora Exascale System up to 122,880 Intel Ponte Vecchio Xe Stacks

The challenge of being able to scale application codes based on the Asynchronous Many-Task (AMT) Uintah framework on the Department of Energy (DOE) Aurora exascale system is addressed in this work by considering a challenging Reverse Monte Carlo Ray Tracing radiation benchmark calculation. This benchmark involves potentially global all-to-all communication and uses adaptive mesh refinement and ray tracing to achieve scalability. This benchmark has been used as part of previous scalability studies on a number of pre-exascale systems and on the DOE Frontier exascale system. This paper describes steps taken to enable this benchmark to run successfully on up to 10,240 nodes and 122,880 Intel® Ponte Vecchio Xe stacks on the DOE Aurora exascale system. This scalability was achieved through a limited number of experiments on Aurora, given machine loads and its uniqueness. These experiments constitute valuable lessons learned to achieve scalability at this level. The resulting scalability runs, while few in number, demonstrate relatively good strong-scaling characteristics. A detailed analysis of these results provides important indications about the path to scalability on Aurora for future work. Overall, these results continue the remarkable ability of this AMT approach to produce scalable solutions for challenging problems at extreme scale on heterogeneous architectures.

Garcia, Marta [Argonne National Laboratory (ANL)]↗

Sci-Vis Framework

SVF is a full featured OpenGL 3d framework that allows for rapid creation of complex visualizations. The SVF framework handles much of the lifecycle and complex tasks required for a 3d visualization. Unlike a game framework SVF was designed to use fewer resources, work well in a windowed environment, and only render when necessary. The scene also takes advantage of multiple threads to free up the UI thread as much as possible. Shapes (actors) in the scene are created by adding or removing functionality (through support objects) during runtime. This allows a highly flexible and dynamic means of creating highly complex actors without the code complexity (it also helps overcome the lack of multiple inheritance in Java.) All classes are highly customizable and there are abstract classes which are intended to be subclassed to allow a developer to create more complex and highly performant actors. There are multiple demos included in the framework to help the developer get started and shows off nearly all of the functionality. Some simple shapes (actors) are already created for you such as text, bordered text, radial text, text area, complex paths, NURBS paths, cube, disk, grid, plane, geometric shapes, and volumetric area. It also comes with various camera types for viewing that can be dragged, zoomed, and rotated. Picking or selecting items in the scene can be accomplished in various ways depending on your needs (raycasting or color picking.) The framework currently has functionality for tooltips, animation, actor pools, color gradients, 2d physics, text, 1d/2d/3d textures, children, blending, clipping planes, view frustum culling, custom shaders, and custom actor states

Arthur Bleeker, PNNL↗

SUMMARY OF 2010 DOE EM INTERNATIONAL PROGRAM STUDIES OF WASTE GLASS STRUCTURE AND PROPERTIES

Collaborative work between the Savannah River National Laboratory (SRNL) and SIA Radon in Russia was divided among three tasks for calendar year 2010. The first task focused on the study of simplified high level waste glass compositions with the objective of identifying the compositional drivers that lead to crystallization and poor chemical durability. The second task focused on detailed characterization of more complex waste glass compositions with unexpectedly poor chemical durabilities. The third task focused on determining the structure of select high level waste glasses made with varying frit compositions in order to improve models under development for predicting the melt rate of the Defense Waste Processing Facility (DWPF) glasses. The majority of these tasks were carried out at SIA Radon. Selection and fabrication of the glass compositions, along with chemical composition measurements and evaluations of durability were carried out at SRNL and are described in this report. SIA Radon provided three summary reports based on the outcome of the three tasks. These reports are included as appendices to this document. Briefly, the result of characterization of the Task 1 glasses may indicate that glass compositions where iron is predominantly tetrahedrally coordinated have more of a tendency to crystallize nepheline or nepheline-like phases. For the Task 2 glasses, the results suggested that the relatively low fraction of tetrahedrally coordinated boron and the relatively low concentrations of Al{sub 2}O{sub 3} available to form [BO{sub 4/2}]{sup -}Me{sup +} and [AlO{sub 4/2}]{sup -}Me{sup +} tetrahedral units are not sufficient to consume all of the alkali ions, and thus these alkali ions are easily leached from the glasses. All of the twelve Task 3 glass compositions were determined to be mainly amorphous, with some minor spinel phases. Several key structural units such as metasilicate chains and rings were identified, which confirms the current modeling approach for the silicate phase. The coordination of aluminum and iron was found to be mainly tetrahedral, with some octahedral iron ions. In all samples, trigonally-coordinated boron was determined to dominate over tetrahedrally-coordinated boron. The results further suggested that BO{sub 4} tetrahedra and BO{sub 3} triangles form complex borate units and may be present as separate constituents. However, no quantification of tetrahedral-to-trigonal boron ratio was made.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Analytical Plans Supporting The Sludge Batch 8 Glass Variability Study Being Conducted By Energy solutions And Cua's Vitreous State Laboratory

Energy Solutions (ES) and its partner, the Vitreous State Laboratory (VSL) of The Catholic University of America (CUA), are to provide engineering and technical services support to Savannah River Remediation, LLC (SRR) for ongoing operation of the Defense Waste Processing Facility (DWPF) flowsheet as well as for modifications to improve overall plant performance. SRR has requested via a statement of work that ES/VSL conduct a glass variability study (VS) for Sludge Batch 8. SRR issued a technical task request (TTR) asking that the Savannah River National Laboratory (SRNL) provide planning and data reduction support for the ES/VSL effort. This document provides two analytical plans for use by ES/VSL: one plan is to guide the measurement of the chemical composition of the study glasses while the second is to guide the measurement of the durability of the study glasses. The measurements generated by ES/VSL are to be provided to SRNL for data reduction and evaluation. SRNL is to review the results of its evaluation with ES/VSL and SRR. The results will subsequently be incorporated into a joint report with ES/VSL as a deliverable to SRR to support the processing of SB8 at DWPF.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Wind Forecasting Improvement Project In Complex Terrain Near the Columbia River Gorge

In 2014 a multi-institution team led by Vaisala, Inc. was selected by the Department of Energy (DOE) to partner with multiple DOE and National Oceanic and Atmospheric Administration (NOAA) laboratories on a project designed to improve the quality of wind power forecasts in areas of complex terrain. This was the second Wind Forecast Improvement Project (WFIP2) funded by DOE and it extended from late 2014 through the middle of 2018. It encompassed an 18-month observational field campaign, numerical weather prediction (NWP) model development, extensive analysis of data and NWP output, and the creation of decision support tool algorithms to convey forecast information to end users in the wind industry. WFIP2 focused on improvements to the representation of near-surface and boundary-layer physics in NOAA’s High-Resolution Rapid Refresh (HRRR) model. Improvements to HRRR, which is run operationally over the continental United States, benefit the wind industry in multiple ways. Forecasts from the operational HRRR are used directly by wind power forecast vendors and the operators of wind plants. In addition, because HRRR is built using the widely-used Weather Research and Forecasting (WRF) model, improvements to its parameterizations become available to commercial and research institutions using WRF for a myriad of purposes. The geographic area studied by WFIP2 was a region of the Columbia River basin located to the east of the Cascade Mountains between Oregon and Washington. Home to over 6 GW of installed capacity for wind energy production, this area also hosts a variety of atmospheric phenomena either unique to or augmented by complex topography. This makes it an attractive test-bed for the analysis of wind forecast in complex terrain, though results found are should be applicable in any area of topographic complexity. WFIP2 succeeded as a collaborative effort, and while this report focuses on the activities of the team led by Vaisala, the work described here is part of a larger whole. The Vaisala team accomplished a number of specific tasks as described in this report, while also contributing to this larger effort.

17 WIND ENERGY↗

Technology Development and Field Monitoring in nZEB - US Country report IEA HPT Annex 49 Task 3

The International Energy Agency (IEA) Heat Pumping Technologies (HPT) Annex 49, “Design and Integration of Heat Pumps for Nearly Zero Energy Buildings,” deals with the application of heat pumps (HPs) as a core component of the HVAC system for nearly or net-zero energy buildings. This report covers the Task 3 activities of the US team. Three institutions are involved on the US team and have worked on the following projects. 1) Oak Ridge National Laboratory (ORNL) summarized development activities since the conclusion of IEA HPT Annex 40 for several integrated HP (IHP) systems - electric ground-source IHP and air-source IHP versions and engine-driven AS-IHP version. 2) The University of Maryland partnered with ORNL and Blue Bear Management to develop a personal cooling device called RoCo that can provide personalized conditioned air to occupants in inadequately or unconditioned environments. With RoCo, building facility management can elevate the HVAC thermostat settings without compromising occupants’ thermal comfort. Researchers have found that a 4°F increase in thermostat settings can save 12–30% energy savings. Therefore, RoCo is a promising technology that can help reduce building energy consumption and facilitate achievement of net-zero energy building performance. 3) The National Institute of Standards and Technology (NIST) is working on a field study effort on the NIST Net-Zero Energy Residential Test Facility. Two air-source split-system HPs were installed in a residential, net-zero energy home that was constructed as a laboratory on the NIST campus in Gaithersburg, Maryland. The first HP was a two-stage, 7 kW (2 ton), 15.8 seasonal energy efficiency ratio (SEER), 9.05 heating seasonal performance factor (HSPF) conventionally ducted system; the second HP was a variable-speed, 10.6 kW (3 ton), 14 SEER, 8.35 HSPF, high-velocity ducted system. These two systems operated side by side, using separate supply ducts and a common return duct, on a weekly alternating schedule to condition the home that was operated with very consistent simulated thermal loads. The team wanted to determine whether the high-velocity system could provide comparable energy-use efficiency to the conventional system. The results of this study showed that it did meet the required loads and had slightly greater efficiency; the average cooling coefficient of performance (COP) was (0.40 ± 0.11) higher, and the average heating COP was statistically equal. A new firmware was provided at the end of the heating season that greatly improved the performance of the high-velocity system; its average heating COP went from (1.8 ± 0.9) to (2.5 ± 1.1) at a 95% confidence level. The new firmware heating COP averaged (1.05 ± 0.23) higher than the old firmware over the same outdoor temperatures. The defrost performance is very different for these two systems, yet they consumed equivalent energy per HDD. The conventional system uses a timed-initiate, temperature-terminate algorithm with auxiliary electric resistive heating, whereas the high-velocity system uses calculated evaporator parameters with a hot-gas bypass before a full reverse-cycle defrost with no supplementary resistive heat.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Cognition at the Point of Sensing

Over the last 15 years, compressive sensing techniques have been developed which have the potential to greatly reduce the amount of data collected by systems while preserving the amount of information obtained. A cost of this efficiency is that a computationally-intensive optimization routine must be used to put the sensed data into a form that a person can interpret. At the same time, machine learning techniques have experienced tremendous growth as well. Machines have demonstrated the ability learn how to effectively perform tasks such as detection and classification at speeds much faster than humanly possible. Our goal in this project was to study the feasibility of using compressive sensing systems "at the edge." That is, how can compressive sensing sensors be deployed such that information is created at the remote sensor rather than sending raw data to a central processing location? Studies were performed to analyze whether machine learning could be done on the compressively sensed data in its raw form. If a machine is performing the task, is it possible to do so without putting the data into a human interpretable form? We show that this is possible for some systems, in particular a compressive sensing snapshot imaging spectrometer. Machine learning tasks were demonstrated to be more effective and more robust to noise when the machine learning algorithm worked on data in its raw form. This system is shown to outperform a traditional spectrometer. Techniques for reducing the complexity of the reconstruction routine were also analyzed. Techniques for such as data regularization, deep neural networks, and matrix completion were studied and shown to have benefits over traditional reconstruction techniques. In this project we showed that compressive sensing sensors are indeed feasible at the edge. As always, sensors and algorithms must be carefully tuned to work in the constrained environment. In this project we developed tools and techniques to enable those analyses.

47 OTHER INSTRUMENTATION↗

Internship Report - Andrew Beathard

This summer, I had the opportunity to work as an intern under Ryan Goldhahn at Lawrence Livermore National Laboratory. Alongside another intern, I was tasked with implementing perception and autonomy algorithms on unmanned aerial systems (UAS) for multiagent cooperative missions and testing their effectiveness at the OS-150 UAS test facility at LLNL. We aimed to engineer a collaborative reaction to a single agent’s identification of an object of interest and test various multiagent algorithms.

42 ENGINEERING↗

FY 22 ORNL Integral Experiment work [Slides]

This presentation covers Oak Ridge National Labatory (ORNL) integral experiment work for fiscal year 2022. The talk highlights the three main Nuclear Criticality Safety Program (NCSP) funded tasks for 2022.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Research to Address Technical Barriers to Expanded Markets for Biodiesel and Biodiesel Blends (CRADA Final Report)

NREL and the Clean Fuels Alliance America (CFAA) will work cooperatively to assess the effects of biodiesel blends on the performance of modern diesel engines. This work will include research to understand the impact of biodiesel blends on the operation and durability of particle filters and NO x control sorbents/catalysts, to quantify the effect on emission control systems performance, and to understand effects on engine component durability. This research was performed at NREL Renewable Fuels and Lubricants (REFUEL) laboratory (an engine testing laboratory) as well as at third party labs paid directly by CFAA with NREL as part of the project management team. Also, research to develop appropriate ASTM standards for biodiesel quality and stability was conducted in NREL’s bench scale fuel chemistry laboratory. The cooperative project involved laboratory testing and research at NREL using biodiesel from a variety of sources and in collaboration with a broad range of other stakeholders. In addition, NREL will work with NBB to set up an Industrial Steering Committee to design the scope for the various tasks and to provide technical oversight to these projects. NREL and NBB will cooperatively communicate the study results to as broad an audience as possible. This research benefits the public by expanding markets for a domestically produced low-carbon intensity fuel for use in diesel engines.

09 BIOMASS FUELS↗

A Prospective Design Method for Nuclear Power: The Evaluation, Requirements, and Goals Outline for Nuclear (ERGON) Method

Human factors researchers at Idaho National Laboratory (INL) have worked on projects spanning control room modernization, operator support systems, visualization design, and novel system creation. These projects demonstrated the need for an explicit design method for nuclear power. Human factors teams found a high standard in the Human Factors Engineering Program Review Model (NUREG-0711) and needed a design methodology which could be successful in gaining approval. Previous work has been synthesized as the Evaluation, Requirements, and Goals Outline for Nuclear (ERGON) method here. Design tasks are broken into four phases: Context and Orientation, Human Factors Review, Prototyping and Evaluation, Iteration and Improvement. ERGON is intended as a flexible and direct design method for many applications in nuclear power. ERGON has been vetted through collaborative research and development with nuclear utilities and as such, the ERGON method can assist utilities to achieve approval from a NUREG-0711 summative evaluation for HSI implementations.

42 ENGINEERING↗