Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “directional data”

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 1,495 records · Page 83

Nuclear Energy Infrastructure Database Description and User’s Manual

In 2014, the Deputy Assistant Secretary for Science and Technology Innovation initiated the Nuclear Energy (NE)–Infrastructure Management Project by tasking the Nuclear Science User Facilities, formerly the Advanced Test Reactor National Scientific User Facility, to create a searchable and interactive database of all pertinent NE-supported and -related infrastructure. This database, known as the Nuclear Energy Infrastructure Database (NEID), is used for analyses to establish needs, redundancies, efficiencies, distributions, etc., to best understand the utility of NE’s infrastructure and inform the content of infrastructure calls. The Nuclear Science User Facilities developed the database by utilizing data and policy direction from a variety of reports from the U.S. Department of Energy, the National Research Council, the International Atomic Energy Agency, and various other federal and civilian resources. The NEID currently contains data on 802 research and development instruments housed in 377 facilities at 84 institutions in the United States and abroad. The effort to maintain and expand the database is ongoing. Detailed information on many facilities must be gathered from associated institutions and added to complete the database. The data must be validated and kept current to capture facility and instrumentation status as well as to cover new acquisitions and retirements. This document provides a short tutorial on the navigation of the NEID web portal at NSUF-Infrastructure.INL.gov.

42 ENGINEERING↗

The Use of the North Alabama Lightning Mapping Array (NALMA) in the Real-Time Operational Warning Environment During the March 2nd, 2012 Severe Weather Outbreak in Northern Alabama

The North Alabama Lightning Mapping Array (NALMA) is a three-dimensional very high frequency (VHF) detection network consisting of 11 sensors spread across north central Alabama and two sensors located in the Atlanta, Georgia region. The primary advantage of this network is that it detects total lightning, or the combination of both cloud-to-ground and intra-cloud lightning, instead of cloud-to-ground lightning alone. This helps to build a complete picture of storm evolution and development, and can serve as a proxy for storm updraft strength, particularly since intra-cloud lightning makes up the majority of all lightning in a typical thunderstorm. While the NALMA data do not directly indicate severe weather, they can indirectly indicate when a storm is strengthening (weakening) due to increases (decreases) in updraft strength, as the updraft is responsible for charging mechanisms within the storm. Data output are VHF radiation sources, which are produced during lightning breakdown processes. These sources are made into 2x2 km source density grids and are ported into the Advanced Weather Interactive Processing System (AWIPS) for National Weather Service (NWS) offices in Huntsville, AL, Nashville, TN, Morristown, TN, and Birmingham, AL, in near real-time. An increase in sources, or source densities, correlates to increased lightning activity and trends in updraft magnitude as long as the storm is within about 125 km of the center of the LMA network. Operationally, these data have been used at the Huntsville NWS office since early 2003 through a collaborative effort with NASA s Short-term Prediction Research and Transition (SPoRT) Center. Since then, total lightning observations have become an essential tool for forecasters during real-time warning operations. One of the operational advantages of the NALMA is the two-minute temporal resolution of the data. This provides forecasters with two to three updates during a typical volume scan of the WSR-88D radar.

White, Kristopher↗

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Deep-learning↗

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-ofthe-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Hesham ElAbd↗

Towards a study protocol: A data-driven workflow to identify error sources in direct ink write mechatronics

Abstract Using Direct Ink Write (DIW) technology in a rapid and large-scale production requires reliable quality control for printed parts. Data streams generated during printing, such as print mechatronics, are massive and diverse which impedes extracting insights. In our study protocol approach, we developed a data-driven workflow to understand the behavior of sensor-measured X- and Y- axes positional errors with process parameters, such as print velocity and velocity control. We uncovered patterns showing that instantaneous changes in the velocity, when the build platform accelerates and decelerates, largely influence the positional errors, especially in the X- axis due to the hardware architecture. Since DIW systems share similar mechatronic inputs and outputs, our study protocol approach is broadly applicable and scalable across multiple systems. Graphical abstract

36 MATERIALS SCIENCE↗

Automated and efficient local adaptive regression for principal component-based reduced-order modeling of turbulent reacting flows

Principal Component Analysis can be used to reduce the cost of Computational Fluid Dynamics simulations of turbulent reacting flows by reducing the dimensionality of the transported variables through projection of the thermochemical state onto a lower-dimensional manifold. However, because of the nonlinearity of the principal component source terms, nonlinear regression techniques must be utilized for the source terms in terms of the principal components. Unfortunately, widely available and utilized nonlinear regression techniques can have prohibitive computational requirements and/or accuracy that is highly dependent on user experience in ad hoc tuning of model architecture and hyperparameters. Here, in this work, a new nonlinear regression approach is proposed that is both computationally efficient and automated so does not require any user input. The approach is evaluated through a priori prediction of principal component source terms using data from a Direct Numerical Simulation of a turbulent nonpremixed n-heptane/air jet flame. In particular, the proposed framework consists of local regressions whose complexity is adapted according to the local nonlinearity of the data: local linear regression when accurate enough and local Artificial Neural Networks when nonlinear regression is required. The number of local clusters for local regression is determined automatically using the Davies-Bouldin index. In addition, Bayesian optimization is utilized for model training (i.e., to select the best architectures and hyperparameters of the nonlinear regressions in an unsupervised fashion), eliminating ad hoc hand-tuning and/or expensive grid searches. Overall, compared to a single, global neural network, the new local adaptive regression approach is shown to have comparable accuracy but 69% less training time due to the utilization of local linear regression and faster training of local neural networks.

42 ENGINEERING↗

Simulating the Compton-Getting effect for hydrogen flux measurements: Implications for IBEX-Hi and -Lo observations

The Interstellar Boundary EXplorer (IBEX), launched in 2008 October, has improved our understanding of the solar wind-local interstellar medium interaction through its detection of neutral atoms, particularly that of hydrogen (H). IBEX is able to create full maps of the sky in six-month intervals as the Earth orbits the Sun, detecting H with energies between ∼0.01 and 6 keV. Due to the relative motion of IBEX to the solar inertial frame, measurements made in the spacecraft frame introduce a Compton-Getting (CG) effect, complicating measurements at the lowest energies. In this paper we provide results from a numerical simulation that calculates fluxes of H atoms at 1 AU in the inertial and spacecraft frames (both ram and anti-ram), at energies relevant to IBEX-Hi and -Lo. We show theory behind the numerical simulations, applying a simple frame transformation to derived flux equations that provides a straightforward way to simulate fluxes in the spacecraft frame. We then show results of H energetic neutral atom fluxes simulated at IBEX-Hi energy passbands 2-6 in all frames, comparing with IBEX-Hi data along selected directions, and also show results simulated at energies relevant to IBEX-Lo. Although simulations at IBEX-Hi energies agree reasonably well with the CG correction method used for IBEX-Hi data, we demonstrate the importance of properly modeling low energy H fluxes due to inherent complexities involved with measurements made in moving frames, as well as dynamic radiation pressure effects close to the Sun.

79 ASTRONOMY AND ASTROPHYSICS↗

A predictive analytic model for the solar modulation of cosmic rays

An important factor limiting our ability to understand the production and propagation of cosmic rays pertains to the effects of heliospheric forces, commonly known as solar modulation. The solar wind is capable of generating time- and charge-dependent effects on the spectrum and intensity of low-energy (≲10 GeV) cosmic rays reaching Earth. Previous analytic treatments of solar modulation have utilized the force-field approximation, in which a simple potential is adopted whose amplitude is selected to best fit the cosmic-ray data taken over a given period of time. Making use of recently available cosmic-ray data from the Voyager 1 spacecraft, along with measurements of the heliospheric magnetic field and solar wind, we construct a time-, charge- and rigidity-dependent model of solar modulation that can be directly compared to data from a variety of cosmic-ray experiments. Here, we provide a simple analytic formula that can be easily utilized in a variety of applications, allowing us to better predict the effects of solar modulation and reduce the number of free parameters involved in cosmic-ray propagation models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

V-band integrated quadriphase modulator

A V-band integrated circuit quadriphase shift keyed modulator/exciter for space communications systems was developed. Intersatellite communications systems require direct modulation at 60 GHz to enhance signal processing capability. For most systems, particularly space applications, small and lightweight components are essential to alleviate severe system design constraints. Thus to achieve wideband, high data rate systems, direct modulation techniques at millimeter waves using solid state integrated circuit technology are an integral part of the overall technology developments.

Grote, A.↗

Laser Doppler velocimeter measurements and laser sheet imaging in an annular combustor model

An experimental study was conducted in annular combustor model to provide a better understanding of the flowfield. Combustor model configurations consisting of primary jets only, annular jets only, and a combination of annular and primary jets were investigated. The purpose of this research was to provide a better understanding of combustor flows and to provide a data base for comparison with computational models. The first part of this research used a laser Doppler velocimeter to measure mean velocity and statistically calculate root-mean-square velocity in two coordinate directions. From this data, one Reynolds shear stress component and a two-dimensional turbulent kinetic energy term was determined. Major features of the flowfield included recirculating flow, primary and annular jet interaction, and high turbulence. The most pronounced result from this data was the effect the primary jets had on the flowfield. The primary jets were seen to reduce flow asymmetries, create larger recirculation zones, and higher turbulence levels. The second part of this research used a technique called marker nephelometry to provide mean concentration values in the combustor. Results showed the flow to be very turbulent and unsteady. All configurations investigated were highly sensitive to alignment of the primary and annular jets in the model and inlet conditions. Any imbalance between primary jets or misalignment of the annular jets caused severe flow asymmetries.

Dwenger, Richard Dale↗

A Data-Driven Algorithm for Enabling Delay Tolerance in Resilient Microgrid Controls Using Dynamic Mode Decomposition

The increased implementation of smart grid technologies in the power distribution grid presents unique opportunities that enable resiliency, but also brings challenges motivating needs for novel solutions and mitigation techniques. The bi-directional power and data flow allow for the grid to operate with increased resiliency, which is the ability to avoid discontinuity of service to end-use loads during extreme events. However, in applications where control of the distribution grid or microgrid relies on communication networks, the degradation of communication systems in the form of loss or high latency can cause maloperation and result in loss of end-use loads. Here this paper presents a novel framework to enable delay tolerance of centralized microgrid control schemes to mitigate communication system latency impacts and guarantee successful control action. We demonstrate the delay tolerance on a control scheme that operates a battery energy storage system (BESS) to offset the sudden loss of generation and maintain system frequency. During periods of severely degraded communication system performance, the proposed delay-tolerant algorithm compensates for the latency by utilizing a data-driven model generated at the device level using dynamic mode decomposition (DMD) to determine the performance of the communications. The DMD technique predicts the system’s frequency using device-level terminal measurements and provides updated control signals. The HELICS cosimulation platform evaluates the cyber-physical interaction of the power system model in GridLAB-D, the centralized control agent in Python, and the discrete network model in NS-3. The framework is tested and validated on the IEEE-123 node system modified to represent a networked remote microgrid model, and the results show an improvement in the dynamic performance

24 POWER TRANSMISSION AND DISTRIBUTION↗

CHAOS. II. Gas-phase abundances in NGC 5194

We have observed NGC 5194 (M51a) as part of the CHemical Abundances of Spirals project. Using the Multi Object Double Spectrographs on the Large Binocular Telescope we are able to measure one or more of the temperature-sensitive auroral lines ([O iii] λ4363, [N ii] λ5755, [S iii] λ6312) and thus measure “direct” gas-phase abundances in 29 individual H ii regions. [O iii] λ4363 is only detected in two H ii regions, both of which show indications of excitation by shocks. We compare our data to previous direct abundances measured in NGC 5194 and find excellent agreement (Δ[log(O/H)]≈0.05) for all but one region. We find no evidence of trends in Ar/O, Ne/O, or S/O within NGC 5194 or compared to other galaxies. We find modest negative gradients in both O/H and N/O with very little scatter (σ ≤ 0.08 dex), most of which can be attributed to random error and not to intrinsic dispersion. The gas-phase abundance gradient is consistent with the gradients observed in other interacting galaxies, which tend to be shallower than gradients measured in isolated galaxies. The N/O ratio (〈log(N/O)〉=−0.62) suggests secondary nitrogen production is responsible for a significantly larger fraction of nitrogen (e.g., factor of 8–10), relative to primary production mechanisms than predicted by theoretical models.

79 ASTRONOMY AND ASTROPHYSICS↗

Testing of High S Matrix Glasses to Expand DFHLW Glass Compositional Ranges (Rev.1)

Gaps in glass composition-property data for direct-feed high-level waste (DFHLW) have recently been identified. One such gap is the region of high sulfur solubility since previous, pretreated, high-level wastes contained very little sulfur. Filling this data gap will significantly broaden the range of process flowsheet options including minimal washing and will allow for optimized waste loading in DFHLW glasses. This report summarizes the data collected during the characterization of the DFHLW High S Glass Matrix (HS24). A glass matrix of 50 glass compositions was developed to evenly cover the DFHLW composition region for high sulfur glass. Matrix glasses were designed to expand the composition region outside the current component concentration and property limits so as to reduce uncertainties at the limits. The 50 matrix glasses were fabricated and tested for properties important to the success of DFHLW vitrification including: compositions, canister centerline cooling (CCC) crystallinity and isothermal crystallinity, density, viscosity, electrical conductivity (EC), product consistency test (PCT) response, toxicity, and sulfate solubility. Melter materials corrosion testing is reported elsewhere. These glasses were intentionally designed to have high SO 3 solubilities (0.7 to 2.2 SO 3 wt%) in compositional regions that had not been previously explored. Forty-eight glasses showed the measured SO 3 content retained >80% of the target SO 3 and the densities of all the glasses ranged from 2.49 g·cm -3 to 2.74 g·cm -3 . While the model predicted nepheline formation in 5 glasses, one of the tested 50 CCC glasses formed nepheline, and 35 glasses formed Cr-containing phases such as spinels and eskolaite. Only five glasses were amorphous after CCC treatment where 44 glasses with detectable crystals contained =10 wt% crystals and only one glass had > 10 wt% crystals. None of the glasses exceeded the allowable T 2% for spinel crystal formation at 950 ºC (i.e., no glasses had >2 wt% spinel at 950 ºC) during isothermal crystal fraction tests where 10 glasses showed no crystalline phases at or below 950 ºC. All the glasses (except one which failed being slightly lower than the target) satisfied the SO 3 constraint while 98 glasses did not meet the viscosity constraints and 4 failed the EC constraints. Six quenched (Q) and six CCC glasses failed the Defense Waste Processing Facility (DWPF) Environmental Assessment (EA) glass PCT threshold and 3 Q and 4 CCC failed the PCT design constraint. One glass exceeded the WTP delisting limits for Cr via EPA Method 1311 (i.e., Toxicity Characteristic Leaching Procedure, TCLP). It should be emphasized that some of these glasses were specifically designed to approach or even exceed certain property constraints, as filling data gaps in these regions will provide the greatest benefit for future model development by improving accuracy and reducing uncertainties. These insights will ultimately support the development of more robust glass formulation strategies, enabling higher waste loading, reducing operational risks, and expanding the processing envelope.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Performance of a class of parallel spread-spectrum code acquisition schemes in the presence of data modulation

An acquisition procedure for data-modulated direct-sequence spread-spectrum systems is investigated. The correlation time is partitioned into subintervals, and the integration results in these subintervals are noncoherently combined for detection. The tradeoff between noncoherent combining loss and data modulation degradation guides the optimum choice of the number of subintervals. Two forms of data modulation are considered, namely, the alternate-data and random-data cases. The parallel acquisition schemes discussed allow multiple code-phase offsets to be examined at each test. The circular state diagram approach is used to analyze the performances of these schemes. The theory presented is valid for a class of such parallel schemes.

Cheng, Unjeng↗

Use of system identification techniques for improving airframe finite element models using test data

A method for using system identification techniques to improve airframe finite element models was developed and demonstrated. The method uses linear sensitivity matrices to relate changes in selected physical parameters to changes in total system matrices. The values for these physical parameters were determined using constrained optimization with singular value decomposition. The method was confirmed using both simple and complex finite element models for which pseudo-experimental data was synthesized directly from the finite element model. The method was then applied to a real airframe model which incorporated all the complexities and details of a large finite element model and for which extensive test data was available. The method was shown to work, and the differences between the identified model and the measured results were considered satisfactory.

Hanagud, Sathya V.↗

Project Description and Publications List for UAH CMMR

This research combines a state of the art X-ray Transmission Microscope, XTM, with a specially designed x-ray transparent horizontal Bridgman furnace to image (with resolutions up to 3 micrometers) the solidification of metal alloys in real-time. The objective is to obtain real-time dynamic data to provide direct measure of the solute profile in the liquid, phase coalescence and growth in the liquid, and the detailed interface morphology (e,g., dendrites and cells) during solidification. We are also enhancing the XTM data with precise solid-liquid interfacial temperature and the thermal gradient measurement techniques, and working on the application of this technology to the study of the fundamentals of solidification in microgravity. Over the last several years we have successfully imaged in real-time: interfacial-morphologies, phase growth, coalescence, incorporation of phases into the growing interface, and the solute boundary layer in the liquid at the solid-liquid interface. We have also measured true local growth rates and can evaluate segregation structures in the solid. Interfacial undercoolings are being measured either with a special Seebeck furnace or with micro-thermocouple arrays we are developing. These later techniques are presently being incorporated with the XTM furnace. This last year emphasized the investigation of the solute layer in the melt during solidification. Methods were developed to quantify the solute concentrations using x-ray absorption and to compare to predictions from simulations. In addition, work is being completed on a brass-board portable XTM that incorporates a vertical Bridgman furnace.

Kaukler, William F.↗

Emergence of Complex Modes in Lightly Damped Structural Systems

In this short presentation format, we will provide a set of useful equations, relationships, and physical interpretations direct from the damped two-mode interaction problem. We will demonstrate that the results from the two-mode interaction problem are in direct agreement with observed test data. Practical methods for estimating the parameters driving the relations direct from modal test data are provided. We finalize with two examples demonstrating weak and strong interactions.

spaceflight hardware↗

Directional acoustic measurements by laser Doppler velocimeters

Laser Doppler velocimeters (LDVs) are used as velocity microphones to measure sound pressure level in the range from 90 to 130 dB, spectral components, and two-point correlation functions for acoustic-noise source identification. Close agreement between LDV and microphone data is observed. Directional sensitivity and the ability to measure remotely make LDVs useful tools for acoustic measurement where placement of any physical probe is difficult or undesirable, as in the diagnosis of jet noise.

Mazumder, M. K.↗