SEARCH · Engineering Papers

Results for “applied computing”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29

All sources NASA NTRS DOE OSTI

DOE OSTI2023

(U) A General-Purpose Code for Correlated Sampling Using Batch Statistics with MCNP6 for Fixed-Source Problems

Correlated sampling can be used to reduce the uncertainty of a difference of tallies by taking advantage of the negative covariance term in the sandwich formula. Booth first showed how correlated sampling can be applied with batch statistics using MCNP’s tally fluctuation chart (TFC) to reduce the uncertainty of a difference of tallies in fixed-source problems. Booth presented a problem in which a 1273% uncertainty in a difference was reduced to 8% by accounting for correlations. Researchers He and Su recently studied correlated sampling using the TFC in MCNP version 5. They determined that the code did not print enough digits in the TFC tally means for accurate batch statistics in some cases. After modifying the source code, they concluded that “correlated sampling can yield a standard deviation of about one magnitude smaller than that predicted by the direct, un-correlated simulation when the changes in system response are small (say about 1%), which is equivalent to saving in CPU time by a factor of 100. Such saving [sic] becomes less significant as the change in system response becomes larger.” He and Su provided the formulas needed to apply batch statistics to compute the correlated uncertainty of a difference of tallies. In this report, we follow up on their work by providing the formulas needed to apply batch statistics to compute the correlated uncertainty of a ratio of tallies and of a difference of two tallies divided by a third tally. We extend these formulas to differences and ratios of ratios. These formulas are applied to reduce the uncertainty associated with calculating a relative sensitivity. He and Su did not investigate the accuracy of their correlated sampling uncertainty estimates. We use their test problems and evaluate the accuracy of the uncertainty estimates by comparing with results obtained from random sampling, and, in simple cases, with theoretical values of the “exact” uncertainties. We find that the uncertainties obtained from batch statistics are accurate as long as at least 100 batches are used. We present a new computer code, COSUBS (COrrelated Sampling Using Batch Statistics), that reads MCNP6 TFCs and applies correlated sampling using batch statistics for the tally combinations that the user specifies. COSUBS is a very general tool that compares all TFCs for a base case and one or two perturbed cases. It computes uncertainties for ratios if given only a base case. This report is organized as follows. The equations to apply batch statistics to the difference of random tallies are reviewed in Sec. II. Section III presents the equations for applying batch statistics to a ratio of random tallies; this is useful for computing relative sensitivities using a one-sided finite difference and the relative sensitivity using the differential operator method. Section IV presents the equations for applying batch statistics to a difference of two random tallies divided by a third; this is useful for computing a relative sensitivities using a central difference. Section V presents the equations for applying batch statistics to a difference of two ratios with four random tallies. Section VI presents the equations for applying batch statistics to a one-sided finite difference estimate of the relative sensitivity of a ratio (this uses four random tallies). Section VII presents the equations for applying batch statistics to a central difference estimate of the relative sensitivity of a ratio (this uses six random tallies). Section VIII presents the equations for applying batch statistics to a sum of random tallies. Section IX discusses how to apply batch statistics using MCNP6. Section X presents COSUBS, describing its command-line options and logic. Sections XI through XVI present numerical results for various test problems. Section XVII is a summary and conclusions. Appendix A derives the theoretical Monte Carlo tally variance given certain assumptions; these variances are used to verify the batch statistics for some of the problems. Appendix B lists the MCNP6 input for the unperturbed example problem. Appendix C presents modifications made to MCNP6.3 to support this work.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Computational study of runaway electrons in MST tokamak discharges with applied resonant magnetic perturbation

A numerical study of magnetohydrodynamics (MHD) and tracer-particle evolution investigates the effects of resonant magnetic perturbations (RMPs) on the confinement of runaway electrons (REs) in tokamak discharges conducted in the Madison Symmetric Torus. In computational results of applying RMPs having a broad toroidal spectrum but a single poloidal harmonic, m = 1 RMP does not suppress REs, whereas m = 3 RMP achieves significant deconfinement but not the complete suppression obtained in the experiment. MHD simulations with the NIMROD code produce sawtooth oscillations, and the associated magnetic reconnection can affect the trajectory of REs starting in the core region. Simulations with m = 3 RMP produce chaotic magnetic topology over the outer region, but the m = 1 RMP produces negligible changes in field topology, relative to applying no RMP. Using snapshots of the MHD simulation fields, full-orbit relativistic electron test particle computations with KORC show ≈50% loss from the m = 3 RMP compared to the 10%–15% loss from the m = 1 RMP. Here, test particle computations of the m = 3 RMP in the time-evolving MHD simulation fields show correlation between MHD activity and late-time particle losses, but total electron confinement is similar to computations using magnetic-field snapshots.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2023

Modeling, analysis, and optimization of complex nuclear processes and facilities via computational methods: The HALEU process case study

Improving and adapting industrial systems to timely meet changing programmatic and market demands is an important goal to achieve, including when operating and maintaining complex nuclear processes and facilities. However, changes to these complex systems are costly, particularly when they are already in place and bounded to stringent requirements and constraints such as when handling radioactive material and contaminated equipment. These conditions often exist when treating spent nuclear fuel remotely within shielded nuclear radiation chambers, commonly referred as hot cells, to condition nuclear material and/or fabricate products for utilization in other nuclear enterprises such as in the manufacture of advanced nuclear fuel. The illustrative case considered here is the production of high assay low enriched uranium (HALEU) products supporting the deployment of advanced nuclear reactors. For the HALEU program, resources invested were and are being systematically analyzed so that these investments are maximized in a facility that is nearly 60 years old. A methodology that has effectively enabled optimized and improvements in the Spent Fuel Treatment (SFT) program, and consequently the HALEU program, involves discrete event simulation as addressed in this article. Here, the quantification of multiple productivity metrics, including material processing rates, cycle times, bottlenecks, number of material transfers as well as equipment, workstation, and material handling utilization, has resulted in a myriad of diverse discoveries and data-informed decisions regarding process layout and constituent, labor levels and schedules, selection of new process units, storage needs, and other critical process configurations. This article describes such a computational capability being applied for decision-making, illustrates its application to an actual process and program, provides illustrative results, and argues how computational methods for the modeling, analysis, and optimization of complex processes and facilities does lead to informed decisions derived from data and not only from intuition.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

NASA NTRS1982

Applications of stereoscopic height computations from dual geosynchronous satellite data/joint NASA-Japan stereo project

Stereoscopic heights of the top of an Oklahoma thunderstorm were computed, finding that high cloud tops are not always characterized by very cold IR temperature. The identical method was also applied to the computation of stereo heights based on GOES West and GMS stereo pairs obtained under the NASA-JAPAN cooperative program. It was found that stereo techniques are extremely useful in understanding the structure of thunderstorms in the United States, as well as that of hurricanes over the South Pacific.

Fujita, T. T.↗

NASA NTRS1994

Computation of turbulent boundary layers employing the defect wall-function method

In order to decrease overall computational time requirements of spatially-marching parabolized Navier-Stokes finite-difference computer code when applied to turbulent fluid flow, a wall-function methodology, originally proposed by R. Barnwell, was implemented. This numerical effort increases computational speed and calculates reasonably accurate wall shear stress spatial distributions and boundary-layer profiles. Since the wall shear stress is analytically determined from the wall-function model, the computational grid near the wall is not required to spatially resolve the laminar-viscous sublayer. Consequently, a substantially increased computational integration step size is achieved resulting in a considerable decrease in net computational time. This wall-function technique is demonstrated for adiabatic flat plate test cases from Mach 2 to Mach 8. These test cases are analytically verified employing: (1) Eckert reference method solutions, (2) experimental turbulent boundary-layer data of Mabey, and (3) finite-difference computational code solutions with fully resolved laminar-viscous sublayers. Additionally, results have been obtained for two pressure-gradient cases: (1) an adiabatic expansion corner and (2) an adiabatic compression corner.

Brown, Douglas L.↗

NASA NTRS1992

Report on the final panel discussion on computational aeroacoustics

Some important conclusions about future prospects for aeroacoustics in general, and for computational aeroacoustics in particular, that were reached in the course of the Final Panel Discussion of the Workshop on Computational Aeroacoustics held from 6 to 9 April 1992 by ICASE and NASA Langley Research Center are summarized by the panel chairman. Aeroacoustics must now be involved in interactions with computational fluid dynamics (as applied not only to deterministic flows but also to the statistical characteristics of turbulence), while additionally incorporating rigorous comparisons with experiment. The new Computational Aeroacoustics will press forward in two parallel ways. In one of them, CFD will be used to determine aeroacoustic source strengths, the associated radiation being derived by the Acoustic Analogy approach in one of its forms. In the other, a direct Computational Aeroacoustics will apply CFD techniques over a region extending beyond the flow field so as to include at least the beginnings of the acoustic far field. There are some particularly important areas of study, including rotor noise, boundary-layer noise, and the noise of supersonic jets, where it is strongly recommended that use of both methods is continued. On the other hand, important problems of the diffraction of radiation from aeroacoustic sources around complicated aircraft shapes will require the use of comprehensively Computational Aeroacoustics, while Acoustic Analogy methods seem better suited to estimating subsonic jet noise. The study of model problems to allow comparisons with experiment will be valuable in both lines of attack.

Lighthill, James↗

DOE OSTI2020

The influence of boundary and edge-plasma modeling in computations of axisymmetric vertical displacement

A number of previously published studies apply extended-MHD computations to model macroscopic dynamics of tokamak vertical displacement events (VDEs). The characteristic plasma-wall contact and resistive-wall diffusion imply sensitivity to boundary conditions in a general sense, but which conditions have significant influence depends on how the plasma is modeled. This work examines these dependencies by comparing results of axisymmetric extended-MHD computations with different sets of boundary conditions, plasma thermal-conduction models, and wall resistivity values. The geometry and plasma parameters of the computations represent a modest-sized tokamak. The forced-displacement scenario and computational setup are similar to those used in previous 3D computations. The results show that for a given wall resistivity, the VDE time-scale is most sensitive to variations in the boundary and thermal-conduction parameters that affect electron thermal transport. The electrical conductivity depends on electron temperature, and the dependence on thermal transport stems from its influence on the electrical circuit that includes the open-field halo current. Lastly, conditions that lead to hotter, broader halo regions slow the evolution. Significant sensitivity to the boundary condition on plasma flow-velocity exists when electron thermal conduction is restricted and electron energy loss is convective, which is expected for conditions at the entrance of the magnetic presheath.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2024

Hutchinson Trace Estimation for high-dimensional and high-order Physics-Informed Neural Networks

Physics-Informed Neural Networks (PINNs) have proven effective in solving partial differential equations (PDEs), especially when some data are available by seamlessly blending data and physics. However, extending PINNs to high-dimensional and even high-order PDEs encounters significant challenges due to the computational cost associated with automatic differentiation in the residual loss function calculation. Herein, we address the limitations of PINNs in handling high-dimensional and high-order PDEs by introducing the Hutchinson Trace Estimation (HTE) method. Starting with the second-order high-dimensional PDEs, which are ubiquitous in scientific computing, HTE is applied to transform the calculation of the entire Hessian matrix into a Hessian vector product (HVP). This approach not only alleviates the computational bottleneck via Taylor-mode automatic differentiation but also significantly reduces memory consumption from the Hessian matrix to an HVP’s scalar output. We further showcase HTE’s convergence to the original PINN loss and its unbiased behavior under specific conditions. Comparisons with the Stochastic Dimension Gradient Descent (SDGD) highlight the distinct advantages of HTE, particularly in scenarios with significant variability and variance among dimensions. We further extend the application of HTE to higher-order and higher-dimensional PDEs, specifically addressing the biharmonic equation. By employing tensor-vector products (TVP), HTE efficiently computes the colossal tensor associated with the fourth-order high-dimensional biharmonic equation, saving memory and enabling rapid computation. The effectiveness of HTE is illustrated through experimental setups, demonstrating comparable convergence rates with SDGD under memory and speed constraints. Additionally, HTE proves valuable in accelerating the Gradient-Enhanced PINN (gPINN) version as well as the Biharmonic equation. Overall, HTE opens up a new capability in scientific machine learning for tackling high-order and high-dimensional PDEs.

Curse of dimensionality↗

NASA NTRS1997

Thermal Response and Ablation Programs for TPS Sizing Computation

The computer programs developed at NASA Ames Research Center for TPS sizing computation have been applied to many NASA's space missions, such as Mars Pathfinder, StarDust, Mars 2001, DS-II, and Saturn Entry Probe. These computer programs include FIAT (Fully Implicit Ablation and Thermal Response Program, MAT (Multi-component Ablation Thermochemistry Program), TPSX (Thermal Protection Systems Expert & Material Properties Database), and TPSGui (Thermal Protection Systems Graphical User Interface). For most planetary missions, the aerothermodynamics and material response are strongly coupled; thus a closed loop iteration technique between the FIAT and CFD (Computational Fluid Dynamics) codes has been developed to obtain the high fidelity bench mark TPS sizing solution. The computer codes and predictive methods are presented and discussed in detail.

Chen, Y. K.↗

NASA NTRS2008

Spiking Neurons for Analysis of Patterns

Artificial neural networks comprising spiking neurons of a novel type have been conceived as improved pattern-analysis and pattern-recognition computational systems. These neurons are represented by a mathematical model denoted the state-variable model (SVM), which among other things, exploits a computational parallelism inherent in spiking-neuron geometry. Networks of SVM neurons offer advantages of speed and computational efficiency, relative to traditional artificial neural networks. The SVM also overcomes some of the limitations of prior spiking-neuron models. There are numerous potential pattern-recognition, tracking, and data-reduction (data preprocessing) applications for these SVM neural networks on Earth and in exploration of remote planets. Spiking neurons imitate biological neurons more closely than do the neurons of traditional artificial neural networks. A spiking neuron includes a central cell body (soma) surrounded by a tree-like interconnection network (dendrites). Spiking neurons are so named because they generate trains of output pulses (spikes) in response to inputs received from sensors or from other neurons. They gain their speed advantage over traditional neural networks by using the timing of individual spikes for computation, whereas traditional artificial neurons use averages of activity levels over time. Moreover, spiking neurons use the delays inherent in dendritic processing in order to efficiently encode the information content of incoming signals. Because traditional artificial neurons fail to capture this encoding, they have less processing capability, and so it is necessary to use more gates when implementing traditional artificial neurons in electronic circuitry. Such higher-order functions as dynamic tasking are effected by use of pools (collections) of spiking neurons interconnected by spike-transmitting fibers. The SVM includes adaptive thresholds and submodels of transport of ions (in imitation of such transport in biological neurons). These features enable the neurons to adapt their responses to high-rate inputs from sensors, and to adapt their firing thresholds to mitigate noise or effects of potential sensor failure. The mathematical derivation of the SVM starts from a prior model, known in the art as the point soma model, which captures all of the salient properties of neuronal response while keeping the computational cost low. The point-soma latency time is modified to be an exponentially decaying function of the strength of the applied potential. Choosing computational efficiency over biological fidelity, the dendrites surrounding a neuron are represented by simplified compartmental submodels and there are no dendritic spines. Updates to the dendritic potential, calcium-ion concentrations and conductances, and potassium-ion conductances are done by use of equations similar to those of the point soma. Diffusion processes in dendrites are modeled by averaging among nearest-neighbor compartments. Inputs to each of the dendritic compartments come from sensors. Alternatively or in addition, when an affected neuron is part of a pool, inputs can come from other spiking neurons. At present, SVM neural networks are implemented by computational simulation, using algorithms that encode the SVM and its submodels. However, it should be possible to implement these neural networks in hardware: The differential equations for the dendritic and cellular processes in the SVM model of spiking neurons map to equivalent circuits that can be implemented directly in analog very-large-scale integrated (VLSI) circuits.

Huntsberger, Terrance↗

DOE OSTI2021

LLNL Response to the DOE ASCR RFI, "Stewardship of Software for Scientific and High-Performance Computing"

For decades, Lawrence Livermore National Laboratory (LLNL) has been engaged in significant research, development, and support for software to enable scientific computing and, particularly, the use of high performance computing (HPC) in the NNSA mission space. In particular, the move in the mid-1990’s to simulation as a leading component of stockpile stewardship through the ASCI and the successor ASC programs, as well as the need for reliable data acquisition and control software for the National Ignition Facility, have been important drivers in building expertise in production-quality software development at LLNL. LLNL has also been a leader in the DOE SciDAC FASTMath Institute and the DOE Exascale Computing Project (ECP), both of which have striven to make scientific computing software – in particular, the enabling technologies underpinning simulation capabilities – more widely adopted and sustainable. As such, we believe that our experience can inform the broader goal of software stewardship for scientific and high-performance computing. LLNL strongly supports the formation of a new DOE ASCR program element in software stewardship and sustainment. Historically, DOE ASCR has funded applied mathematics and computer science research that has led to the development of important new capabilities and algorithms that are expressed as artifacts in research software. Such frameworks, libraries, and tools have seldom been directly funded to address the important issues of code maintenance, documentation, robustness, and community building. Software engineering and support have typically been done on the side in support of the ASCR-driven research products. DOE funding priorities have been slow to recognize that good software engineering, the kind that ensures research investments have more adoption and longevity, requires significant resources. Based upon our experiences, we have prepared this response to highlight the concerns and issues we believe to be important as DOE ASCR considers its role in scientific software stewardship. We believe that role is important and will require a significant investment of new funding to legitimately support the technologies past and future DOE ASCR investments have and will produce to facilitate their uptake and adoption in the broader scientific computing community. Following a summary of our involvement in scientific software development, the remainder our response is organized around the nine topics specifically identified in the RFI.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Brochure on the 2024 ASCR Workshop on Energy-Efficient Computing for Science

Large-scale computing has enabled numerous scientific discoveries, including ground-breaking achievements facilitated by the US Department of Energy (DOE) supercomputers and advances in applied mathematics and computer science. While important advances were made in energy efficiency to enable exascale computing, continued efforts are needed to dramatically improve the energy efficiency of the next generation of high-performance computing (HPC) systems and, more broadly, AI data centers. Without substantial improvements in energy efficiency, the energy consumption associated with computing could become a limiting factor for future scientific discovery, national security, and technological advancement.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

NASA NTRS1986

Three-dimensional, conservative, Euler computations using patched grid systems and explicit methods

The method of 'zonal approach' (in which the flow field is partitioned into regions with independent grids) for computation of flow over complex geometries, such as aircraft configuration, requires application of a grid-interfacing procedure. A three-dimensional conservative, boundary scheme for patched grids, applicable in generalized coordinates for arbitrary point distributions on a planar zonal surface is presented. The computation technique is derived within the framework of the Osher (1983) upwind scheme, using Euler equations. The three-dimensional interfacing method is applied to the computation of flow about a wing-canard combination using a two-zone, patched grid.

Hessenius, K. A.↗

NASA NTRS1990

Analytical learning and term-rewriting systems

Analytical learning is a set of machine learning techniques for revising the representation of a theory based on a small set of examples of that theory. When the representation of the theory is correct and complete but perhaps inefficient, an important objective of such analysis is to improve the computational efficiency of the representation. Several algorithms with this purpose have been suggested, most of which are closely tied to a first order logical language and are variants of goal regression, such as the familiar explanation based generalization (EBG) procedure. But because predicate calculus is a poor representation for some domains, these learning algorithms are extended to apply to other computational models. It is shown that the goal regression technique applies to a large family of programming languages, all based on a kind of term rewriting system. Included in this family are three language families of importance to artificial intelligence: logic programming, such as Prolog; lambda calculus, such as LISP; and combinatorial based languages, such as FP. A new analytical learning algorithm, AL-2, is exhibited that learns from success but is otherwise quite different from EBG. These results suggest that term rewriting systems are a good framework for analytical learning research in general, and that further research should be directed toward developing new techniques.

Laird, Philip↗

DOE OSTI2021

Audacity of huge: overcoming challenges of data scarcity and data quality for machine learning in computational materials discovery

Machine learning (ML)-accelerated discovery requires large amounts of high-fidelity data to reveal predictive structure–property relationships. For many properties of interest in materials discovery, the challenging nature and high cost of data generation has resulted in a data landscape that is both scarcely populated and of dubious quality. Data-driven techniques starting to overcome these limitations include the use of consensus across functionals in density functional theory, the development of new functionals or accelerated electronic structure theories, and the detection of where computationally demanding methods are most necessary. When properties cannot be reliably simulated, large experimental data sets can be used to train ML models. In the absence of manual curation, increasingly sophisticated natural language processing and automated image analysis are making it possible to learn structure–property relationships from the literature. Finally, models trained on these data sets will improve as they incorporate community feedback.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Hedging direct simulation Monte Carlo bets via event splitting

We propose a new scheme for simulation of collisions with multiple possible outcomes in variable-weight DSMC computations. The scheme is applied to a 0-D ionization rate coefficient computation, and 1-D electrical breakdown simulation. We show that the scheme offers a significant (up to an order of magnitude) improvement in the level of stochastic noise over the usual acceptance-rejection algorithm, even when controlling for the slight additional computational costs. Furthermore, the benefits and performance of the scheme are analyzed in detail, and possible extensions are proposed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2024

Deep learning with mixup augmentation for improved pore detection during additive manufacturing

In additive manufacturing (AM), process defects such as keyhole pores are difficult to anticipate, affecting the quality and integrity of the AM-produced materials. Hence, considerable efforts have aimed to predict these process defects by training machine learning (ML) models using passive measurements such as acoustic emissions. This work considered a dataset in which keyhole pores of a laser powder bed fusion (LPBF) experiment were identified using X-ray radiography and then registered both in space and time to acoustic measurements recorded during the LPBF experiment. Due to AM’s intrinsic process controls, where a pore-forming event is relatively rare, the acoustic datasets collected during monitoring include more non-pores than pores. In other words, the dataset for ML model development is imbalanced. Moreover, this imbalanced and sparse data phenomenon remains ubiquitous across many AM monitoring schemes since training data is nontrivial to collect. Hence, we propose a machine learning approach to improve this dataset imbalance and enhance the prediction accuracy of pore-labeled data. Specifically, we investigate how data augmentation helps predict pores and non-pores better. This imbalance is improved using recent advances in data augmentation called Mixup, a weak-supervised learning method. Convolutional neural networks (CNNs) are trained on original and augmented datasets, and an appreciable increase in performance is reported when testing on five different experimental trials. When ML models are trained on original and augmented datasets, they achieve an accuracy of 95% and 99% on test datasets, respectively. We also provide information on how dataset size affects model performance. Lastly, we investigate the optimal Mixup parameters for augmentation in the context of CNN performance.

36 MATERIALS SCIENCE↗

NASA NTRS1979

Optimal nodal point distribution for improved accuracy in computational fluid dynamics

In applying finite-difference techniques to flow field problems, the accuracy attained for a fixed number of node points can be improved using unequally-spaced node points. The distribution of these node points is chosen here by minimizing a measure of local truncation error with respect to the parameters which define a transformation between the computational space of equally-spaced node points and the physical space of unequally-spaced node points. The problem then becomes a nonlinear programming problem. Numerical results are presented for two one-dimensional test problems: the Blasius boundary layer problem and the inviscid Burgers' equation.

Pierson, B. L.↗