SEARCH · Search NASA

Results for “Approximate 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 1,603 records · Page 89

All sources NASA NTRS DOE OSTI

NASA NTRSDate not supplied

Higher-Order Approximations for Stabilizing Zero-Energy Modes in Peridynamics Crystal Plasticity Models with Large Horizon Interactions

The non-ordinary state-based peridynamics theory combines non-local dynamic techniques with a desirable correspondence material principle, allowing for the use of continuum mechanics constitutive models. Such an approach presents a unique capability for solving problems involving discontinuities (e.g., strain localization, fracture, and fragmentation). However, the correspondence-based peridynamics models often suffer from zero-energy mode instabilities in numerical implementation, primarily due to the approximations of the non-local deformation gradient tensor. This paper focuses on a computational scheme for eliminating the zero-energy mode oscillations using a choice of influence functions that improve the truncation error in a higher-order Taylor series expansion of the deformation gradient. The novelty here is a tensor-based derivation of the linear constraint equations, which can be used to systematically identify the particle interaction weight functions for various user-specified horizon radii. In this paper, the proposed higher-order stabilization scheme is demonstrated for multi-dimensional examples involving polycrystalline and composite microstructures, along with comparisons against conventional finite element methods. The proposed stabilization scheme is shown to be highly effective in suppressing the spurious zero-energy mode oscillations in all numerical examples while enabling efficient simulations of strain localizations across material interfaces.

Non-Ordinary State-Based Peridynamics↗

NASA NTRSDate not supplied

Machine-Learned Committor Functions for Reactive Molecular Dynamics

Reactive molecular dynamics (MD) is a powerful tool for atomistic-scale modeling of a diverse range of chemical processes. However, scaling these simulations to large systems and long times scales remains a challenge because of the complexity of the potential energy function required. The authors previously developed a heuristic approach, called REACTER, that incorporates reactivity in MD simulations in a less general but much more computationally efficient manner. REACTER uses standard, fixed valence force fields as the underlying potentialenergy surface for describing all interatomic interactions but adds a procedure for enforcing user-defined reactions that occur when certain geometric constraints on relative atomic positions are satisfied. Further, these bonding changes can be accepted or rejected with a probability related tothe local thermal energy. This work seeks to generalize this approach by replacing the set of user defined geometric constraints and energetic criteria with a committor function that specifies the probability of a reaction occurring on the basis of the local atomic configuration. The committor function is a useful mathematical tool for modeling rare events but, unfortunately, is very difficult to compute for realistic systems in a general way. This work describes a method for approximating the committor function using a machine learning approach, specifically a deep neural network trained with data from reactive MD and DFT-based dynamics simulations. This network is coupled to the existing REACTER protocol, as implemented in the LAMMPS MD package, and used to make on-the-fly predictions of reaction probabilities without the more extensive user input previously required. The new method is demonstrated using the polymerization of polystyrene as a case study. Although very dependent on the quality and quantity of training data, machine-learned committor functions show promise as a method for incorporating reaction probability from higher level calculations into highly scalable MD simulations.

polymer simulations↗

NASA NTRS2022

Model Predictive Control in the Three-body Problem Using Invariant Funnels As Terminal Sets

This paper describes a method for augmenting Model Predictive Control techniques using invariant funnels computed in the Circular Restricted Three-Body Problem. We use ellipsoids that roughly approximate the boundary of the invariant funnel as convex terminal sets for a short look-ahead optimization problem at each time step. We apply this method to a hypothetical low-thrust mission to land on Jupiter’s mooon Europa and show that including the invariant funnels as terminal sets reduces the amount of control effort required by almost an order of magnitude.

Close, Sigrid↗

NASA NTRSDate not supplied

Sequential Filtering in the Presence of Uniform Measurement Errors

This paper presents a sequential filtering strategy using observations corrupted with uniform measurement noise. While the Kalman filter remains the best linear estimator of the state, other filtering techniques provide minimum variance optimal estimates, a trait only enjoyed by the Kalman filter when the underlying noises are, in fact, Gaussian. This work develops a new approximate optimal estimator for uniform measurement noises. The resulting recursion requires just slightly more computational time to complete a measurement update than the Kalman filter, which generally cannot be claimed by other optimal strategies such as the particle or Gaussian mixture filters.

James S. McCabe↗

NASA NTRSDate not supplied

Sequential Filtering in the Presence of Uniform Measurement Errors

James S McCabe↗

NASA NTRS2024

Achieving High Efficiency in Reduced Order Modeling for Large Scale Polycrystal Plasticity Simulations

Reduced order models for the nonlinear response of heterogeneous microstructures typically require a construction (or training) stage to build the reduced order basis. In this manuscript, an efficient model construction strategy for the eigenstrain homogenization method (EHM) is presented. The proposed strategy relies on a parallel, element-by-element, conjugate gradient solver. Near linear scaling has been achieved with respect to the number of degrees of freedom used to resolve the microstructure. Linear scaling with respect to the number of pre-analyses required to construct the reduced order model (ROM) follows from the EHM formulation. Furthermore, a parallel implementation for fast evaluation of the constructed ROM has been developed using shared memory parallelization. It has been shown that for large microstructures with ≈ 10,000 grains, the total computational cost of evaluating the nonlinear response of a polycrystal could be reduced by approximately an order of magnitude using 32 cores with respect to serial ROM simulation. The present methodology has been verified using an additively manufactured polycrystalline microstructure of a nickel-based superalloy, Inconel 625. The capability of the developed framework to construct a ROM for such large microstructures, as well as the ability of the ROM to predict average and local quantities of interest has been demonstrated.

microscale↗

NASA NTRS1987

Response of joint dominated space structures

An approximate method is developed for estimating the transient response of nonlinear systems in terms of linearized modes of response. Its advantages are that it is computationally more efficient than the time integration method and that it is possible to view the design problem in the more traditional physical terms of modal response. The major drawback of the approximate method is loss of accuracy. It seems that both approximate methods and time integration have their roles in design. Approximate methods provide efficient tools for performing parametric studies and they supply physical insights into how to optimize system performance that are not easily inferred from strictly numerical methods. Time integration provides a method for assessing the accuracy of the approximate solution for key simulations and for fine tuning the final design. In the procedure presented the nonlinear system is approximated by an equivalent linear system in which the system parameters are constant over the range of transient response.

Source record↗

NASA NTRS1986

Parallel structures in human and computer memory

If one thinks of our experiences as being recorded continuously on film, then human memory can be compared to a film library that is indexed by the contents of the film strips stored in it. Moreover, approximate retrieval cues suffice to retrieve information stored in this library. One recognizes a familiar person in a fuzzy photograph or a familiar tune played on a strange instrument. A computer memory that would allow a computer to recognize patterns and to recall sequences the way humans do is constructed. Such a memory is remarkably similiar in structure to a conventional computer memory and also to the neural circuits in the cortex of the cerebellum of the human brain. It is concluded that the frame problem of artificial intelligence could be solved by the use of such a memory if one were able to encode information about the world properly.

Kanerva, P.↗

NASA NTRS1992

Finite element Euler calculations of unsteady transonic cascade flows

A Galerkin finite element procedure incorporating an explicit Runge-Kutta time-stepping scheme has been developed in this work to solve unsteady transonic flow in cascades. The computational domain is discretized by a globally unstructured but locally structured blade-fitted deformable mesh. The Galerkin approximation is applied to the unsteady Euler equations based on a mixed Eulerian-Lagrangian description. The semi-discretized equations are integrated forward in time using a multistage Runge-Kutta scheme. An artificial dissipation operator of the type proposed by Jameson is adapted in the current scheme to capture shocks and suppress nonphysical oscillations. Phase-shifted boundary conditions are used to reduce the computational domain to a single reference passage. Results for both steady and unsteady transonic flows through cascades are presented and compared to existing finite volume solutions.

Hsiao, Chingteng↗

NASA NTRSDate not supplied

Development and Validation of a Slosh-Based Ullage Collapse Model

Ullage collapse modeling is an essential tool for the success of spacecraft powered by cryogenic propellants. The degree of ullage collapse can have a major impact on the design of cryogenic propellant tanks as well as the mission’s concept of operations. Additional commodities (e.g., helium gas for pressure control) and their quantities are dependent on the amount of heat transfer and phase change of propellants expected during flight. Currently, high-fidelity computational fluid dynamics (CFD) with heat transfer and phase change is the most accurate method for predicting ullage collapse. CFD simulations of this kind, however, can become very costly for problems requiring large domains, long simulation times, or significant liquid motion. Consequently, there is a need for a reduced order model to provide fast responses which can be used in engineering design decisions. The proposed numerical model is a nodal approach, augmented by outputs from CFD analyses that are generated rapidly and reliably. These outputs include transient values for liquid-vapor interface area, interface velocity, and liquid position using a volume of fluid (VoF) method. There are several benefits of combining a nodal approach with slosh predictions from CFD simulations: leveraging of the Loci/STREAM-VoF solver that has been validated in high-g and low-g environments, rapid modeling of VoF with isothermal propellant simplifications and relatively coarse meshes, and improved model predictions from the nodal tool. The nodal approach, informed by the slosh predictions from CFD, assumes uniform conditions in the liquid and gas domains. Modifications to account for temperature stratification in the liquid and gas may be implemented as a future improvement. Thermodynamic changes within the tank are solved at user-defined timesteps. First, impacts due to mass entering or leaving the control volume are captured. Second, heat transfer impacts are accounted for by tracking the volume of the gas which has been contacted by liquid in that timestep and rebalancing the internal energy. Finally, phase change is approximated using the area and velocity of the interface. The tank conditions are re-computed by conserving mass and energy. Calibration of the model to several test flights and experiments are required to baseline the predictive power and uncertainty. The final product will enable designers and engineers to create quick-turnaround predictions for ullage collapse and inform design feasibility.

Computational Fluid Dynamics↗

NASA NTRS1991

Computation of viscous blast wave flowfields

A method to determine unsteady solutions of the Navier-Stokes equations was developed and applied. The structural finite-volume, approximately factored implicit scheme uses Newton subiterations to obtain the spatially and temporally second-order accurate time history of the interaction of blast-waves with stationary targets. The inviscid flux is evaluated using MacCormack's modified Steger-Warming flux or Roe flux difference splittings with total variation diminishing limiters, while the viscous flux is computed using central differences. The use of implicit boundary conditions in conjunction with a telescoping in time and space method permitted solutions to this strongly unsteady class of problems. Comparisons of numerical, analytical, and experimental results were made in two and three dimensions. These comparisons revealed accurate wave speed resolution with nonoscillatory discontinuity capturing. The purpose of this effort was to address the three-dimensional, viscous blast-wave problem. Test cases were undertaken to reveal these methods' weaknesses in three regimes: (1) viscous-dominated flow; (2) complex unsteady flow; and (3) three-dimensional flow. Comparisons of these computations to analytic and experimental results provided initial validation of the resultant code. Addition details on the numerical method and on the validation can be found in the appendix. Presently, the code is capable of single zone computations with selection of any permutation of solid wall or flow-through boundaries.

Atwood, Christopher A.↗

NASA NTRS1988

An investigation of turbulence models

The accuracy to which a turbulent boundary layer or wake can be predicted numerically depends on the validity of the turbulence closure model used. The modeling of turbulence physics is one of the most difficult problems in computational fluid dynamics (CFD). In fact, it is one of the pacing factors in the development of CFD. In general, there are three main approaches to the description of trubulence physics. First is turbulence modeling in which the Reynolds averaged Navier-Stokes equations are used and some closure approximation is made for the Reynolds stresses. A second approach to turbulence is large eddy simulation (LES) in which the computational mesh is taken to be fine enough that the large scale structure of the turbulence can be calculated directly. An empirical assumption must be made for the small scale sub-grid turbulence. The third approach is direct simulation. In this technique the Navier-Strokes equations are solved directly on a mesh which if fine enough to resolve the smallest length scale of the turbulence. The Reynolds averaged equations are not used and no closure assumption is required. These last two approaches require extensive computer resources and as such are not engineering tools. The purpose of the work was to investigate the various engineering turbulence models for accuracy and ease of programming. This involved comparison of the models with each other and with experimental data.

Doria, Michael L.↗

NASA NTRS1991

Probabilistic structural mechanics research for parallel processing computers

Aerospace structures and spacecraft are a complex assemblage of structural components that are subjected to a variety of complex, cyclic, and transient loading conditions. Significant modeling uncertainties are present in these structures, in addition to the inherent randomness of material properties and loads. To properly account for these uncertainties in evaluating and assessing the reliability of these components and structures, probabilistic structural mechanics (PSM) procedures must be used. Much research has focused on basic theory development and the development of approximate analytic solution methods in random vibrations and structural reliability. Practical application of PSM methods was hampered by their computationally intense nature. Solution of PSM problems requires repeated analyses of structures that are often large, and exhibit nonlinear and/or dynamic response behavior. These methods are all inherently parallel and ideally suited to implementation on parallel processing computers. New hardware architectures and innovative control software and solution methodologies are needed to make solution of large scale PSM problems practical.

Sues, Robert H.↗

NASA NTRS2014

Real-Time Global Flood Estimation Using Satellite-Based Precipitation and a Coupled Land Surface and Routing Model

A widely used land surface model, the Variable Infiltration Capacity (VIC) model, is coupled with a newly developed hierarchical dominant river tracing-based runoff-routing model to form the Dominant river tracing-Routing Integrated with VIC Environment (DRIVE) model, which serves as the new core of the real-time Global Flood Monitoring System (GFMS). The GFMS uses real-time satellite-based precipitation to derive flood monitoring parameters for the latitude band 50 deg. N - 50 deg. S at relatively high spatial (approximately 12 km) and temporal (3 hourly) resolution. Examples of model results for recent flood events are computed using the real-time GFMS (http://flood.umd.edu). To evaluate the accuracy of the new GFMS, the DRIVE model is run retrospectively for 15 years using both research-quality and real-time satellite precipitation products. Evaluation results are slightly better for the research-quality input and significantly better for longer duration events (3 day events versus 1 day events). Basins with fewer dams tend to provide lower false alarm ratios. For events longer than three days in areas with few dams, the probability of detection is approximately 0.9 and the false alarm ratio is approximately 0.6. In general, these statistical results are better than those of the previous system. Streamflow was evaluated at 1121 river gauges across the quasi-global domain. Validation using real-time precipitation across the tropics (30 deg. S - 30 deg. N) gives positive daily Nash-Sutcliffe Coefficients for 107 out of 375 (28%) stations with a mean of 0.19 and 51% of the same gauges at monthly scale with a mean of 0.33. There were poorer results in higher latitudes, probably due to larger errors in the satellite precipitation input.

global flood estimation↗

NASA NTRS2004

Experimental, Theoretical, and Computational Investigation of Separated Nozzle Flows

A detailed experimental, theoretical, and computational study of separated nozzle flows has been conducted. Experimental testing was performed at the NASA Langley 16-Foot Transonic Tunnel Complex. As part of a comprehensive static performance investigation, force, moment, and pressure measurements were made and schlieren flow visualization was obtained for a sub-scale, non-axisymmetric, two-dimensional, convergent- divergent nozzle. In addition, two-dimensional numerical simulations were run using the computational fluid dynamics code PAB3D with two-equation turbulence closure and algebraic Reynolds stress modeling. For reference, experimental and computational results were compared with theoretical predictions based on one-dimensional gas dynamics and an approximate integral momentum boundary layer method. Experimental results from this study indicate that off-design overexpanded nozzle flow was dominated by shock induced boundary layer separation, which was divided into two distinct flow regimes; three- dimensional separation with partial reattachment, and fully detached two-dimensional separation. The test nozzle was observed to go through a marked transition in passing from one regime to the other. In all cases, separation provided a significant increase in static thrust efficiency compared to the ideal prediction. Results indicate that with controlled separation, the entire overexpanded range of nozzle performance would be within 10% of the peak thrust efficiency. By offering savings in weight and complexity over a conventional mechanical exhaust system, this may allow a fixed geometry nozzle to cover an entire flight envelope. The computational simulation was in excellent agreement with experimental data over most of the test range, and did a good job of modeling internal flow and thrust performance. An exception occurred at low nozzle pressure ratios, where the two-dimensional computational model was inconsistent with the three-dimensional separation observed in the experiment. In general, the computation captured the physics of the shock boundary layer interaction and shock induced boundary layer separation in the nozzle, though there were some differences in shock structure compared to experiment. Though minor, these differences could be important for studies involving flow control or thrust vectoring of separated nozzles. Combined with other observations, this indicates that more detailed, three-dimensional computational modeling needs to be conducted to more realistically simulate shock-separated nozzle flows.

Hunter, Craig A.↗

NASA NTRS1988

Euler/Navier-Stokes zonal scheme with applications to separated and isolated-vortex flows

An Euler/Navier-Stokes zonal scheme with boundary-layer compatibility conditions is developed to economically compute separated and vortex flows. The scheme is based on dividing the flow region into zones, where different levels of mathematical approximations of the governing equations are used in each zone. The scheme is applied to two specific problems: the two dimensional flow over a blunt leading-edge plate and the quasi-axisymmetric flow of an isolated vortex core. In the first problem, the computational domain is divided into inner and outer zones where the Navier-Stokes and Euler equations are used, respectively. On the downstream boundary of the computational domain, boundary-layer compatibility conditions are used. In the second problem, boundary-layer-like equations for slender, compressible, vortex flows are developed. A compatibility condition has been used to ensure consistency of the boundary and initial conditions. The outer boundary conditions of the flow are derived from Euler equations for a stream surface.

Kandil, Osama A.↗

NASA NTRS1961

A General Computer Program for Ionospheric Ray-Tracing

The study of the ionosphere using the Faraday rotation effect has been undertaken recently by means of rocket, satellite, and moon echo experiments. Different approximations have been used by different authors, resulting in methods with different degrees of complexity, and it is possible to say that the more accurate a method is, the more difficult its application becomes. However, the use of modern high-speed digital computers offers the possibility of using more complex methods in the solution of this problem, The program described in this report was written for the ILLIAC, the digital computer of the University of Illinois. Only the general features common to most digital computers will be mentioned. This program was prepared having in mind the analysis of the Faraday rotation effect, as recorded from artificial satellites. It is intended to be as general as possible in the conditions imposed on the assumed propagating medium: specifically there are no restrictions on the models of the electron density distribution and the earth's magnetic fields. long as the ray theory Ls valid. The program will trace separately the ordinary and the extraordinary mode, and it will find the virtual phase path length of a ray of each mode between the transmitter (satellite) and a receiver (station). The difference between respective phase path-lengths is related to the Faraday rotation through a constant which depends on the frequency.

Gonzales, Victor H.↗

NASA NTRS1979

Boundary conditions for exterior acoustic problems

Problems of acoustics are generally posed in unbounded regions. Numerical computations require that the infinite regions be truncated with artificial finite boundaries. Boundary conditions must be developed at these artificial boundaries which approximate the conditions for outgoing waves. These conditions should be asymptotic in the distance of the boundaries from the source terms and should give rise to well posed problems interior to fairly arbitrary regions, which can be chosen on the basis of computational efficiency rather than for mathematical (i.e., well posedness) reasons. A family of such conditions is presented and their properties are discussed. These conditions can be thought of as extensions of the Sommerfeld radiation condition for outgoing waves.

Bayliss, A.↗