Search NASA⌕ Search

SEARCH · Search NASA

Results for “functional convergence”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRS1989

An iterative method for systems of nonlinear hyperbolic equations

An iterative algorithm for the efficient solution of systems of nonlinear hyperbolic equations is presented. Parallelism is evident at several levels. In the formation of the iteration, the equations are decoupled, thereby providing large grain parallelism. Parallelism may also be exploited within the solves for each equation. Convergence of the interation is established via a bounding function argument. Experimental results in two-dimensions are presented.

Scroggs, Jeffrey S.↗
NASA NTRS1995

A 3DGRAPE/AL: The Ames/Langley technology upgrade

This paper describes a new three-dimensional structured multiple-block volume grid generator called 3DGRAPE/AL. It is a significantly improved version of the previously-released and widely-distributed program 3DGRAPE, with many of the improvements taken from the grid-generator program 3DMAGGS. It generates volume grids by iteratively solving the Poisson Equations in three-dimensions. The right-hand-side terms are designed so that user-specified grid cell heights and user-specified grid cell skewness near boundary surfaces result automatically, with little user intervention. Versatility was a high priority in this code's development, and as a result it can generate grids in almost any three-dimensional physical domain. Improvements include added kinds of forcing functions, improved control of cell skewness, improved initial conditions, convergence acceleration, the ability to take as input the output from GRIDGEN, and a simple but powerful graphical user interface (GUI).

Sorenson, Reese L.↗
NASA NTRS2005

CFD Code Validation of Wall Heat Fluxes for a G02/GH2 Single Element Combustor

This paper puts forth the case for the need for improved injector design tools to meet NASA s Vision for Space Exploration goals. Requirements for this improved tool are outlined and discussed. The potential for Computational Fluid Dynamics (CFD) to meet these requirements is noted along with its current shortcomings, especially relative to demonstrated solution accuracy. The concept of verification and validation is introduced as the primary process for building and quantifying the confidence necessary for CFD to be useful as an injector design tool. The verification and validation process is considered in the context of the Marshall Space Flight Center (MSFC) Combustion Devices CFD Simulation Capability Roadmap via the Simulation Readiness Level (SRL) concept. The portion of the validation process which demonstrates the ability of a CFD code to simulate heat fluxes to a rocket engine combustor wall is the focus of the current effort. The FDNS and Loci-CHEM codes are used to simulate a shear coaxial single element G02/GH2 injector experiment. The experiment was conducted a t a chamber pressure of 750 psia using hot propellants from preburners. A measured wall temperature profile is used as a boundary condition to facilitate the calculations. Converged solutions, obtained from both codes by using wall functions with the K-E turbulence model and integrating to the wall using Mentor s baseline turbulence model, are compared to the experimental data. The initial solutions from both codes revealed significant issues with the wall function implementation associated with the recirculation zone between the shear coaxial jet and the chamber wall. The FDNS solution with a corrected implementation shows marked improvement in overall character and level of comparison to the data. With the FDNS code, integrating to the wall with Mentor s baseline turbulence model actually produce a degraded solution when compared to the wall function solution with the K--E model. The Loci-CHEM solution, produced by integrating to the wall with Mentor s baseline turbulence model, matches both the heat flux rise rate in the near injector region and the peak heat flux level very well. However, it moderately over predicts the heat fluxes downstream of the reattachment point. The Loci-CHEM solution achieved by integrating to the wall with Mentor s baseline turbulence model was clearly superior to the other solutions produced in this effort.

Lin, Jeff↗
NASA NTRS2011

Model Checking a Self-Stabilizing Distributed Clock Synchronization Protocol for Arbitrary Digraphs

This report presents the mechanical verification of a self-stabilizing distributed clock synchronization protocol for arbitrary digraphs in the absence of faults. This protocol does not rely on assumptions about the initial state of the system, other than the presence of at least one node, and no central clock or a centrally generated signal, pulse, or message is used. The system under study is an arbitrary, non-partitioned digraph ranging from fully connected to 1-connected networks of nodes while allowing for differences in the network elements. Nodes are anonymous, i.e., they do not have unique identities. There is no theoretical limit on the maximum number of participating nodes. The only constraint on the behavior of the node is that the interactions with other nodes are restricted to defined links and interfaces. This protocol deterministically converges within a time bound that is a linear function of the self-stabilization period.

Malekpour, Mahyar R.↗
NASA NTRS1984

Legendre-Tau approximation for functional differential equations. Part 3: Eigenvalue approximations and uniform stability

The stability and convergence properties of the Legendre-tau approximation for hereditary differential systems are analyzed. A charactristic equation is derived for the eigenvalues of the resulting approximate system. As a result of this derivation the uniform exponential stability of the solution semigroup is preserved under approximation. It is the key to obtaining the convergence of approximate solutions of the algebraic Riccati equation in trace norm.

Ito, K.↗
NASA NTRS1975

A variable-metric algorithm employing linear and quadratic penalties

A variable-metric algorithm is described that uses both linear and quadratic penalty terms for handling nonlinear constraints. Quadratic penalty coefficients are adjusted in a process which maintains a positive-definite matrix of second partial derivatives of the function without generating the large positive eigenvalues which cause zigzagging and slow convergence. The schemes suggested use inferred second-order properties not only in terms of the variable metric of the Davidson-Fletcher-Powell algorithm (or its relatives) but by estimating of second directional derivatives by fitting cubics to various functions along search directions.

Kelley, H. J.↗
NASA NTRS1992

A new algorithm for L2 optimal model reduction

In this paper the quadratically optimal model reduction problem for single-input, single-output systems is considered. The reduced order model is determined by minimizing the integral of the magnitude-squared of the transfer function error. It is shown that the numerator coefficients of the optimal approximant satisfy a weighted least squares problem and, on this basis, a two-step iterative algorithm is developed combining a least squares solver with a gradient minimizer. Convergence of the proposed algorithm to stationary values of the quadratic cost function is proved. The formulation is extended to handle the frequency-weighted optimal model reduction problem. Three examples demonstrate the optimization algorithm.

Spanos, J. T.↗
NASA NTRS1979

Experiments of shock associated noise of supersonic jets

The paper examines the aeroacoustics associated with model nozzles operating supersonically. In particular, the shock structure and radiated shock noise of Mach 1.5 and 2.0 nozzles are compared with those of a convergent nozzle over a wide nozzle pressure ratio range corresponding to a fully expanded Mach number between unity and 2.37. The nozzles were operated unheated both with and without a tab for screech tone suppression. The measurements show differences between the shock cell spacing of convergent and convergent-divergent nozzles, and the scaling relation appears to be a function of the exit-to-throat velocity ratio of each nozzle type. The acoustic measurements indicate the extent of the pressure ratio range where a C-D nozzle achieves a noise reduction benefit. At the design point of the Mach 1.5 nozzle, the total integrated sound power from this nozzle is 6 dB less than a convergent nozzle operating at the same pressure ratio and thrust

Seiner, J. M.↗
NASA NTRS1986

Larger Convergence Zones for Newton's Method

Iterative technique applies over wider range of initial guesses. New theorem describes convergence zone of Newton's iterative method for finding zeros of real function. Involves two points, Xp and Xp*, called primary conjugate points. If exact solution lies between these points (Xp is less than Xz is less than Xp*) and no other conjugate points in interval, then according to theorem, subsequent iterations will converge upon exact solution if initial guess lies in interval.

Campbell, C. W.↗
NASA NTRS1995

A General Method for Solving Systems of Non-Linear Equations

The method of steepest descent is modified so that accelerated convergence is achieved near a root. It is assumed that the function of interest can be approximated near a root by a quadratic form. An eigenvector of the quadratic form is found by evaluating the function and its gradient at an arbitrary point and another suitably selected point. The terminal point of the eigenvector is chosen to lie on the line segment joining the two points. The terminal point found lies on an axis of the quadratic form. The selection of a suitable step size at this point leads directly to the root in the direction of steepest descent in a single step. Newton's root finding method not infrequently diverges if the starting point is far from the root. However, the current method in these regions merely reverts to the method of steepest descent with an adaptive step size. The current method's performance should match that of the Levenberg-Marquardt root finding method since they both share the ability to converge from a starting point far from the root and both exhibit quadratic convergence near a root. The Levenberg-Marquardt method requires storage for coefficients of linear equations. The current method which does not require the solution of linear equations requires more time for additional function and gradient evaluations. The classic trade off of time for space separates the two methods.

Nachtsheim, Philip R.↗
NASA NTRS2002

Evolvable, reconfigurable hardware for future space systems

This paper overviews Evolvable Hardware (EHW) technology, examining its potential for enhancing survivability and flexibility of future space systems. EHW refers to selfconfiguration of electronic hardware by evolutionary/genetic search mechanisms. Evolvable Hardware can maintain existing functionality in the presence of faults and degradations due to aging, temperature and radiation. It can also configure itself for new functionality when required for mission changes or encountered opportunities. The paper illustrates hardware evolution in silicon using a JPL-designed programmable device reconfigurable at transistor level as the platform and a genetic algorithm running on a DSP as the reconfiguration mechanism. Rapid reconfiguration allows convergence to circuit solutions in the order of seconds. The experiments demonstrate functional recovery from faults as well as from degradation at extreme temperatures indicating the possibility of expanding the operational range of extreme electronics through evolved circuit solutions.

Evolvable Hardware EHW↗
NASA NTRS1978

A study of parameter identification

A set of definitions for deterministic parameter identification ability were proposed. Deterministic parameter identificability properties are presented based on four system characteristics: direct parameter recoverability, properties of the system transfer function, properties of output distinguishability, and uniqueness properties of a quadratic cost functional. Stochastic parameter identifiability was defined in terms of the existence of an estimation sequence for the unknown parameters which is consistent in probability. Stochastic parameter identifiability properties are presented based on the following characteristics: convergence properties of the maximum likelihood estimate, properties of the joint probability density functions of the observations, and properties of the information matrix.

Herget, C. J.↗
NASA NTRS1988

Quasi-Newton methods for parameter estimation in functional differential equations

A state-space approach to parameter estimation in linear functional differential equations is developed using the theory of linear evolution equations. A locally convergent quasi-Newton type algorithm is applied to distributed systems with particular emphasis on parameters that induce unbounded perturbations of the state. The algorithm is computationally implemented on several functional differential equations, including coefficient and delay estimation in linear delay-differential equations.

Brewer, Dennis W.↗
DOE OSTI2025

Application-specific machine-learned interatomic potentials: exploring the trade-off between DFT convergence, MLIP expressivity, and computational cost

Machine-learned interatomic potentials (MLIPs) are revolutionizing computational materials science and chemistry by offering an efficient alternative to ab initio molecular dynamics (MD) simulations. However, fitting high-quality MLIPs remains a challenging, time-consuming, and computationally intensive task where numerous trade-offs have to be considered, e.g., How much and what kind of atomic configurations should be included in the training set? Which level of ab initio convergence should be used to generate the training set? Which loss function should be used for fitting the MLIP? Which machine learning architecture should be used to train the MLIP? The answers to these questions significantly impact both the computational cost of MLIP training and the accuracy and computational cost of subsequent MLIP MD simulations. In this study, we use a configurationally diverse beryllium dataset and quadratic spectral neighbor analysis potential. We demonstrate that joint optimization of energy versus force weights, training set selection strategies, and convergence settings of the ab initio reference simulations, as well as model complexity can lead to a significant reduction in the overall computational cost associated with training and evaluating MLIPs. This opens the door to computationally efficient generation of high-quality MLIPs for a range of applications which demand different accuracy versus training and evaluation cost trade-offs.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Factorization Machine‐Based Active Learning for Functional Materials Design with Optimal Initial Data

The optimization of functional materials is important to enhance their properties, but their complex geometries pose great challenges to optimization. Data-driven algorithms efficiently navigate such complex design spaces by learning relationships between material structures and performance metrics to discover high-performance functional materials. Surrogate-based active learning, continually improving its surrogate model by iteratively including high-quality data points, has emerged as a cost-effective data-driven approach. Furthermore, it can be coupled with quantum computing to enhance optimization processes, especially when paired with a special form of surrogate model (i.e., quadratic unconstrained binary optimization), formulated by factorization machine (FM). However, current practices often overlook the variability in design space sizes when determining the initial data size for optimization. In this work, we investigate the optimal initial data sizes required for efficient convergence across various design space sizes. By employing averaged piecewise linear regression, we identify initiation points where convergence begins, highlighting the crucial role of employing adequate initial data in achieving efficient optimization. These results contribute to the efficient optimization of functional materials by ensuring faster convergence and reducing computational costs in FM-based active learning.

active learning↗
NASA NTRS2006

Accuracy Quantification of the Loci-CHEM Code for Chamber Wall Heat Transfer in a GO2/GH2 Single Element Injector Model Problem

A robust rocket engine combustor design and development process must include tools which can accurately predict the multi-dimensional thermal environments imposed on solid surfaces by the hot combustion products. Currently, empirical methods used in the design process are typically one dimensional and do not adequately account for the heat flux rise rate in the near-injector region of the chamber. Computational Fluid Dynamics holds promise to meet the design tool requirement, but requires accuracy quantification, or validation, before it can be confidently applied in the design process. This effort presents the beginning of such a validation process for the Loci-CHEM CFD code. The model problem examined here is a gaseous oxygen (GO2)/gaseous hydrogen (GH2) shear coaxial single element injector operating at a chamber pressure of 5.42 MPa. The GO2/GH2 propellant combination in this geometry represents one the simplest rocket model problems and is thus foundational to subsequent validation efforts for more complex injectors. Multiple steady state solutions have been produced with Loci-CHEM employing different hybrid grids and two-equation turbulence models. Iterative convergence for each solution is demonstrated via mass conservation, flow variable monitoring at discrete flow field locations as a function of solution iteration and overall residual performance. A baseline hybrid was used and then locally refined to demonstrate grid convergence. Solutions were obtained with three variations of the k-omega turbulence model.

West, Jeff↗
NASA NTRS2004

On Study of Air/Space-borne Dual-Wavelength Radar for Estimates of Rain Profiles

In this study, a framework is discussed to apply air/space-borne dual-wavelength radar for the estimation of characteristic parameters of hydrometeors. The focus of our study is on the Global Precipitation Measurements (GPM) precipitation radar, a dual-wavelength radar that operates at Ku (13.8 GHz) and Ka (35 GHz) bands. As the droplet size distributions (DSD) of rain are expressed as the Gamma function, a procedure is described to derive the median volume diameter (D(sub 0)) and particle number concentration (N(sub T)) of rain. The correspondences of an important quantity of dual-wavelength radar, defined as deferential frequency ratio (DFR), to the D(sub 0) in the melting region are given as a function of the distance from the 0 C isotherm. A self-consistent iterative algorithm that shows a promising to account for rain attenuation of radar and infer the DSD without use of surface reference technique (SRT) is examined by applying it to the apparent radar reflectivity profiles simulated from the DSD model and then comparing the estimates with the model (true) results. For light to moderate rain the self-consistent rain profiling approach converges to unique and correct solutions only if the same shape factors of Gamma functions are used both to generate and retrieve the rain profiles, but does not converges to the true solutions if the DSD form is not chosen correctly. To further examine the dual-wavelength techniques, the self-consistent algorithm, along with forward and backward rain profiling algorithms, is then applied to the measurements taken from the 2nd generation Precipitation Radar (PR-2) built by Jet Propulsion Laboratory. It is found that rain profiles estimated from the forward and backward approaches are not sensitive to shape factor of DSD Gamma distribution, but the self-consistent method is.

Liao, Liang↗