Search NASA⌕ Search

SEARCH · Search NASA

Results for “Generalized Linear Model”

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 1,225 records · Page 68

Proteus two-dimensional Navier-Stokes computer code, version 2.0. Volume 2: User's guide

A computer code called Proteus 2D was developed to solve the two-dimensional planar or axisymmetric, Reynolds-averaged, unsteady compressible Navier-Stokes equations in strong conservation law form. The objective in this effort was to develop a code for aerospace propulsion applications that is easy to use and easy to modify. Code readability, modularity, and documentation were emphasized. The governing equations are solved in generalized nonorthogonal body-fitted coordinates, by marching in time using a fully-coupled ADI solution procedure. The boundary conditions are treated implicitly. All terms, including the diffusion terms, are linearized using second-order Taylor series expansions. Turbulence is modeled using either an algebraic or two-equation eddy viscosity model. The thin-layer or Euler equations may also be solved. The energy equation may be eliminated by the assumption of constant total enthalpy. Explicit and implicit artificial viscosity may be used. Several time step options are available for convergence acceleration. The documentation is divided into three volumes. This is the User's Guide, and describes the program's features, the input and output, the procedure for setting up initial conditions, the computer resource requirements, the diagnostic messages that may be generated, the job control language used to run the program, and several test cases.

Towne, Charles E.↗

Proteus two-dimensional Navier-Stokes computer code, version 2.0. Volume 1: Analysis description

A computer code called Proteus 2D was developed to solve the two-dimensional planar or axisymmetric, Reynolds-averaged, unsteady compressible Navier-Stokes equations in strong conservation law form. The objective in this effort was to develop a code for aerospace propulsion applications that is easy to use and easy to modify. Code readability, modularity, and documentation were emphasized. The governing equations are solved in generalized nonorthogonal body-fitted coordinates, by marching in time using a fully-coupled ADI solution procedure. The boundary conditions are treated implicitly. All terms, including the diffusion terms, are linearized using second-order Taylor series expansions. Turbulence is modeled using either an algebraic or two-equation eddy viscosity model. The thin-layer or Euler equations may also be solved. The energy equation may be eliminated by the assumption of constant total enthalpy. Explicit and implicit artificial viscosity may be used. Several time step options are available for convergence acceleration. The documentation is divided into three volumes. This is the Analysis Description, and presents the equations and solution procedure. The governing equations, the turbulence model, the linearization of the equations and boundary conditions, the time and space differencing formulas, the ADI solution procedure, and the artificial viscosity models are described in detail.

Towne, Charles E.↗

Proteus three-dimensional Navier-Stokes computer code, version 1.0. Volume 2: User's guide

A computer code called Proteus 3D was developed to solve the three-dimensional, Reynolds-averaged, unsteady compressible Navier-Stokes equations in strong conservation law form. The objective in this effort was to develop a code for aerospace propulsion applications that is easy to use and easy to modify. Code readability, modularity, and documentation were emphasized. The governing equations are solved in generalized nonorthogonal body-fitted coordinates, by marching in time using a fully-coupled ADI solution procedure. The boundary conditions are treated implicitly. All terms, including the diffusion terms, are linearized using second-order Taylor series expansions. Turbulence is modeled using either an algebraic or two-equation eddy viscosity model. The thin-layer or Euler equations may also be solved. The energy equation may be eliminated by the assumption of constant total enthalpy. Explicit and implicit artificial viscosity may be used. Several time step options are available for convergence acceleration. The documentation is divided into three volumes. This User's Guide describes the program's features, the input and output, the procedure for setting up initial conditions, the computer resource requirements, the diagnostic messages that may be generated, the job control language used to run the program, and several test cases.

Towne, Charles E.↗

Proteus two-dimensional Navier-Stokes computer code, version 2.0. Volume 3: Programmer's reference

A computer code called Proteus 2D was developed to solve the two-dimensional planar or axisymmetric, Reynolds-averaged, unsteady compressible Navier-Stokes equations in strong conservation law form. The objective in this effort was to develop a code for aerospace propulsion applications that is easy to use and easy to modify. Code readability, modularity, and documentation were emphasized. The governing equations are solved in generalized nonorthogonal body-fitted coordinates, by marching in time using a fully-coupled ADI solution procedure. The boundary conditions are treated implicitly. All terms, including the diffusion terms, are linearized using second-order Taylor series expansions. Turbulence is modeled using either an algebraic or two-equation eddy viscosity model. The thin-layer or Euler equations may also be solved. The energy equation may be eliminated by the assumption of constant total enthalpy. Explicit and implicit artificial viscosity may be used. Several time step options are available for convergence acceleration. The documentation is divided into three volumes. The Programmer's Reference contains detailed information useful when modifying the program. The program structure, the Fortran variables stored in common blocks, and the details of each subprogram are described.

Towne, Charles E.↗

Microstructurally based model of fatigue initiation and growth

A model to calculate fatigue life is developed based on the assumption that fatigue life is entirely composed of crack growth from an initial microstructural inhomogeneity. Specifically, growth is considered to start from either an ellipsoidal void, a cracked particle, or a debonded particle. The capability of predicting fatigue life from material microstructure is based on linear elastic fracture mechanics principles, the sizes of the crack-initiating microstructural inhomogeneities, and an initiation parameter that is proportional to the cyclic plastic zone size. A key aspect of this modeling approach is that it is linked with a general purpose probability program to analyze the effect of the distribution of controlling microstructural features within the material. This enables prediction of fatigue stress versus life curves for various specimen geometries using distributional statistics obtained from characterizations of the microstructure. Results are compared to experimental fatigue data from an aluminum alloy.

Brockenbrough, J. R.↗

Application of Parametrized Post-Newtonian Methods to the Gravitational IS of Satellite Energy Exchange Data

Project Satellite Energy Exchange (SEE) is a free-flying, high altitude satellite that utilizes space to construct a passive, low-temperature, nano-g environment in order to accurately measure the poorly known gravitational constant G plus other gravitational parameters that are difficult to measure in an earth-based laboratory. Eventually data received from SEE must be analyzed using a model of the gravitational interaction including parameters that describe deviations from general relativity and experiment. One model that can be used to fit tile data is the Parametrized post- Newtonian (PPN) approximation of general relativity (GR) which introduces ten parameters which have specified values in (GR). It is the lowest-order, consistent approximation that contains non linear terms. General relativity predicts that the Robertson parameters, gamma (light deflection), and beta (advance of the perihelion), are both 1 in GR. Another eight parameters, alpha(sub k), k=1,2,3 and zeta(sub k), k=1,2,3,4 and Xi are all zero in GR. Non zero values for alpha(sub k) parameters predict preferred frame effects; for zeta(sub k) violations of globally conserved quantities such as mass, momentum and angular momentum; and for Xi a contribution from the Whitehead theory of gravitation, once thought to be equivalent to GR. In addition, there is the possibility that there may be a preferred frame for the universe. If such a frame exists, then all observers must measure the velocity omega of their motion with respect to this universal rest frame. Such a frame is somewhat reminiscent of the concept of the ether which was supposedly the frame in which the velocity of light took the value c predicted by special relativity. The SEE mission can also look for deviations from the r(exp -2) law of Newtonian gravity, adding parameters alpha and lamda for non Newtonian behavior that describe the magnitude and range of the r(exp -2) deviations respectively. The foundations of the GR supposedly agree with Newtonian gravity to first order so that the parameters alpha and lamda are zero in GR. More important, however, GR subsequently depends on this Newtonian approximation to build up the non linear higher-order terms which forms the basis of the PPN frame work.

Smalley, Larry L.↗

Modeling and Analysis of Large Amplitude Flight Maneuvers

Analytical methods for stability analysis of large amplitude aircraft motion have been slow to develop because many nonlinear system stability assessment methods are restricted to a state-space dimension of less than three. The proffered approach is to create regional cell-to-cell maps for strategically located two-dimensional subspaces within the higher-dimensional model statespace. These regional solutions capture nonlinear behavior better than linearized point solutions. They also avoid the computational difficulties that emerge when attempting to create a cell map for the entire state-space. Example stability results are presented for a general aviation aircraft and a micro-aerial vehicle configuration. The analytical results are consistent with characteristics that were discovered during previous flight-testing.

Anderson, Mark R.↗

Data-Driven Learning for the Mori--Zwanzig Formalism: A Generalization of the Koopman Learning Framework

A theoretical framework which unifies the conventional Mori--Zwanzig formalism and the approximate Koopman learning of deterministic dynamical systems from noiseless observation is presented. In this framework, the Mori--Zwanzig formalism, developed in statistical mechanics to tackle the hard problem of construction of reduced-order dynamics for high-dimensional dynamical systems, can be considered as a natural generalization of the Koopman description of the dynamical system. We next show that, similar to the approximate Koopman learning methods, data-driven methods can be developed for the Mori--Zwanzig formalism with Mori's linear projection operator. We have developed two algorithms to extract the key operators, the Markov and the memory kernel, using time series of a reduced set of observables in a dynamical system. We have adopted the Lorenz `96 system as a test problem and solved for the above operators. These operators exhibit complex behaviors, which are unlikely to be captured by traditional modeling approaches in Mori--Zwanzig analysis. The nontrivial generalized fluctuation-dissipation relationship, which relates the memory kernel with the two-time correlation statistics of the orthogonal dynamics, was numerically verified as a validation of the solved operators. Here we present numerical evidence that the generalized Langevin equation, a key construct in the Mori--Zwanzig formalism, is more advantageous in predicting the evolution of the reduced set of observables than the conventional approximate Koopman operators.

97 MATHEMATICS AND COMPUTING↗

Acoustic Codes in 2D Spherical Coordinate

Finite-difference methods are widely used to simulate infrasound propagation in the atmosphere. Flexibility of finite-difference scheme allows implementation of highly heterogeneous media for sound propagation as well as complex source models for sound generation. While full 3-D finite-difference methods have been utilized for local infrasound propagation with pronounced topography, 2-D modeling approach has been preferred for regional and global propagation as full 3-D methods generally require enormous computational resources. Infrasound propagation is often simulated with a second-order finite difference scheme. This lowest-order finite-difference scheme is robust and straightforward to implement complex boundary conditions, but the solution includes large error with numerical dispersion and dissipation. This large numerical error may make the second-order finite-difference not optimal for long range infrasound propagation modeling as the numerical dispersion degrades the accuracy of the solution unacceptably. Here, we developed a high-order finite-difference solver for long-range infrasound simulation. The high-order scheme is particularly popular for linear wave modeling in aeroacoustics owing to its low-dispersive and low-dissipative behavior. We develop and evaluate a high-order finite difference scheme in 2-D axisymmetric coordinates. The axisymmetry allows to approximate 3-D spherical sound propagation and amplitude attenuation by a 2-D method. AC2Dr is developed to simulate infrasound propagation in realistic atmosphere, but can be used for linear acoustic waves in general materials with background flow. AC2Dr in an axisymmetric coordinates allows for spherical radiation of acoustic waves from compact sources.

Sjogreen, Bjorn↗

Verification and application of resonance broadened quasi-linear (RBQ) model with multiple Alfvénic instabilities

The resonance broadened quasilinear (RBQ) model for the problem of relaxing the hot ion distribution function in constant-of-motion 3D space [Gorelenkov et al., Nucl. Fusion 58, 082016 (2018)] is presented with the self-consistent evolution of multiple Alfvén eigenmode amplitudes. The RBQ model represents the generalization of the earlier published model [Berk et al., Nucl. Fusion 35, 1661 (1995)] by carefully examining the wave particle interaction in the presence of realistic Alfvén eigenmode (AE) structures and pitch angle scattering with the help of the guiding center code ORBIT. One aspect of the generalization is that the RBQ model goes beyond the local perturbative-pendulumlike approximation for the wave particle dynamics near the resonance. An iterative procedure is introduced to account for eigenstructures varying within the resonances. It is found that a radially localized mode structure implies a saturation level 2–3 times smaller than that predicted by an earlier bump-on-tail quasilinear model that employed uniform mode structures. We apply the RBQ code to a DIII-D plasma with an elevated q-profile where the beam ion profiles exhibit stiff transport properties [Collins et al., Phys. Rev. Lett. 116, 095001 (2016)]. Finally, the properties of AE driven fast ion distribution relaxation are studied for validations of the applied RBQ model in DIII-D discharges. Initial results show that the model is robust, is numerically efficient, and can predict fast ion relaxation in present and future burning plasma experiments.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

An approximation technique for jet impingement flow

The analytical approximate solution of a non-linear jet impingement flow model will be demonstrated. We will show that this is an improvement over the series approximation obtained via the Adomian decomposition method, which is itself, a powerful method for analysing non-linear differential equations. The results of these approximations will be compared to the Runge-Kutta approximation in order to demonstrate their validity.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Large-scale white noise and cosmology

The generation of white noise on large scales is a generic property of the dynamics of physical systems described by local nonlinear partial differential equations. Nonlinearities prevent the small-scale dynamics from being erased by smoothing. Unresolved small-scale dynamics act as an uncorrelated (white or Poissonian) noise (seemingly stochastic but actually deterministic) contribution to large-scale dynamics. This white noise exists even when the dynamics is very nearly linear. In cases where the power spectrum is sub-Poissonian on large scales, this noise will dominate on the largest scale power no matter the amplitude of the inhomogeneities. Such is the case in the standard model of cosmology, where the primordial density power spectrum is expected to have an almost Harrison-Zel’dovich, P [ k ] ∼ k , spectrum on a much broader range of scales than can be observed. Even though linear gravitational evolution dominates nonlinear corrections by a factor of ∼ 10 5 , the nonobservation of white noise on the Hubble scale precludes the extrapolation of this power law below the comoving 1 pc scale. More generally, observation or nonobservation of large-scale white noise provides a powerful probe of the Universe on very small scales in the early Universe. Gravitational radiation, phase transitions, vorticity, and running of the spectral index are all phenomena that can be probed with large-scale white noise. Large-scale white noise is a nonoptional feature of all cosmological models but one which has not heretofore been appreciated.

Barenboim, Gabriela [Valencia U., IFIC; Valencia U↗

Action-angle formulation of generalized, orbit-based, fast-ion diagnostic weight functions

Due to the usually complicated and anisotropic nature of the fast-ion distribution function, diagnostic velocity-space weight functions, which indicate the sensitivity of a diagnostic to different fast-ion velocities, are used to facilitate the analysis of experimental data. Additionally, when velocity-space weight functions are discretized, a linear equation relating the fast-ion density and the expected diagnostic signal is formed. In a technique known as velocity-space tomography, many measurements can be combined to create an ill-conditioned system of linear equations that can be solved using various computational methods. However, when velocity-space weight functions (which by definition ignore spatial dependencies) are used, velocity-space tomography is restricted, both by the accuracy of its forward model and also by the availability of spatially overlapping diagnostic measurements. In this work, we extend velocity-space weight functions to a full 6D generalized coordinate system and then show how to reduce them to a 3D orbit-space without loss of generality using an action-angle formulation. Moreover, we show how diagnostic orbit-weight functions can be used to infer the full fast-ion distribution function, i.e., orbit tomography. In depth derivations of orbit weight functions for the neutron, neutral particle analyzer, and fast-ion D-α diagnostics are also shown

74 ATOMIC AND MOLECULAR PHYSICS↗

Light-Front Holography, Light-Front Wavefunctions, and Novel QCD Phenomena

Light-Front Holography is one of the most remarkable features of the AdS/CFT correspondence. In spite of its present limitations it provides important physical insights into the nonperturbative regime of QCD and its transition to the perturbative domain. This novel framework allows hadronic amplitudes in a higher dimensional anti-de Sitter (AdS) space to be mapped to frame-independent light-front wavefunctions of hadrons in physical space-time. The model leads to an effective confining light-front QCD Hamiltonian and a single-variable light-front Schroedinger equation which determines the eigenspectrum and the light-front wavefunctions of hadrons for general spin and orbital angular momentum. The coordinate z in AdS space is uniquely identified with a Lorentz-invariant coordinate {zeta} which measures the separation of the constituents within a hadron at equal light-front time and determines the off-shell dynamics of the bound-state wavefunctions, and thus the fall-off as a function of the invariant mass of the constituents. The soft-wall holographic model modified by a positive-sign dilaton metric, leads to a remarkable one-parameter description of nonperturbative hadron dynamics - a semi-classical frame-independent first approximation to the spectra and light-front wavefunctions of meson and baryons. The model predicts a Regge spectrum of linear trajectories with the same slope in the leading orbital angular momentum L of hadrons and the radial quantum number n. The hadron eigensolutions projected on the free Fock basis provides the complete set of valence and non-valence light-front Fock state wavefunctions {Psi}{sub n/H} (x{sub i}, k{sub {perpendicular}i}, {lambda}{sub i}) which describe the hadron's momentum and spin distributions needed to compute the direct measures of hadron structure at the quark and gluon level, such as elastic and transition form factors, distribution amplitudes, structure functions, generalized parton distributions and transverse momentum distributions. The effective confining potential also creates quark-antiquark pairs from the amplitude q {yields} q{bar q}q. Thus in holographic QCD higher Fock states can have any number of extra q{bar q} pairs. We discuss the relevance of higher Fock-states for describing the detailed structure of space and time-like form factors. The AdS/QCD model can be systematically improved by using its complete orthonormal solutions to diagonalize the full QCD light-front Hamiltonian or by applying the Lippmann-Schwinger method in order to systematically include the QCD interaction terms. A new perspective on quark and gluon condensates is also obtained.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Zonal flow as pattern formation

Zonal flows are well known to arise spontaneously out of turbulence. We show that for statistically averaged equations of the stochastically forced generalized Hasegawa-Mima model, steady-state zonal flows, and inhomogeneous turbulence fit into the framework of pattern formation. There are many implications. First, the wavelength of the zonal flows is not unique. Indeed, in an idealized, infinite system, any wavelength within a certain continuous band corresponds to a solution. Second, of these wavelengths, only those within a smaller subband are linearly stable. Unstable wavelengths must evolve to reach a stable wavelength; this process manifests as merging jets.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Constructing tensor network influence functionals for general quantum dynamics

Here, we describe an iterative formalism to compute influence functionals that describe the general quantum dynamics of a subsystem beyond the assumption of linear coupling to a quadratic bath. We use a space-time tensor network representation of the influence functional and investigate its approximability in terms of its bond dimension and time-like entanglement in the tensor network description. We study two numerical models, the spin-boson model and a model of interacting hard-core bosons in a 1D harmonic trap. We find that the influence functional and the intermediates involved in its construction can be efficiently approximated by low bond dimension tensor networks in certain dynamical regimes, which allows the quantum dynamics to be accurately computed for longer times than with direct time evolution methods. However, as one iteratively integrates out the bath, the correlations in the influence functional can first increase before decreasing, indicating that the final compressibility of the influence functional is achieved via non-trivial cancellation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

TOUGH3: A new efficient version of the TOUGH suite of multiphase flow and transport simulators

The TOUGH suite of nonisothermal multiphase flow and transport simulators has been updated by various developers over many years to address a vast range of challenging subsurface problems. The increasing complexity of the simulated processes as well as the growing size of model domains that need to be handled call for an improvement in the simulator's computational robustness and efficiency. Moreover, modifications have been frequently introduced independently, resulting in multiple versions of TOUGH that (1) led to inconsistencies in feature implementation and usage, (2) made code maintenance and development inefficient, and (3) caused confusion to users and developers. TOUGH3—a new base version of TOUGH—addresses these issues. It consolidates both the serial (TOUGH2 V2.1) and parallel (TOUGH2-MP V2.0) implementations, enabling simulations to be performed on desktop computers and supercomputers using a single code. New PETSc parallel linear solvers are added to the existing serial solvers of TOUGH2 and the Aztec solver used in TOUGH2-MP. The PETSc solvers generally perform better than the Aztec solvers in parallel and the internal TOUGH3 linear solver in serial. TOUGH3 also incorporates many new features, addresses bugs, and improves the flexibility of data handling. So due to the improved capabilities and usability, TOUGH3 is more robust and efficient for solving tough and computationally demanding problems in diverse scientific and practical applications related to subsurface flow modeling.

97 MATHEMATICS AND COMPUTING↗

A computational study of an atomizing liquid sheet

Linear instability predictions of liquid sheets injected into a gas medium are well established in the literature. These analyses are often used in Lagrangian-Eulerian spray simulations, a prominent simulation method, to model the dynamics occurring in the near-nozzle region. In the present work, these instability predictions are re-examined by first generalizing the treatment of interfacial conditions and related assumptions with a two-phase Orr-Sommerfeld (OS) system, and second, by employing highly resolved-Volume-of-Fluid (VoF) simulations. After presenting some validation exercises for both the VoF and OS solvers, the OS predictions are compared to earlier studies from the literature leading to reasonable agreement in the limit as the boundary layer thickness tends to zero. Results from VoF simulations of liquid sheet injection are used to characterize the range of scales of the liquid structures immediately before atomization. The mean value in this range is found to be approximately two to three orders of magnitude larger than the corresponding predictions from previous studies. A two-phase mixing layer under the same physical conditions is used to examine this disparity, revealing that within the linear regime, relatively good agreement exists between the VoF and OS predicted instability mechanisms. However, the most unstable mode in the linear regime is too small to cause a fracture or atomization of the liquid sheet and hence cannot be directly responsible for the atomization. The generation of a much larger mode, which emerges well beyond the linear regime, is the one causing breakup.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗