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 1,243 records · Page 69

Study on the well-posedness, convergence and the stability of the semi-implicit upwind numerical solver for the multi-fluid model

The well-posedness, convergence and the stability of the two-fluid code has been studied for a long time. Most of the investigations concern the semi-implicit upwind solution scheme for the six equation two-fluid model such as used in RELAP5 3 or TRACE 21. Since the system code, SPACE 2, adopts one more field, a droplet field, it consists of nine equations (3 mass, 3 momentum and 3 energy balance equations) and thus more involved investigations are necessary to confirm the stability and convergence. For this objective, the old issue of the well-posedness, convergence and the stability is revisited and some general guidelines to develop a well-posed numerical multi-fluid model are derived as follows; (1) Hyperbolicity of the corresponding system of partial differential equations is not a necessary condition for the development of a numerical model for multi-phase flow, but whether or not it is hyperbolic can provide guidance relative to initial conditions, boundary conditions, and expected high frequency behavior of the model. (2) A necessary condition for a well-posed numerical model is stability in the von Neumann sense, i.e. growth factor less than 1.0 for the shortest wave-length, 2{Delta}x. (3) The smallest node size used for convergence studies should be of the order of the characteristic dimension of the average description, i.e. smaller nodes can be used so long as they do not result in unphysical growth of wave-lengths less than the characteristic dimension. The usual mathematical definition of convergence i.e. the behavior of the solution as the node size approaches zero, is not appropriate for the discrete averaged numerical model, since there is diminished physical meaning to behavior at wavelengths less than the characteristic dimension of the average description. Under these guidelines, dispersion analysis and von Neumann stability analysis are performed for the three field multi-fluid, semi-implicit, upwind numerical model to show that the necessary conditions for well-posedness are met. To study the non-linear stability and the convergence, various runs with the torus problem using the SPACE code are utilized to confirm the frequency cascading effects that augment non-linear stability. A robust mechanism for the flow regime change is also a very important factor for developing non-linearly stable and convergent code. A open pipe flow problem is also simulated to investigate the non-linear stability effects in a flow geometry more typical of real safety code applications. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

High frequency impedance characteristics of a tunable microplasma device

Computational studies on high frequency impedance characteristics of a microplasma device are reported. While microplasma is ignited using a primary excitation signal, frequency response of plasma impedance is determined by a secondary high frequency probe signal with significantly lower voltage amplitude such that it does not influence the plasma parameters. Here, the computational model utilizing the drift–diffusion approximation is first validated by comparing with experimental data for microplasmas ignited at pressures ranging from 1 to 5 Torr. In spite of quantitative discrepancies, good overall agreement is obtained between the measured frequency response of impedance of the discharge. Comparisons are also presented for various plasma parameters including mean electron number density, sheath thickness, mean electron temperature, and collision frequency that were inferred from the impedance measurements. The computational model is then used to perform simulations of near-atmospheric pressure microplasmas with the probe signal frequency ranging from 3 to 20 GHz. The simulations demonstrate the presence of a resonance frequency at which the impedance vanishes. More importantly, it is shown that this resonant frequency can be tuned effectively by suitably modifying the operating parameters (gap size, pressure, and excitation voltage). The simulated impedance characteristics are used to determine the effective plasma inductance and capacitance using a non-linear fitting approach, thereby showing the dependence of these electrical parameters on the plasma operating conditions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Large deformations of Tr(Φ 3 ) and the world at infinity

The amplitudes of the non-linear sigma model can be obtained from those of Tr(Φ 3 ) theory by sending the kinematic (Mandelstam) variables to infinity in a certain direction. In this paper we characterize the behavior of Tr(Φ 3 ) amplitudes under a general class of large kinematic shifts called g-vector shifts. The objects that live in this world at infinity retain certain key amplitude-like properties, most notably factorization, and admit descriptions in terms of polytopes, but they are not generally amplitudes of any cognizable theory. We identify particular g-vector shifts that lead at infinity to mixed amplitudes involving two pions and any number of scalars, allowing us to provide polytopal descriptions of these amplitudes.

effective field theories↗

Reducing parametric backscattering by polarization rotation

When a laser passes through underdense plasmas, Raman and Brillouin Backscattering can reflect a substantial portion of the incident laser energy. This is a major loss mechanism, for example, in employing lasers in inertial confinement fusion. But, by slow rotation of the incident linear polarization, the overall reflectivity can be reduced significantly. Particle in cell simulations show that, for parameters similar to those of indirect drive fusion experiments, polarization rotation reduces the reflectivity by a factor of 5. A general, fluid-model based analytical estimation for the reflectivity reduction agrees with simulations. However, in identifying the source of the backscatter reduction, it is difficult to disentangle the rotating polarization from the frequency separation based approach used to engineer the beam's polarization. Though the backscatter reduction arises similarly to other approaches that employ frequency separation, in the case here, the intensity remains constant in time.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The invariance of the total direct DNA strand break yield

Purpose: The invariance of the total direct strand break yield when DNA is irradiated by different types of particles and energies has been reported by previous works. This study is intended to explain the physical causes of this behavior. Methods: The geant4-dna extension of the geant4 general purpose Monte Carlo simulation toolkit has been used to determine direct strand break yields induced by protons and alpha particles impacting on a B-DNA geometrical model, including five organization levels of the human genetic material. The linear energy transfer (LET) of such particles ranges from 4.8 keV/{mu}m (10 MeV protons) to about 235 keV/{mu}m (2 MeV alpha particles), at 5.225 {mu}m depth (near the center of the region of interest). Direct total, single and double strand break probabilities have been determined in a liquid water homogeneous medium with a 1.06 g/cm {sup 3} density. The energetic spectra of single strand breaks (SSB), the number of energy deposition events, and the SSB/event ratio were determined. Results: The target-hit probability was found to be independent of both the type and the energy of the incident particle, even if this latter is a secondary electron. This probability is determined by the geometrical properties of the system. The total strand break yield and the number of energy deposition events required to reach a certain absorbed dose were found nearly independent of the type and energy of the incident ion (proton or alpha). In contrast, the double strand break (DSB) yield was found strongly dependent on the LET of the incident radiation. Conclusions: The SSB generation process is homogeneous and independent of the LET of the particles involved, at least within the proton and alpha particle energy range here studied. The target-hit probability is only determined by the ratio between the total volume occupied by targets and that of the ROI where the radiation deposits its energy. The maximum separation distance between two adjacent SSBs to produce a DSB is the parameter that breaks the homogeneity of the target-hit process, making the DSB production process strongly heterogeneous.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗

SCEPTRE v. 2.0 (Sandia Computational Engine for Particle Transport for Radiat

SCEPTRE solves the linear Boltzmann transport equation for general one-, two- and three-dimensional unstructured-mesh Cartesian geometries. SCEPTRE can handle any particle type for which multigroup-Legendre cross sections are available. However, the code is designed primarily to model the transport of photons, electrons, and positrons through matter. SCEPTRE 2.0 includes several auxiliary packages for pre- and post-processing, regression testing, stochastic radiation transport, thermal radiation transport, and charged-particle transport in the presence of electromagnetic fields. SCEPTRE 2.0 includes sophisticated transport-synthetic acceleration and ray-effects mitigation capabilities.

Pautz, Shawn↗

Impacts of the horizontal and vertical grids on the numerical solutions of the dynamical equations – Part 1: Nonhydrostatic inertia–gravity modes

We have used a normal-mode analysis to investigate the impacts of the horizontal and vertical discretizations on the numerical solutions of the nonhydrostatic anelastic inertia–gravity modes on a midlatitude f plane. The dispersion equations are derived from the linearized anelastic equations that are discretized on the Z, C, D, CD, (DC), A, E and B horizontal grids, and on the L and CP vertical grids. The effects of both horizontal grid spacing and vertical wavenumber are analyzed, and the role of nonhydrostatic effects is discussed. We also compare the results of the normal-mode analyses with numerical solutions obtained by running linearized numerical models based on the various horizontal grids. The sources and behaviors of the computational modes in the numerical simulations are also examined. Our normal-mode analyses with the Z, C, D, A, E and B grids generally confirm the conclusions of previous shallow-water studies for the cyclone-resolving scales (with low horizontal wavenumbers). We conclude that, aided by nonhydrostatic effects, the Z and C grids become overall more accurate for cloud-resolving resolutions (with high horizontal wavenumbers) than for the cyclone-resolving scales. A companion paper, Part 2, discusses the impacts of the discretization on the Rossby modes on a midlatitude β plane.

58 GEOSCIENCES↗

Equicontrollability and the model following problem

Equicontrollability and its application to the linear time-invariant model-following problem are discussed. The problem is presented in the form of two systems, the plant and the model. The requirement is to find a controller to apply to the plant so that the resultant compensated plant behaves, in an input-output sense, the same as the model. All systems are assumed to be linear and time-invariant. The basic approach is to find suitable equicontrollable realizations of the plant and model and to utilize feedback so as to produce a controller of minimal state dimension. The concept of equicontrollability is a generalization of control canonical (phase variable) form applied to multivariable systems. It allows one to visualize clearly the effects of feedback and to pinpoint the parameters of a multivariable system which are invariant under feedback. The basic contributions are the development of equicontrollable form; solution of the model-following problem in an entirely algorithmic way, suitable for computer programming; and resolution of questions on system decoupling.

Curran, R. T.↗

Critical issues in viscous flow computations

In developing computer programs to numerically solve the Navier-Stokes equations, the purpose of the computation must be clearly kept in mind. In the Air Force, the purpose is to provide design information on non-linear aerodynamic phenomenon for aircraft that perform throughout the flight corridor. This translates into the requirement for a computer program which can solve the time averaged compressible Navier-Stokes equations (with a turbulence model) in three dimensions for generalized geometries. The intended application of the results then controls the priorities in addressing critical issues. Recurrent problem areas encountered in the study of viscous flow include: (1) grid generation for arbitrary geometry; (2) numerical difficulties; (3) turbulence models; (4) accuracy and efficiency; and (5) smearing of discontinuities.

Hankey, W. L., Jr.↗

A frost formation model and its validation under various experimental conditions

A numerical model that was used to calculate the frost properties for all regimes of frost growth is described. In the first regime of frost growth, the initial frost density and thickness was modeled from the theories of crystal growth. The 'frost point' temperature was modeled as a linear interpolation between the dew point temperature and the fog point temperature, based upon the nucleating capability of the particular condensing surfaces. For a second regime of frost growth, the diffusion model was adopted with the following enhancements: the generalized correlation of the water frost thermal conductivity was applied to practically all water frost layers being careful to ensure that the calculated heat and mass transfer coefficients agreed with experimental measurements of the same coefficients.

Dietenberger, M. A.↗

Derivation and definition of a linear aircraft model

A linear aircraft model for a rigid aircraft of constant mass flying over a flat, nonrotating earth is derived and defined. The derivation makes no assumptions of reference trajectory or vehicle symmetry. The linear system equations are derived and evaluated along a general trajectory and include both aircraft dynamics and observation variables.

Duke, Eugene L.↗

Microgravity vibration isolation: Optimal preview and feedback control

In order to achieve adequate low-frequency vibration isolation for certain space experiments an active control is needed, due to inherent passive-isolator limitations. Proposed here are five possible state-space models for a one-dimensional vibration isolation system with a quadratic performance index. The five models are subsets of a general set of nonhomogeneous state space equations which includes disturbance terms. An optimal control is determined, using a differential equations approach, for this class of problems. This control is expressed in terms of constant, Linear Quadratic Regulator (LQR) feedback gains and constant feedforward (preview) gains. The gains can be easily determined numerically. They result in a robust controller and offers substantial improvements over a control that uses standard LQR feedback alone.

Hampton, R. D.↗

Probabilistic nonlinear finite element analysis of composite structures

A probabilistic finite element analysis procedure for laminated composite shells is developed. A total Lagrangian finite element formulation, employing a degenerated three-dimensional laminated composite shell element with the full Green-Lagrange strains and first-order shear deformable kinematics, is used. The first-order second-moment technique for probabilistic finite element analysis of random fields is employed, and results are presented in the form of mean and variance of the structural response. Reliability calculations are made by using the first-order reliability method combined with sensitivity derivatives from the finite element analysis. Both ply-level and micromechanics-level random variables are incorporated, the latter by means of the Aboudi micromechanics model. Two sample problems are solved to verify the accuracy of the procedures developed and to quantify the variability of certain material type/structure combinations. In general, the procedure is quite effective in determining the response statistics and reliability for linear and geometric nonlinear behavior of laminated composite shells.

Engelstad, S. P.↗

Analytic MHD Theory for Earth's Bow Shock at Low Mach Numbers

A previous MHD theory for the density jump at the Earth's bow shock, which assumed the Alfven M(A) and sonic M(s) Mach numbers are both much greater than 1, is reanalyzed and generalized. It is shown that the MHD jump equation can be analytically solved much more directly using perturbation theory, with the ordering determined by M(A) and M(s), and that the first-order perturbation solution is identical to the solution found in the earlier theory. The second-order perturbation solution is calculated, whereas the earlier approach cannot be used to obtain it. The second-order terms generally are important over most of the range of M(A) and M(s) in the solar wind when the angle theta between the normal to the bow shock and magnetic field is not close to 0 deg or 180 deg (the solutions are symmetric about 90 deg). This new perturbation solution is generally accurate under most solar wind conditions at 1 AU, with the exception of low Mach numbers when theta is close to 90 deg. In this exceptional case the new solution does not improve on the first-order solutions obtained earlier, and the predicted density ratio can vary by 10-20% from the exact numerical MHD solutions. For theta approx. = 90 deg another perturbation solution is derived that predicts the density ratio much more accurately. This second solution is typically accurate for quasi-perpendicular conditions. Taken together, these two analytical solutions are generally accurate for the Earth's bow shock, except in the rare circumstance that M(A) is less than or = 2. MHD and gasdynamic simulations have produced empirical models in which the shock's standoff distance a(s) is linearly related to the density jump ratio X at the subsolar point. Using an empirical relationship between a(s) and X obtained from MHD simulations, a(s) values predicted using the MHD solutions for X are compared with the predictions of phenomenological models commonly used for modeling observational data, and with the predictions of a modified phenomenological model proposed recently. The similarities and differences between these results are illustrated using plots of X and a(s) predicted for the Earth's bow shock. The plots show that the new analytic solutions agree very well with the exact numerical MHD solutions and that these MHD solutions should replace the corresponding phenomenological relations in comparisons with data. Furthermore, significant differences exist between the standoff distances predicted at low M(A) using the MHD models versus those predicted by the new modified phenomenological model. These differences should be amenable to observational testing.

Grabbe, Crockett L.↗

Alternative Representations of Convective Processes in the NASA GEOS-5 AGCM

The gap in explicit resolution of phenomena between global climate models and cloud resolving models is shrinking at a steady pace with global integrations of several km in resolution now practical for at least some time scales. In moving toward finer resolution the long standing problem of convective parameterization is being examined along with the assumption of convective quasi-equilibrium and how this can be reconciled with the stochastic and intermittent nature of convection. In this context we examine the nature of parameterized convection in the NASA Goddard Earth Observing System (GEOS-5) Atmospheric General Circulation Model. Our analysis uses both coarse (2.5 degree) and fine scale 0.25 degree spatial resolution integrations. Two basic formulations are compared: the default option is the Relaxed Arakawa-Schubert (RAS) scheme which invokes a sequence of linearly entraining plumes and quasi-equilibrium closure. An optional modification of this method (the "Stochastic Tokioka" constraint) places a random lower limit on plume entrainment. An alternative representation is the Kain-Fritsch parameterization which was originally developed for mesoscale numerical modeling strategies. Here entrainment is determined by a crude buoyancy sorting approach that allows the plume spectrum to be more responsive to ambient vertical stratification of moisture. Diagnostics of the model behavior are referenced to recent observational evidence of continuous phase transition behavior. In particular we examine the relationship between column water vapor and probablility of convective presence and intensity. Sensitivity of the statistics of convective behavior to parcel mixing/entrainment formulations and parcel initial thermodynamics are considered. Observational statistics from A-Train and TRMM sensors provide validation of the model integrations.

Robertson, Franklin↗

Constraining Dvali-Gabadadze-Porrati gravity from observational data

The accelerating expansion of our Universe at present could be driven by an unknown energy component (dark energy) or a modification of general relativity (modified gravity). In this paper we revisit the constraints on a phenomenological model which interpolates between the pure {lambda}CDM model and the Dvali-Gabadadze-Porrati braneworld model with an additional parameter {alpha}. Combining the cosmic microwave background , baryon acoustic oscillations, and type Ia supernovae, as well as some high-redshift observations, such as the gamma-ray bursts and the measurements of linear growth factors, we obtain the tight constraint on the parameter {alpha}=0.254{+-}0.153 (68% C.L.), which implies that the flat Dvali-Gabadadze-Porrati model is incompatible with the current observations, while the pure {lambda}CDM model still fits the data very well. Finally, we simulate the future measurements with higher precisions and find that the constraint on {alpha} can be improved by a factor of 2 when compared to the present constraints.

99 GENERAL AND MISCELLANEOUS↗

Mixed-strategy approach to band-edge analysis and modeling in semiconductors

In semiconductor physics, the Urbach rule describes an exponential dependence of the UV-Vis absorption coefficient on photon energy at the band edge. The rule is so ubiquitously applied that a single common mechanism, variability in the local potential, is generally considered to be the cause of exponential band edges. Variability in the local potential may be caused by alloy inhomogeneity, structural defects, impurities, thermal excitation, or any other mechanism that interrupts ideal crystallinity and creates disorder. A crucial point of the Urbach rule is its development for band edges which approach linearity when plotted on a logarithmic scale. The rule was not developed in the context of nonlinear band edges with complex fine structure. Even so, this model has been applied extensively to band edges with such features. Using differential analysis and deconvolution, we develop an analytical technique capable of separating the components responsible for the fine structure observed. This approach reduces challenging signal-artifact convolution problems to a simple linear combination. Furthermore, we develop a model that can be applied to the nonlinear band-edge signals extracted. In this work, we show that in the limiting case of linearity, our model is consistent with and produces the same results as the linear Urbach model. Our model is capable of bypassing “tailing” as an indirect measure of the spread in the local potential and gives a direct measure of this quantity.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗