Search NASA⌕ Search

SEARCH · Search NASA

Results for “mathematical regularization”

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 253 records · Page 14

(U) Correlated Sampling Using Batch Statistics to Reduce the Uncertainty of Combinations of KSEN Outputs with MCNP6

The relative sensitivity of k eff to the densities of nuclides in a material are combined to compute relative sensitivities to other inputs. When computed in a single Monte Carlo run, the nuclide density sensitivities are correlated, and the statistical uncertainties propagated to other inputs will be incorrect unless those correlations are accounted for. Equations are presented to apply correlated sampling using batch statistics for the sum of an arbitrary number of random tallies and the difference of two random tallies when each is multiplied by a different constant. When correlated sampling is used on a recent benchmark evaluation, the correct statistical uncertainties for certain combinations of sensitivities are dramatically smaller than the incorrect uncertainties. A user-controlled, regular output of MCNP6’s KSEN sensitivities that allows batch statistics to be applied to combinations would be extremely valuable.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Data-driven Thermal Model Inference with ARMAX, in Smart Environments, based on Normalized Mutual Information

Understanding the models that characterize the thermal dynamics in a smart building is important for the comfort of its occupants and for its energy optimization. Here, a significant amount of research has attempted to utilize thermodynamics (physical) models for smart building control, but these approaches remain challenging due to the stochastic nature of the intermittent environmental disturbances. This paper presents a novel data-driven approach for indoor thermal model inference, which combines an Autoregressive Moving Average with eXogenous inputs model (ARMAX) with a Normalized Mutual Information scheme (NMI). Based on this information-theoretic method, NMI, causal dependencies between the indoor temperature and exogenous inputs are explicitly obtained as a guideline for the ARMAX model to find the dominating inputs. For validation, we use three datasets based on building energy systems-against which we compare our method to an autoregressive model with exogenous inputs (ARX), a regularized ARMAX model, and state-space models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Algorithms for optimal dyadic decision trees

A new algorithm for constructing optimal dyadic decision trees was recently introduced, analyzed, and shown to be very effective for low dimensional data sets. This paper enhances and extends this algorithm by: introducing an adaptive grid search for the regularization parameter that guarantees optimal solutions for all relevant trees sizes, revising the core tree-building algorithm so that its run time is substantially smaller for most regularization parameter values on the grid, and incorporating new data structures and data pre-processing steps that provide significant run time enhancement in practice.

97 MATHEMATICS AND COMPUTING↗

On the Ill-Posed Nature of Inverse Problems: An Exactly Solvable Paradigm Inverse Neutron Diffusion Problem Illustrating the Solution's Non-Computability - Paper 48

In general, a physical system and/or the result of an indirect experimental measurement are modeled mathematically in terms of: (a) A system of linear and/or nonlinear equations that relate the system's independent variables and parameters to the system's state (i.e., dependent) variables; (b) Probability distributions, moments thereof, inequality and/or equality constraints that define the range of variations of the system's parameters; (c) One or several quantities, customarily referred to as system responses (or objective functions, or indices of performance), which are computed using the mathematical model. The 'direct problem' solves the 'parameter-to-output' mapping that describes the 'cause-to-effect' relationship in the respective physical process. In practice, the model parameters and/or responses are experimentally measured quantities and are therefore subject to uncertainties. Many measurement problems are 'inverse' to the direct problem in that they seek to determine the properties of the medium (i.e., various cross sections), or the properties of the source, and/or the size of the medium on its boundaries from measurements of quantities that depend on the neutron flux (or from direct measurements of the neutron flux). In general, two problems are called inverses of one another if the formulation of each involves all or part of the solution of the other. Some authors further group such inverse problems into 'invasive', when the interior flux is accessible for measurements as opposed to 'non-invasive', in which only fluxes on the boundary of (or exterior to) the medium can be measured. Examples of inverse radiative transfer problems have been reviewed by McCormick, while examples of inverse source problems for neutron transport have been provided by Sanchez and McCormick. More recently, Bledsoe et al., have presented two methods (the Levenberg-Marquard, and 'differential evolution method') for addressing numerically inverse transport problems. However, although these works (as well as other works cited therein) mentioned the detrimental effects of 'small perturbations in the input data', none of the works published thus far on inverse problems for particle transport or diffusion actually showed explicitly the effects that uncertainties in the measured responses (e.g., detector measurements) would actually have on the 'inverse' deduction (from such measurements) of the underlying medium's properties, sources, etc. This work highlights explicitly the nature of the difficulties encountered in so-called 'inverse problems' by presenting a paradigm inverse neutron diffusion problem admitting an exact solution, in which the 'inverse problem' is to determine the medium's diffusion coefficient from flux measurements. It will be shown explicitly (analytically) that this is not possible to achieve in a unique manner, even in the absence of measurement uncertainties, unless additional error-free information is provided about the quantity to be determined (in this case, the medium's diffusion coefficient). It if further shown that, even if this additional error-free information about the diffusion coefficient to be determined is available, the unavoidable experimental measurements errors, albeit it vanishingly small, in the measured flux render the 'exact solution' useless for computational purposes, since the exact solution turns out to be a divergent series. As discussed in this paper's ENDNOTES section, the paradigm problem presented in this work also underscores the need for 'regularization', which is a qualifier indicating that the inverse problem is to be solved in some approximate manner. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗

The Small-Mass Limit and White-Noise Limit of an Infinite Dimensional Generalized Langevin Equation

We study asymptotic properties of the Generalized Langevin Equation (GLE) in the presence of a wide class of external potential wells with a power-law decay memory kernel. When the memory can be expressed as a sum of exponentials, a class of Markovian systems in infinite-dimensional spaces is used to represent the GLE. The solutions are shown to converge in probability in the small-mass limit and the white-noise limit to appropriate systems under minimal assumptions, of which no global Lipschitz condition is required on the potentials. With further assumptions about space regularity and potentials, we obtain L{sup 1} convergence in the white-noise limit.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Regularity of Solutions to Quasilinear Parabolic Systems with Time-Nonsmooth Principal Matrix and the Neumann Boundary Condition

We consider a quasilinear parabolic system of equations with nondiagonal principal matrix in a model parabolic cylinder with the Neumann condition on the plane part Γ of the lateral surface of the cylinder. We prove the partial regularity (the Hölder continuity) of the weak solution in a neighborhood of Γ by the method of A(t)-caloric approximations adapted to the problem with the Neumann boundary condition.

97 MATHEMATICS AND COMPUTING↗

Sparse approximation of fitting surface by elastic net

The goal of this paper is to develop a computational model and corresponding efficient algorithm for obtaining a sparse representation of the fitting surface to the given scattered data. The basic idea of the model is to utilize the principal shift invariant space and the balanced l{sub 1},l{sub 2} norm minimization (named elastic net). The elastic net can be solved efficiently by an adapted split Bregman iteration algorithm. Numerical experiments indicate that by choosing appropriate regularization parameters, the model can efficiently provide an acceptable compromise between the minimization of the data mismatch term and the sparsity of the solution.

97 MATHEMATICS AND COMPUTING↗

Extrinsic geometric properties of the Rozendorn surface, an isometric immersion of the Lobachevskii plane in E{sup 5}

The lengths of the normal curvature vectors on the Rozendorn surface F{sup 2} are shown to be uniformly bounded above on the whole of the surface. A regular three-dimensional submanifold F{sup 3}, F{sup 2} subset of F{sup 3} subset of E{sup 5}, is constructed in the form of a regular leaf whose sectional curvatures in the two-dimensional directions tangent to F{sup 2} are strictly negative and bounded away from zero. Bibliography: 9 titles.

97 MATHEMATICS AND COMPUTING↗

Effects of cross correlations between inhomogeneities of the parameters of an isotropic medium on the spectrum and damping of elastic waves

The dispersion and damping laws have been investigated for elastic waves in an isotropic medium with one- and three-dimensional inhomogeneities of the density p(x) of the material and the elastic force constants {mu}(x) and {lambda}(x) with allowance for the cross correlations between these inhomogeneities. It has been demonstrated that the positive cross correlations between {mu}(x) and {lambda}(x), as well as the negative cross correlations between p(x) and {mu}(x) or p(x) and {lambda}(x), lead to an enhancement of the modification of the dispersion law and an increase in the damping of waves. The positive cross correlations between p(x) and {mu}(x) or p(x) and {lambda}(x), as well as the negative cross correlations between {mu}(x) and {lambda}(x), result in the opposite effects: a weakening of the modification of the dispersion law and a decrease in the damping. An analysis of the results obtained in this paper and in our recent work [15] has made it possible to formulate the general regularity of the effects of cross correlations, irrespective of the physical nature of the waves: the effects of cross correlations between inhomogeneities of any two parameters of the material on the wave spectrum depend on whether both parameters related by the cross correlations belong to the same part of the Hamiltonian (i.e., if they both belong to either the kinetic part or the potential part of the Hamiltonian) or they belong to different parts of the Hamiltonian. The positive cross correlations lead to a greater modification of the dispersion law and to an increase in the damping of waves in the former case and to a decrease in these characteristics in the latter case. Correspondingly, the negative cross correlations in each of these cases result in the opposite effects. This regularity has been explained qualitatively.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

ECP Software Technology Capability Assessment Report

The Exascale Computing Project (ECP) Software Technology (ST) Focus Area is responsible for developing critical software capabilities that will enable successful execution of ECP applications, and for providing key components of a productive and sustainable Exascale computing ecosystem that will position the US Department of Energy (DOE) and the broader high performance (HPC) community with a rm foundation for future extreme-scale computing capabilities. This ECP ST Capability Assessment Report (CAR) provides an overview and assessment of current ECP ST capabilities and activities, giving stakeholders and the broader HPC community information that can be used to assess ECP ST progress and plan their own efforts accordingly. ECP ST leaders commit to updating this document on regular basis (targeting approximately every six months). Highlights from the report are presented here.

97 MATHEMATICS AND COMPUTING↗

Preinflationary and inflationary fast-roll eras and their signatures in the low CMB multipoles

We study the entire coupled evolution of the inflaton {phi}(t) and the scale factor a(t) for general initial conditions {phi}(t{sub 0}) and d{phi}(t{sub 0})/dt at a given initial time t{sub 0}. The generic early Universe evolution has three stages: decelerated fast roll followed by inflationary fast roll and then inflationary slow roll (an attractor always reached for generic initial conditions). This evolution is valid for all regular inflaton potentials v({phi}). In addition, we find a special (extreme) slow-roll solution starting at t=-{infinity} in which the fast-roll stages are absent. At some time t=t{sub *}, the evolution backwards in time from t{sub 0} reaches generically a mathematical singularity where a(t) vanishes and the Hubble parameter becomes singular. We determine the general behavior near the singularity. The classical homogeneous inflaton description turns to be valid for t-t{sub *}>10t{sub Planck} well before the beginning of inflation, quantum loop effects are negligible there. The singularity is never reached in the validity region of the classical treatment and therefore it is not a real physical phenomenon here. Fast-roll and slow-roll regimes are analyzed in detail including the equation of state evolution, both analytically and numerically. The characteristic time scale of the fast-roll era turns to be t{sub 1}=(1/m){radical}(V(0)/[3M{sup 4}]){approx}10{sup 4}t{sub Planck}, where V is the double-well inflaton potential, m is the inflaton mass, and M the energy scale of inflation. The whole evolution of the fluctuations along the decelerated and inflationary fast-roll and slow-roll eras is computed. The Bunch-Davies initial conditions are generalized for the present case in which the potential felt by the fluctuations can never be neglected. The fluctuations feel a singular attractive potential near the t=t{sub *} singularity (as in the case of a particle in a central singular potential) with exactly the critical strength (-1/4) allowing the fall to the center. Precisely, the fluctuations exhibit logarithmic behavior describing the fall to t=t{sub *}. The power spectrum gets dynamically modified by the effect of the fast-roll eras and the choice of Bunch-Davies initial conditions at a finite time through the transfer function D(k) of initial conditions. The power spectrum vanishes at k=0.D(k) presents a first peak for k{approx}2/{eta}{sub 0} ({eta}{sub 0} being the conformal initial time), then oscillates with decreasing amplitude and vanishes asymptotically for k{yields}{infinity}. The transfer function D(k) affects the low cosmic microwave background multipoles C{sub l}: the change {Delta}C{sub l}/C{sub l} for 1{<=}l{<=}5 is computed as a function of the starting instant of the fluctuations t{sub 0}. Cosmic microwave background quadrupole observations indicate large suppressions, which are well reproduced for the range t{sub 0}-t{sub *} > or approx. 0.05/m{approx_equal}10 100t{sub Planck}.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Shape and Topology Optimization in Stokes Flow with a Phase Field Approach

In this paper we introduce a new formulation for shape optimization problems in fluids in a diffuse interface setting that can in particular handle topological changes. By adding the Ginzburg–Landau energy as a regularization to the objective functional and relaxing the non-permeability outside the fluid region by introducing a porous medium approach we hence obtain a phase field problem where the existence of a minimizer can be guaranteed. This problem is additionally related to a sharp interface problem, where the permeability of the non-fluid region is zero. In both the sharp and the diffuse interface setting we can derive necessary optimality conditions using only the natural regularity of the minimizers. We also pass to the limit in the first order conditions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Time-dependent Pauli equation in the presence of the Aharonov-Bohm effect

We use the Lewis-Riesenfeld theory to determine the exact form of the wavefunctions of a two-dimensional Pauli equation of a charged spin 1/2 particle with time-dependent mass and frequency in the presence of the Aharonov-Bohm effect and a two-dimensional time-dependent harmonic oscillator. We find that the irregular solution at the origin as well as the regular one contributes to the phase of the wavefunction.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Wassersplines for Neural Vector Field-Controlled Animation

Much of computer-generated animation is created by manipulating meshes with rigs. While this approach works well for animating articulated objects like animals, it has limited flexibility for animating less structured free-form objects. Here we introduce Wassersplines, a novel trajectory inference method for animating unstructured densities based on recent advances in continuous normalizing flows and optimal transport. The key idea is to train a neurally-parameterized velocity field that represents the motion between keyframes. Trajectories are then computed by advecting keyframes through the velocity field. We solve an additional Wasserstein barycenter interpolation problem to guarantee strict adherence to keyframes. Our tool can stylize trajectories through a variety of PDE-based regularizers to create different visual effects. We demonstrate our tool on various keyframe interpolation problems to produce temporally-coherent animations without meshing or rigging.

97 MATHEMATICS AND COMPUTING↗

KKT Preconditioners for PDE-Constrained Optimization with the Helmholtz Equation

This paper considers preconditioners for the linear systems that arise from optimal control and inverse problems involving the Helmholtz equation. Specifically, we explore an all-at-once approach. The main contribution centers on the analysis of two block preconditioners. Variations of these preconditioners have been proposed and analyzed in prior works for optimal control problems where the underlying partial differential equation is a Laplace-like operator. In this paper, we extend some of the prior convergence results to Helmholtz-based optimization applications. Our analysis examines situations where control variables and observations are restricted to subregions of the computational domain. We prove that solver convergence rates do not deteriorate as the mesh is refined or as the wavenumber increases. More specifically, for one of the preconditioners we prove accelerated convergence as the wavenumber increases. Additionally, in situations where the control and observation subregions are disjoint, we observe that solver convergence rates have a weak dependence on the regularization parameter. We give a partial analysis of this behavior. We illustrate the performance of the preconditioners on control problems motivated by acoustic testing.

97 MATHEMATICS AND COMPUTING↗

Analysis of a Stabilized CNLF Method with Fast Slow Wave Splittings for Flow Problems

In this work, we study Crank-Nicolson leap-frog (CNLF) methods with fast-slow wave splittings for Navier-Stokes equations (NSE) with a rotation/Coriolis force term, which is a simplification of geophysical flows. We propose a new stabilized CNLF method where the added stabilization completely removes the method's CFL time step condition. A comprehensive stability and error analysis is given. We also prove that for Oseen equations with the rotation term, the unstable mode (for which u(n+1) + u(n-1) equivalent to 0) of CNLF is asymptotically stable. Numerical results are provided to verify the stability and the convergence of the methods.

97 MATHEMATICS AND COMPUTING↗

Cryo-STEM Tomography with Inpainting

Traditionally, microscopists have worked with the Nyquist-Shannon theory of sampling, which states that to be able to reconstruct the image fully it needs to be sampled at a rate of at least twice the highest spatial frequency in the image. This sampling rate assumes that the image is sampled at regular intervals and that every pixel contains information that is crucial for the image (it even assumes that noise is important). Images in general, and especially low dose S/TEM images, contain significantly less information than can be encoded by a grid of pixels (which is why image compression works). Mathematically speaking, the image data has a low dimensional or sparse representation. Through the application of compressive sensing methods [1,2,3] this representation can be found using pre-designed measurements that are usually random for implementation simplicity. These measurements and the compressive sensing reconstruction algorithms have the added benefit of reducing noise. This reconstruction approach can be extended into higher dimensions, whereby the random sampling in each 2-D image can be extended into: a sequence of tomographic projections (i.e. tilt images); a sequence of video frames (i.e. incorporating temporal resolution and dynamics); spectral resolution (i.e. energy filtering an image to see the distribution of elements); and ptychography (i.e. sampling a full diffraction image at each location in a 2-D grid across the sample). This approach has been employed experimentally for materials science samples requiring low-dose imaging [2], and can be readily applied to biological samples. Figure 1 shows the resolution possible in a complex biological system, mouse pancreatic islet beta cells [4], when tomogram slices are reconstructed using subsampling. Reducing the number of pixels (1/6 pix and 1/3*1/3) shows minimal degradation compared to the reconstructions using all pixels (all data and 1/3 tilt). Although subsampling 1/6 of the tilts (1/6 of overall dose) degrades the reconstruction to the point that the cellular structures cannot be identified. Using 1/3 of both the pixels and the tilts provides a high quality image at 1/9 the overall dose even for this most basic and rapid demonstration of the CS methods. Figure 2 demonstrates the theoretical tomogram reconstruction quality (vertical axis) as undersampling (horizontal axis) is increased; we examined subsampling pixels and tilt-angles individually and a combined approach in which both pixels and tilts are sub-sampled. Note that subsampling pixels maintains high quality reconstructions (solid lines). Using the inpainting algorithm to obtain tomograms can automatically reduce the dose applied to the system by an order of magnitude. Perhaps the best way to understand the impact is to consider that by using inpainting (and with minimal hardware changes), a sample that can normally withstand a dose of ~10 e/Å2 can potentially be imaged with an “equivalent quality” to a dose level of 103 e/Å2. To put this in perspective, this is approaching the dose level used for the most advanced images, in terms of spatial resolution, for inorganic systems. While there are issues for biological specimens beyond dose (structural complexity being the most important one), this sampling approach allows the methods that are traditionally used for materials science to be applied to biological systems [5]. References: [1] A Stevens, H Yang, L Carin et al. Microscopy 63(1), (2014), pp. 41. [2] L Kovarik, A Stevens, A Liyu et al. Appl. Phys. Lett. 109, 164102 (2016) [3] A Stevens, L Kovarik, P Abellan et al. Adv. Structural and Chemical Imaging 1(10), (2015), pp. 1. [4] MD Guay, W Czaja, MA Aronova et al. Scientific Reports 6, 27614 (2016) [5] Supported by the Chemical Imaging, Signature Discovery, and Analytics in Motion Initiatives at PNNL. PNNL is operated by Battelle Memorial Inst. for the US DOE; contract DE-AC05-76RL01830.

Stevens, Andrew↗

On the Stochastic Wave Equation with Nonlinear Damping

We discuss an initial boundary value problem for the stochastic wave equation with nonlinear damping. We establish the existence and uniqueness of a solution. Our method for the existence of pathwise solutions consists of regularization of the equation and data, the Galerkin approximation and an elementary measure-theoretic argument. We also prove the existence of an invariant measure when the equation has pure nonlinear damping.

97 MATHEMATICS AND COMPUTING↗