Search NASA⌕ Search

SEARCH · Search NASA

Results for “discretization error”

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 127 records · Page 7

Co-designing Spectral Transformation Oracles with Hybrid Oscillator-Qubit Quantum Processors: From Algorithms to Compilation

We co-design a family of quantum eigenvalue transformation oracles that can be efficiently implemented on hybrid discrete- or continuous-variable (qubit or qumode) hardware. To illustrate the oracle’s representation-theoretic power and near-term experimental accessibility, we encode a Gaussian imaginary time-evolution spectral filter. As a result, we define a continuous linear combination of unitaries block encoding. Due to the ancillary qumode’s infinite-dimensional nature, continuous-variable qumodes constitute a powerful compilation tool for encoding continuous spectral functions without discretization errors while minimizing resource requirements. We then focus on the ubiquitous task of preparing eigenstates in quantum spin models. For completeness, we provide an end-to-end compilation which expresses high-level oracles in terms of an experimentally realizable instruction set architecture in both 1D and 2D. Finally, we examine the leading-order effects of physical errors and highlight open research directions. Our algorithms scale linearly with the spatial extent of the target system and are applicable to both near-term and large-scale quantum processors.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Continuous-variable quantum computation of the O(3) model in 1+1 dimensions

We formulate the $O(3)$ non-linear sigma model in $1+1$ dimensions as a limit of a three-component scalar field theory restricted to the unit sphere in the large squeezing limit. This allows us to describe the model in terms of the continuous variable (CV) approach to quantum computing. Here we construct the ground state and excited states using the coupled cluster ansatz and find excellent agreement with the exact diagonalization results for a small number of lattice sites. We then present the simulation protocol for the time evolution of the model using CV gates, estimate the discretization error, and present numerical results obtained from a photonic quantum simulator. We expect that the methods developed in this work will be useful for exploring interesting dynamics for a wide class of sigma models and gauge theories, as well as for simulating scattering events on quantum hardware in the coming decade.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Dynamical mean field theory simulations with the adaptive sampling configuration interaction method

In the pursuit of accurate descriptions of strongly correlated quantum many-body systems, dynamical mean field theory (DMFT) has been an invaluable tool for elucidating the spectral properties and quantum phases of both phenomenological models and ab initio descriptions of real materials. Key to the DMFT process is the self-consistent map of the original system into an Anderson impurity model, the ground state of which is computed using an impurity solver. The power of the method is thus limited by the complexity of the impurity model the solver can handle. Simulating realistic systems generally requires many correlated sites. By adapting the recently proposed adaptive sampling configuration interaction (ASCI) method as an impurity solver, we enable much more efficient zero-temperature DMFT simulations. The key feature of the ASCI method is that it selects only the most relevant Hilbert space degrees of freedom to describe the ground state. This reduces the numerical complexity of the calculation, which will allow us to pursue future DMFT simulations with more correlated impurity sites than in previous works. Here, we present the ASCI-DMFT method and example calculations on the one- and two-dimensional Hubbard models that exemplify its efficient convergence and timing properties. We show that the ASCI approach is several orders of magnitude faster than the current best published ground-state DMFT simulations, which allows us to study the bath discretization error in simulations with small clusters, as well as to address cluster sizes beyond the current state of the art. Finally, our approach can also be adapted for other embedding methods such as density matrix embedding theory and self-energy embedding theory.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Calculation of the shear viscosity in SU(3) gluodynamics

We perform a lattice Monte Carlo calculation of the two-point functions of the energy-momentum tensor at finite temperature in the SU(3) gauge theory. Unprecedented precision is obtained thanks to a multilevel algorithm. The lattice operators are renormalized nonperturbatively and the classical discretization errors affecting the correlators are corrected for. A conservative upper bound for the shear viscosity to entropy density ratio is derived, {eta}/s<1.0, and our best estimate is {eta}/s=0.134(33) at T=1.65T{sub c} under the assumption of smoothness of the spectral function in the low-frequency region.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Calibration of moving puncture simulations

We present single and binary black-hole simulations that follow the 'moving-puncture' paradigm of simulating black-hole spacetimes without excision, and use 'moving boxes' mesh refinement. Focusing on binary black-hole configurations where the simulations cover roughly two orbits, we address five major issues determining the quality of our results: numerical discretization error, finite extraction radius of the radiation signal, physical appropriateness of initial data, gauge choice, and computational performance. We also compare results we have obtained with the BAM code described here with the independent LEAN code.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Light-quark masses from unquenched lattice QCD

We calculate the light meson spectrum and the light quark masses by lattice QCD simulation, treating all light quarks dynamically and employing the Iwasaki gluon action and the nonperturbatively O(a)-improved Wilson quark action. The calculations are made at the squared lattice spacings at an equal distance a{sup 2}{approx_equal}0.005, 0.01, and 0.015 fm{sup 2}, and the continuum limit is taken assuming an O(a{sup 2}) discretization error. The light meson spectrum is consistent with experiment. The up, down, and strange quark masses in the MS scheme at 2 GeV are m=(m{sub u}+m{sub d})/2=3.55{sub -0.28}{sup +0.65} MeV and m{sub s}=90.1{sub -6.1}{sup +17.2} MeV where the error includes statistical and all systematic errors added in quadrature. These values contain the previous estimates obtained with the dynamical u and d quarks within the error.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

B{yields}D*l{nu} form factor at zero recoil from three-flavor lattice QCD: A model independent determination of |V{sub cb}|

We present the first lattice QCD calculation of the form factor for B{yields}D*l{nu} with three flavors of sea quarks. We use an improved staggered action for the light valence and sea quarks (the MILC configurations), and the Fermilab action for the heavy quarks. The form factor is computed at zero recoil using a new double ratio method that yields the form factor more directly than the previous Fermilab method. Other improvements over the previous calculation include the use of much lighter light-quark masses, and the use of lattice (staggered) chiral perturbation theory in order to control the light-quark discretization errors and chiral extrapolation. We obtain for the form factor, F{sub B{yields}}{sub D*}(1)=0.921(13)(20), where the first error is statistical and the second is the sum of all systematic errors in quadrature. Applying a 0.7% electromagnetic correction and taking the latest PDG average for F{sub B{yields}}{sub D*}(1)|V{sub cb}| leads to |V{sub cb}|=(38.7{+-}0.9{sub exp}{+-}1.0{sub theo})x10{sup -3}.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Spectrum of the Wilson Dirac operator at finite lattice spacings

We consider the effect of discretization errors on the microscopic spectrum of the Wilson Dirac operator using both chiral perturbation theory and chiral random matrix theory. A graded chiral Lagrangian is used to evaluate the microscopic spectral density of the Hermitian Wilson Dirac operator as well as the distribution of the chirality over the real eigenvalues of the Wilson Dirac operator. It is shown that a chiral random matrix theory for the Wilson Dirac operator reproduces the leading zero-momentum terms of Wilson chiral perturbation theory. All results are obtained for a fixed index of the Wilson Dirac operator. The low-energy constants of Wilson chiral perturbation theory are shown to be constrained by the Hermiticity properties of the Wilson Dirac operator.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Improvement via hypercubic smearing in triplet and sextet QCD

We study nonperturbative improvement in SU(3) lattice gauge theory coupled to fermions in the fundamental and two-index symmetric representations. Our lattice action is defined with hypercubic smeared links incorporated into the Wilson-clover fermion kernel. Using standard Schroedinger-functional techniques, we estimate the clover coefficient c{sub SW} and find that discretization errors are much smaller than in thin-link theories.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Scaling study of the gluon propagator in Coulomb gauge QCD on isotropic and anisotropic lattices

We calculate the transverse and time-time components of the instantaneous gluon propagator in Coulomb gauge QCD by using an SU(3) quenched lattice simulation on isotropic and anisotropic lattices. We find that the gluon propagators suffer from strong discretization effects on the isotropic lattice; on the other hand, those on the anisotropic lattices give a better scaling. Moreover, on these two type of lattices the transverse parts are significantly suppressed in the infrared region and have a turnover at about 500 [MeV]. The high resolution to the temporal direction due to the anisotropy yields small discretization errors for the time-time gluon propagators, which also show an infrared enhancement as expected in the Gribov-Zwanziger confinement scenario.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

SU(2) lattice gluon propagators at finite temperatures in the deep infrared region and Gribov copy effects

We study numerically the SU(2) Landau gauge transverse and longitudinal gluon propagators at nonzero temperatures T both in confinement and deconfinement phases. Special attention is paid to the Gribov copy effects in the IR region. Applying a powerful gauge fixing algorithm, we find that the Gribov copy effects for the transverse propagator D{sub T}(p) are very strong in the infrared, while the longitudinal propagator D{sub L}(p) shows very weak (if any) Gribov copy dependence. The value D{sub T}(0) tends to decrease with growing lattice size; however, D{sub T}(0) is nonzero in the infinite volume limit, in disagreement with the suggestion made in [I. Zahed and D. Zwanziger, Phys. Rev. D 61, 037501 (2000).]. We show that in the infrared region, D{sub T}(p) is not consistent with the pole-type formula not only in the deconfinement phase but also for T<T{sub c}. We introduce a new definition of the magnetic infrared mass scale ('magnetic screening mass') m{sub M}. The electric mass m{sub E} has been determined from the momentum space longitudinal gluon propagator. We study also the (finite) volume and temperature dependence of the propagators as well as discretization errors.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Nucleon transverse momentum-dependent parton distributions in lattice QCD: Renormalization patterns and discretization effects

Lattice QCD calculations of transverse momentum-dependent parton distribution functions (TMDs) in nucleons are presented in this paper, based on the evaluation of nucleon matrix elements of quark bilocal operators with a staple-shaped gauge connection. Both time-reversal odd effects, namely, the generalized Sivers and Boer-Mulders transverse momentum shifts, as well as time-reversal even effects, namely, the generalized transversity and one of the generalized worm-gear shifts, are studied. Results are obtained on two different n f = 2 + 1 flavor ensembles with approximately matching pion masses but very different discretization schemes: domain-wall fermions (DWF) with lattice spacing a = 0.084 fm and pion mass 297 MeV, and Wilson-clover fermions with a = 0.114 fm and pion mass 317 MeV. Comparison of the results on the two ensembles yields insight into the length scales at which lattice discretization errors are small, and into the extent to which the renormalization pattern obeyed by the continuum QCD TMD operator continues to apply in the lattice formulation. For the studied TMD observables, the results are found to be consistent between the two ensembles at sufficiently large separation of the quark fields within the operator, whereas deviations are observed in the local limit and in the case of a straight link gauge connection, which is relevant to the studies of parton distribution functions. Finally and furthermore, the lattice estimates of the generalized Sivers shift obtained here are confronted with, and are seen to tend towards, a phenomenological estimate extracted from experimental data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Physics-Informed Neural Networks for Heat Transfer Problems

Abstract Physics-informed neural networks (PINNs) have gained popularity across different engineering fields due to their effectiveness in solving realistic problems with noisy data and often partially missing physics. In PINNs, automatic differentiation is leveraged to evaluate differential operators without discretization errors, and a multitask learning problem is defined in order to simultaneously fit observed data while respecting the underlying governing laws of physics. Here, we present applications of PINNs to various prototype heat transfer problems, targeting in particular realistic conditions not readily tackled with traditional computational methods. To this end, we first consider forced and mixed convection with unknown thermal boundary conditions on the heated surfaces and aim to obtain the temperature and velocity fields everywhere in the domain, including the boundaries, given some sparse temperature measurements. We also consider the prototype Stefan problem for two-phase flow, aiming to infer the moving interface, the velocity and temperature fields everywhere as well as the different conductivities of a solid and a liquid phase, given a few temperature measurements inside the domain. Finally, we present some realistic industrial applications related to power electronics to highlight the practicality of PINNs as well as the effective use of neural networks in solving general heat transfer problems of industrial complexity. Taken together, the results presented herein demonstrate that PINNs not only can solve ill-posed problems, which are beyond the reach of traditional computational methods, but they can also bridge the gap between computational and experimental heat transfer.

Engineering↗

Laplacian Smoothing Stochastic Gradient Markov Chain Monte Carlo

As an important Markov chain Monte Carlo (MCMC) method, the stochastic gradient Langevin dynamics (SGLD) algorithm has achieved great success in Bayesian learning and posterior sampling. Furthermore, SGLD typically suffers from a slow convergence rate due to its large variance caused by the stochastic gradient. In order to alleviate these drawbacks, we leverage the recently developed Laplacian smoothing technique and propose a Laplacian smoothing stochastic gradient Langevin dynamics (LS-SGLD) algorithm. We prove that for sampling from both log-concave and non-log-concave densities, LS-SGLD achieves strictly smaller discretization error in 2-Wasserstein distance, although its mixing rate can be slightly slower. Experiments on both synthetic and real datasets verify our theoretical results and demonstrate the superior performance of LS-SGLD on different machine learning tasks including posterior sampling, Bayesian logistic regression, and training Bayesian convolutional neural networks.

97 MATHEMATICS AND COMPUTING↗

A Fast Solver for the Fractional Helmholtz Equation

The purpose of this paper is to study a Helmholtz problem with a spectral fractional Laplacian, instead of the standard Laplacian. Recently, it has been established that such a fractional Helmholtz problem better captures the underlying behavior in Geophysical Electromagnetics. In this work, we establish the well-posedness and regularity of this problem. We introduce a hybrid spectral-finite element approach to discretize it and show well-posedness of the discrete system. In addition, we derive a priori discretization error estimates. Finally, we introduce an efficient solver that scales aswell as the best possible solver for the classical integer-order Helmholtz equation. We conclude withseveral illustrative examples that confirm our theoretical findings.

97 MATHEMATICS AND COMPUTING↗

A Probabilistic Scheme for Semilinear Nonlocal Diffusion Equations with Volume Constraints

This work presents a probabilistic scheme for solving semilinear nonlocal diffusion equations with volume constraints and integrable kernels. The nonlocal model of interest is defined by a time-dependent semilinear partial integro-differential equation (PIDE), in which the integro-differential operator consists of both local convection-diffusion and nonlocal diffusion operators. Here, our numerical scheme is based on the direct approximation of the nonlinear Feynman–Kac formula that establishes a link between nonlinear PIDEs and stochastic differential equations. The exploitation of the Feynman–Kac representation avoids solving dense linear systems arising from nonlocal operators. Compared with existing stochastic approaches, our method can achieve first-order convergence after balancing the temporal and spatial discretization errors, which is a significant improvement of existing probabilistic/stochastic methods for nonlocal diffusion problems. Error analysis of our numerical scheme is established. The effectiveness of our approach is shown in two numerical examples. The first example considers a three-dimensional nonlocal diffusion equation to numerically verify the error analysis results. The second example presents a physics problem motivated by the study of heat transport in magnetically confined fusion plasmas.

97 MATHEMATICS AND COMPUTING↗

Omega_h

Omega_h is a C++11 library that implements tetrahedron and triangle mesh adaptativity, with a focus on scalable HPC performance using (optionally) MPI, OpenMP, or CUDA. It is intended to provided adaptive functionality to existing simulation codes. Mesh adaptivity allows one to minimize both discretization error and number of degrees of freedom live during the simulation, as well as enabling moving object and evolving geometry simulations. Omega_h will do this for you in a way that is fast, memory-efficient, and portable across many different architectures, including NVIDIA GPUs.

Ibanez, Daniel [Sandia National Laboratories] (000↗

HPGMG

HPGMG is a scalable reimplementation of miniGMG (see previous software disclosure) and is a compact geometric multigrid (MG benchmark designed to proxy the performance characteristics of the solves found in adaptive mesh refinement multigrid (AMR MG) applications. It solves the equation a*alpha*u - b div beta grad u = f on a large cubical domain using cell-centered values and either homogenous dirichlet or periodic boundaries. a and b are scalar constants, alpha and beta are space varying constants, f is the right hand side, and u is the solution. The cubical domain is divided into cubical subdomains which are distributed across the supercomputer. Whereas miniGMG enforced a static spatial decomposition across all levels in the MG V-cycle, HPGMG is free to redistribute data at each level including the possibility of moving all data to a single process (memory permitting). Thus, where miniGMG's U-cycle demanded a coarse grid solve across the entire supercomputer (a performance bottleneck), HPGMG can avoid this by combining distributed data and continuing the V-cycle. Technically, HPGMG can still implement a U-cycle or a Krylov bottom solve, but this solve can now be done on a global problem of only 2^3-8^3 cells. Additionally, HPGMG supports the MG F-cycle which avoids the iterative process of the v-cycle with a direct approach that can reach the discretization error in one pass. This provides a substantial performance boosts and opens avenues for exploiting systems that include hierarchical memories or heterogeneous processors. Like miniGMG, the righthand side f in HPGMG is generated from a continuous function for u. Thus, one can solve.

Williams, Samuel↗