Search NASA⌕ Search

SEARCH · Search NASA

Results for “numerical diffusion”

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 1,297 records · Page 72

Exponential Runge-Kutta Parareal for non-diffusive equations

Parareal is a well-known parallel-in-time algorithm that combines a coarse and fine propagator within a parallel iteration. It allows for large-scale parallelism that leads to significantly reduced computational time compared to serial time-stepping methods. However, like many parallel-in-time methods it can fail to converge when applied to non-diffusive equations such as hyperbolic systems or dispersive nonlinear wave equations. Here, this paper explores the use of exponential integrators within the Parareal iteration. Exponential integrators are particularly interesting candidates for Parareal because of their ability to resolve fast-moving waves, even at the large stepsizes used by coarse propagators. This work begins with an introduction to exponential Parareal integrators followed by several motivating numerical experiments involving the nonlinear Schrödinger equation. These experiments are then analyzed using linear analysis that approximates the stability and convergence properties of the exponential Parareal iteration on nonlinear problems. The paper concludes with two additional numerical experiments involving the dispersive Kadomtsev-Petviashvili equation and the hyperbolic Vlasov-Poisson equation. These experiments demonstrate that exponential Parareal methods offer improved time-to-solution compared to serial exponential integrators when solving certain non-diffusive equations.

97 MATHEMATICS AND COMPUTING↗

A meshless stochastic method for Poisson–Nernst–Planck equations

A plethora of biological, physical, and chemical phenomena involve transport of charged particles (ions). Its continuum-scale description relies on the Poisson–Nernst–Planck (PNP) system, which encapsulates the conservation of mass and charge. The numerical solution of these coupled partial differential equations is challenging and suffers from both the curse of dimensionality and difficulty in efficiently parallelizing. We present a novel particle-based framework to solve the full PNP system by simulating a drift–diffusion process with time- and space-varying drift. We leverage Green’s functions, kernel-independent fast multipole methods, and kernel density estimation to solve the PNP system in a meshless manner, capable of handling discontinuous initial states. The method is embarrassingly parallel, and the computational cost scales linearly with the number of particles and dimension. We use a series of numerical experiments to demonstrate both the method’s convergence with respect to the number of particles and computational cost vis-à-vis a traditional partial differential equation solver.

Chemistry↗

A study of dust grain screening in a weakly ionized plasma based on the numerical solution of the Vlasov-Bhatnagar-Gross-Krook kinetic equations

Screening of a dust grain in a weakly ionized plasma is studied for a wide range of collisional regimes. The problem is considered on the basis of the Vlasov-Bhatnagar-Gross-Krook equations for plasma particles. The equations are solved numerically on parallel processors by means of a high-order finite-volume method. The computations are carried out for different pressures of plasma background and different grain sizes. The values of the total grain charge, distributions of the electric potential, and basic macroscopic parameters of plasma particles (concentration, temperature, and velocity) are obtained. For two limiting cases of the collisionless and strongly collisional background, the obtained results are found to be in good agreement with those computed using the commonly known models (nonlinear collisionless models of Bystrenko et al.[Phys. Lett. A 299, 383 (2002)] and Ratynskaia et al.[Phys. Plasmas 13, 104508 (2006)] and drift-diffusion approach). For the transitional collisional regime, it is shown that the electric potential also has the Coulomb-type asymptotical behavior related to the presence of charging currents and existence of some residual (unscreened) grain charge. Both total and residual grain charges are found to increase as the collisional regime tends to the continuum limit. In addition, the influence of collisions on the ion distribution function near the grain surface is demonstrated.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Kinetic Monte Carlo simulation of dopant-defect systems under submicrosecond laser thermal processes

An innovative Kinetic Monte Carlo (KMC) code has been developed, which rules the post-implant kinetics of the defects system in the extremely far-from-the equilibrium conditions caused by the laser irradiation close to the liquid-solid interface. It considers defect diffusion, annihilation and clustering. The code properly implements, consistently to the stochastic formalism, the fast varying local event rates related to the thermal field T(r,t) evolution. This feature of our numerical method represents an important advancement with respect to current state of the art KMC codes. The reduction of the implantation damage and its reorganization in defect aggregates are studied as a function of the process conditions. Phosphorus activation efficiency, experimentally determined in similar conditions, has been related to the emerging damage scenario.

36 MATERIALS SCIENCE↗

Where did the tumor start? An inverse solver with sparse localization for tumor growth models

In this work, we present a numerical scheme for solving an inverse problem for parameter estimation in tumor growth models for glioblastomas, a form of aggressive primary brain tumor. The growth model is a reaction–diffusion partial differential equation (PDE) for the tumor concentration. We use a PDE-constrained optimization formulation for the inverse problem. The unknown parameters are the reaction coefficient (proliferation), the diffusion coefficient (infiltration), and the initial condition field for the tumor PDE. Segmentation of magnetic resonance imaging (MRI) scans drive the inverse problem where segmented tumor regions serve as partial observations of the tumor concentration. Like most cases in clinical practice, we use data from a single time snapshot. Moreover, the precise time relative to the initiation of the tumor is unknown, which poses an additional difficulty for inversion. We perform a frozen-coefficient spectral analysis and show that the inverse problem is severely ill-posed. We introduce a biophysically motivated regularization on the structure and magnitude of the tumor initial condition. In particular, we assume that the tumor starts at a few locations (enforced with a sparsity constraint on the initial condition of the tumor) and that the initial condition magnitude in the maximum norm is equal to one. We solve the resulting optimization problem using an inexact quasi-Newton method combined with a compressive sampling algorithm for the sparsity constraint. Our implementation uses PETSc and AccFFT libraries. We conduct numerical experiments on synthetic and clinical images to highlight the improved performance of our solver over a previously existing solver that uses standard two-norm regularization for the calibration parameters. The existing solver is unable to localize the initial condition. Our new solver can localize the initial condition and recover infiltration and proliferation. In clinical datasets (for which the ground truth is unknown), our solver results in qualitatively different solutions compared to the two-norm regularized solver.

97 MATHEMATICS AND COMPUTING↗

The hysteresis response of soil CO 2 concentration and soil respiration to soil temperature

Diurnal hysteresis between soil temperature (T s ) and both CO 2 concentration ([CO 2 ]) and soil respiration rate (R s ) were reported across different field experiments. However, the causes of these hysteresis patterns remain a subject of debate, with biotic and abiotic factors both invoked as explanations. Here, to address these issues, a CO 2 gas transport model is developed by combining a layer-wise mass conservation equation for subsurface gas phase CO 2 , Fickian diffusion for gas transfer, and a CO 2 source term that depends on soil temperature, moisture, and photosynthetic rate. Using this model, a hierarchy of numerical experiments were employed to disentangle the causes of the hysteretic [CO 2 ]-T s and CO 2 flux T s (i.e., F-T s ) relations. Model results show that gas transport alone can introduce both [CO 2 ]-T s and F-T s hystereses and also confirm prior findings that heat flow in soils lead to [CO 2 ] and F being out of phase with T s , thereby providing another reason for the occurrence of both hystereses. The area (A hys ) of the [CO 2 ]-T s hysteresis near the surface increases, while the A hys of the Rs-Ts hysteresis decreases as soils become wetter. Moreover, a time-lagged carbon input from photosynthesis deformed the [CO 2 ]-T s and R s -T s patterns, causing a change in the loop direction from counterclockwise to clockwise with decreasing time lag. An asymmetric 8-shaped pattern emerged as the transition state between the two loop directions. Lastly, tracing the pattern and direction of the hysteretic [CO 2 ]-T s and R s -T s relations can provide new ways to fingerprint the effects of photosynthesis stimulation on soil microbial activity and detect time lags between rhizospheric respiration and photosynthesis.

58 GEOSCIENCES↗

An Efficient Numerical Algorithm for Solving Data Driven Feedback Control Problems

The goal of this paper is to solve a class of stochastic optimal control problems numerically, in which the state process is governed by an Itô type stochastic differential equation with control process entering both in the drift and the diffusion, and is observed partially. The optimal control of feedback form is determined based on the available observational data. In this work, we call this type of control problems the data driven feedback control. The computational framework that we introduce to solve such type of problems aims to find the best estimate for the optimal control as a conditional expectation given the observational information. To make our method feasible in providing timely feedback to the controlled system from data, we develop an efficient stochastic optimization algorithm to implement our computational framework.

97 MATHEMATICS AND COMPUTING↗

Operator inference for non-intrusive model reduction of systems with non-polynomial nonlinear terms

Here in this work we present a non-intrusive model reduction method to learn low-dimensional models of dynamical systems with non-polynomial nonlinear terms that are spatially local and that are given in analytic form. In contrast to state-of-the-art model reduction methods that are intrusive and thus require full knowledge of the governing equations and the operators of a full model of the discretized dynamical system, the proposed approach requires only the non-polynomial terms in analytic form and learns the rest of the dynamics from snapshots computed with a potentially black-box full-model solver. The proposed method learns operators for the linear and polynomially nonlinear dynamics via a least-squares problem, where the given non-polynomial terms are incorporated on the right-hand side. The least-squares problem is linear and thus can be solved efficiently in practice. The proposed method is demonstrated on three problems governed by partial differential equations, namely the diffusion–reaction Chafee–Infante model, a tubular reactor model for reactive flows, and a batch-chromatography model that describes a chemical separation process. The numerical results provide evidence that the proposed approach learns reduced models that achieve comparable accuracy as models constructed with state-of-the-art intrusive model reduction methods that require full knowledge of the governing equations.

42 ENGINEERING↗

The interaction of high-speed turbulence with flames: Global properties and internal flame structure

We study the dynamics and properties of a turbulent flame, formed in the presence of subsonic, high-speed, homogeneous, isotropic Kolmogorov-type turbulence in an unconfined system. Direct numerical simulations are performed with Athena-RFX, a massively parallel, fully compressible, high-order, dimensionally unsplit, reactive flow code. A simplified reaction-diffusion model represents a stoichiometric H{sub 2}-air mixture. The system being modeled represents turbulent combustion with the Damkoehler number Da=0.05 and with the turbulent velocity at the energy injection scale 30 times larger than the laminar flame speed. The simulations show that flame interaction with high-speed turbulence forms a steadily propagating turbulent flame with a flame brush width approximately twice the energy injection scale and a speed four times the laminar flame speed. A method for reconstructing the internal flame structure is described and used to show that the turbulent flame consists of tightly folded flamelets. The reaction zone structure of these is virtually identical to that of the planar laminar flame, while the preheat zone is broadened by approximately a factor of two. Consequently, the system evolution represents turbulent combustion in the thin reaction zone regime. The turbulent cascade fails to penetrate the internal flame structure, and thus the action of small-scale turbulence is suppressed throughout most of the flame. Finally, our results suggest that for stoichiometric H{sub 2}-air mixtures, any substantial flame broadening by the action of turbulence cannot be expected in all subsonic regimes. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

PINN surrogate of Li-ion battery models for parameter inference, Part II: Regularization and application of the pseudo-2D model

Bayesian parameter inference is useful to improve Li-ion battery diagnostics and can help formulate battery aging models. However, it is computationally intensive and cannot be easily repeated for multiple cycles, multiple operating conditions, or multiple replicate cells. To reduce the computational cost of Bayesian calibration, numerical solvers for physics-based models can be replaced with faster surrogates. A physics-informed neural network (PINN) is developed as a surrogate for the pseudo-2D (P2D) battery model calibration. For the P2D surrogate, additional training regularization was needed as compared to the PINN single-particle model (SPM) developed in Part I. Both the PINN SPM and P2D surrogate models are exercised for parameter inference and compared to data obtained from a direct numerical solution of the governing equations. A parameter inference study highlights the ability to use these PINNs to calibrate scaling parameters for the cathode Li diffusion and the anode exchange current density. By realizing computational speed-ups of ~2250x for the P2D model, as compared to using standard integrating methods, the PINN surrogates enable rapid state-of-health diagnostics. Finally, in the low-data availability scenario, the testing error was estimated to ~2 mV for the SPM surrogate and ~10 mV for the P2D surrogate which could be mitigated with additional data.

25 ENERGY STORAGE↗

Generalized fluid theory including non-Maxwellian kinetic effects

The results obtained by the plasma physics community for the validation and the prediction of turbulence and transport in magnetized plasmas come mainly from the use of very central processing unit (CPU)-consuming particle-in-cell or (gyro)kinetic codes which naturally include non-Maxwellian kinetic effects. To date, fluid codes are not considered to be relevant for the description of these kinetic effects. Here, after revisiting the limitations of the current fluid theory developed in the 19th century, we generalize the fluid theory including kinetic effects such as non-Maxwellian super-thermal tails with as few fluid equations as possible. The collisionless and collisional fluid closures from the nonlinear Landau Fokker–Planck collision operator are shown for an arbitrary collisionality. Indeed, the first fluid models associated with two examples of collisionless fluid closures are obtained by assuming an analytic non-Maxwellian distribution function. One of the main differences with the literature is our analytic representation of the distribution function in the velocity phase space with as few hidden variables as possible thanks to the use of non-orthogonal basis sets. These new non-Maxwellian fluid equations could initiate the next generation of fluid codes including kinetic effects and can be expanded to other scientific disciplines such as astrophysics, condensed matter or hydrodynamics. As a validation test, we perform a numerical simulation based on a minimal reduced INMDF fluid model. The result of this test is the discovery of the origin of particle and heat diffusion. The diffusion is due to the competition between a growing INMDF on short time scales due to spatial gradients and the thermalization on longer time scales. Here, the results shown here could provide the insights to break some of the unsolved puzzles of turbulence.

42 ENGINEERING↗

Linearized model Fokker-Planck collision operators for gyrokinetic simulations. I. Theory

A new analytically and numerically manageable model collision operator is developed specifically for turbulence simulations. The like-particle collision operator includes both pitch-angle scattering and energy diffusion and satisfies the physical constraints required for collision operators: it conserves particles, momentum, and energy, obeys Boltzmann's H-theorem (collisions cannot decrease entropy), vanishes on a Maxwellian, and efficiently dissipates small-scale structure in the velocity space. The process of transforming this collision operator into the gyroaveraged form for use in gyrokinetic simulations is detailed. The gyroaveraged model operator is shown to have more suitable behavior at small scales in phase space than previously suggested models. Model operators for electron-ion and ion-electron collisions are also presented.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Self-focusing of a Gaussian electromagnetic beam in a complex plasma

This paper analyzes the propagation of a Gaussian electromagnetic beam in a complex plasma; the paraxial approach has been invoked for the study of the propagation characteristics. The analytical model emphasizes the open nature of complex plasmas and incorporates the charge, number density, and energy balance of the plasma constituents along with the plasma neutrality. Diffusion due to the density and temperature gradients, thermal conduction, Ohmic heating, and energy exchange in collisions/accretion have also been taken into account. For a numerical appreciation of the results, the critical curves for the propagation of the beam in dark plasmas have been discussed. The dependence of the beam width parameter on distance of propagation has been evaluated for three typical cases viz., of steady divergence, oscillatory divergence, and self-focusing. The equality of electron and ion density, an assumption inherent in the earlier analysis, has been discarded on account of the presence of charged dust particles.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Geometrical effects on the electron residence time in semiconductor nano-particles

We have used random walk (RW) numerical simulations to investigate the influence of the geometry on the statistics of the electron residence time τ{sub r} in a trap-limited diffusion process through semiconductor nano-particles. This is an important parameter in coarse-grained modeling of charge carrier transport in nano-structured semiconductor films. The traps have been distributed randomly on the surface (r{sup 2} model) or through the whole particle (r{sup 3} model) with a specified density. The trap energies have been taken from an exponential distribution and the traps release time is assumed to be a stochastic variable. We have carried out (RW) simulations to study the effect of coordination number, the spatial arrangement of the neighbors and the size of nano-particles on the statistics of τ{sub r}. It has been observed that by increasing the coordination number n, the average value of electron residence time, τ{sup ¯}{sub r} rapidly decreases to an asymptotic value. For a fixed coordination number n, the electron's mean residence time does not depend on the neighbors' spatial arrangement. In other words, τ{sup ¯}{sub r} is a porosity-dependence, local parameter which generally varies remarkably from site to site, unless we are dealing with highly ordered structures. We have also examined the effect of nano-particle size d on the statistical behavior of τ{sup ¯}{sub r}. Our simulations indicate that for volume distribution of traps, τ{sup ¯}{sub r} scales as d{sup 2}. For a surface distribution of traps τ{sup ¯}{sub r} increases almost linearly with d. This leads to the prediction of a linear dependence of the diffusion coefficient D on the particle size d in ordered structures or random structures above the critical concentration which is in accordance with experimental observations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Large eddy simulation of a supersonic lifted hydrogen flame: Impacts of Lewis, turbulent Schmidt and Prandtl numbers

Parametric large eddy simulations (LES) of a supersonic lifted hydrogen flame are reported. The emphases are on two aspects: impacts of (1) Lewis number (Lei of the ith species) and (2) turbulent Schmidt and Prandtl numbers (⁠Sc t and Pr t ⁠) on supersonic turbulent flame and flow structures. Five cases are considered: species-specific Le i , Sc t = Sr t = 1.0 (C 0 ); unity Le i , Sc t = Pr t = 1.0 (C 1 ); species-specific Sc t = 0.5, Pr t = 1.0 (C 2 ); species-specific Le i , Sc t = 1.0, Pr t = 0.5 (C 3 ); and species-specific Le i , Sc t = Pr t = 0.5 (C 4 ). Numerical results of instantaneous and/or time-averaged species mole fractions, mixture fraction, heat release rate, flame base location, and mixed modes of premixed and diffusion combustion are compared between cases C 0 and C 1 . Differences in auto-ignition locations and strengths and flame structures and stabilization specify the impacts of Lewis number. They are triggered by different predictions of species mass and thermal diffusions at fuel-coflow and/or coflow-ambient air mixing layers. These differences are rationalized by a scale analysis of mass/thermal diffusion and convection for case C 0 , which suggests the relatively low but non-negligible former against the latter. Cases C 0 and C 2 –C 4 barely see differences in terms of instantaneous and/or time-averaged temperature, velocity, and mixed combustion modes except for further downstream areas where combustion occurs. Both Sc t and Pr t impose less significant influences than Lewis number, as sub-grid scale mass/thermal diffusion is subordinate to its resolved counterpart according to their scale analysis for case C 4 .

42 ENGINEERING↗

NEAR-INFRARED SPECTROSCOPY OF INFRARED-EXCESS STELLAR OBJECTS IN THE YOUNG SUPERNOVA REMNANT G54.1+0.3

We present the results of broadband near-infrared spectroscopic observations of the recently discovered mysterious stellar objects in the young supernova remnant G54.1+0.3. These objects, which show significant mid-infrared-excess emission, are embedded in a diffuse loop structure of {approx}1' in radius. Their near-infrared spectra reveal characteristics of late O- or early B-type stars with numerous H and He I absorption lines, and we classify their spectral types to be between O9 and B2 based on an empirical relation derived here between the equivalent widths of the H lines and stellar photospheric temperatures. The spectral types, combined with the results of spectral energy distribution fits, constrain the distance to the objects to be 6.0 {+-} 0.4 kpc. The photometric spectral types of the objects are consistent with those from the spectroscopic analyses, and the extinction distributions indicate a local enhancement of matter in the western part of the loop. If these objects originate via triggered formation by the progenitor star of G54.1+0.3, then their formations likely began during the later evolutionary stages of the progenitor, although a rather earlier formation may still be possible. If the objects and the progenitor belong to the same cluster of stars, then our results constrain the progenitor mass of G54.1+0.3 to be between 18 and {approx}35 M{sub Sun} and suggest that G54.1+0.3 was either a Type IIP supernova or, with a relatively lower possibility, Type Ib/c from a binary system.

79 ASTRONOMY AND ASTROPHYSICS↗

Grassy ELM regime at low pedestal collisionality in high-power tokamak plasma

Spontaneous mitigation of giant ELMs and appearance of grassy ELMs have been observed repeatedly at low pedestal collisionality (${\nu }_{\mathrm{e},\mathrm{p}\mathrm{e}\mathrm{d}}^{{\ast}}\sim 0.15$) in the high-power (P inj > 13 MW) hybrid scenario in DIII-D. Higher β p and higher q 95 appear to be beneficial to achieving the grassy ELM regime. The grassy ELM H-mode plasma shows high energy confinement performance (H 98y2 up to over 1.5) under the conditions of high neutral beam torque and high core rotation. The pedestal width appears to exceed the EPED1.0 model prediction by more than 50%. Pedestal stability analysis performed with the ELITE code indicates that the stability against low-n kink/peeling modes is improved with increased auxiliary heating power and the operational point in the grassy ELM regime is located near the ballooning boundary. The pedestal stability characteristics during the grassy ELM crashes have been investigated in comparison with the giant ELM crashes based on plasma profiles experimentally measured with high resolution and accuracy. It has been found that the underlying mechanism for the observed small-amplitude ELM crashes is mainly the expansion of the ballooning stability boundary induced by an initial radially localized collapse in the pedestal, which helps to stop the growth of instabilities and further collapse of the pedestal. The effect of electron density pedestal on mitigating edge localized instabilities has been analyzed by numerical simulation, suggesting that the electron density pedestal characterized by high n e,sep /n e,ped and low density gradient helps to stabilize peeling-ballooning modes because of a low pressure pedestal gradient and to lower the ballooning boundary mainly because of a low ion diamagnetic frequency in the pedestal region, thus triggering ballooning instabilities and producing the intrinsic grassy ELMs. Numerical simulation of the Chinese fusion engineering test reactor (CFETR) with the SOLPS code indicates that the separatrix density might be insensitive to the electron diffusivity in the pedestal region and increase with the power flowing from the core region to the edge region. Furthermore, pedestal stability analysis suggests that the flat density pedestal with high separatrix density obtained in the high-power plasma in CFETR would make the operational point close to the ballooning boundary, which is considered to help destabilize ballooning instabilities and facilitate the access to the grassy ELM regime.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Randomized Sketching Algorithms for Low-Memory Dynamic Optimization

This paper develops a novel limited-memory method to solve dynamic optimization problems. The memory requirements for such problems often present a major obstacle, particularly for problems with PDE constraints such as optimal flow control, full waveform inversion, and optical tomography. In these problems, PDE constraints uniquely determine the state of a physical system for a given control; the goal is to find the value of the control that minimizes an objective. While the control is often low dimensional, the state is typically more expensive to store. This paper suggests using randomized matrix approximation to compress the state as it is generated and shows how to use the compressed state to reliably solve the original dynamic optimization problem. Concretely, the compressed state is used to compute approximate gradients and to apply the Hessian to vectors. The approximation error in these quantities is controlled by the target rank of the sketch. This approximate first- and second-order information can readily be used in any optimization algorithm. As an example, we develop a sketched trust-region method that adaptively chooses the target rank using a posteriori error information and provably converges to a stationary point of the original problem. Numerical experiments with the sketched trust-region method show promising performance on challenging problems such as the optimal control of an advection-reaction-diffusion equation and the optimal control of fluid flow past a cylinder.

97 MATHEMATICS AND COMPUTING↗