Search NASA⌕ Search

SEARCH · Search NASA

Results for “gradient methods”

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 1,189 records · Page 66

Differentiable lagrangian shock hydrodynamics with application to stable shock acceleration of density interfaces

We develop a gradient based optimization approach for the equations of compressible, Lagrangian hydrodynamics and demonstrate how it can be employed to automatically uncover strategies to control hydrodynamic instabilities arising from shock acceleration of density interfaces. Strategies for controlling the Richtmyer-Meshkov instability (RMI) are of great benefit for inertial confinement fusion (ICF) where shock interactions with many small imperfections in the density interface lead to instabilities which rapidly grow over time. These instabilities lead to mixing which, in the case of laser driven ICF, quenches the runaway fusion process ruining the potential for positive energy return. Here, we demonstrate that control of these instabilities can be achieved by optimization of initial conditions with ( > 100) parameters. Optimizing over a large parameter space like this is not possible with gradient-free optimization strategies. This requires computation of the gradient of the outputs of a numerical solution to the equations of Lagrangian hydrodynamics with respect to the inputs. We show that the efficient computation of these gradients is made possible via a judicious application of (i) adjoint methods, the exact formal representation of sensitivities involving partial differential equations, and (ii) automatic differentiation (AD), the algorithmic calculation of derivatives of functions. Careful regularization of multiple operators including artificial viscosity and timestep control is required. We perform design optimization of > 100 parameter energy field driving the Richtmyer Meshkov instability showing significant suppression while simultaneously enhancing the acceleration of the interface relative to a nominal baseline case.

Hydrophysics↗

AN INTERNAL CONVECTIVE HEATING TECHNIQUE FOR DIFFUSION BONDING ARRAYED MICROCHANNEL ARCHITECTURES

Diffusion bonding cycle times can be a large factor in the production cost of metal microchannel devices. The challenge is to significantly minimize bonding cycle times through rapid heating and cooling within the bonding process. A novel method is described which takes advantage of the internal flow passages within microchannel devices for convective heat transfer during the bonding process. The internal convective heating (ICH) technique makes use of heated inert gas to provide the microchannel assembly with rapid and uniform heat input. Results demonstrate that the ICH technique is feasible, capable of producing microchannels with higher dimensional integrity and shorter bonding cycle times than traditional vacuum hot press methods. Results suggest that this may be due to smaller thermal gradients within microchannel devices during the ICH bonding cycle.

42 ENGINEERING↗

Coupling High-Throughput Experiments and Regression Algorithms to Optimize PGM-Free ORR Electrocatalyst Synthesis

Over the past decades, significant improvement has been achieved in the performance of platinum group metal-free (PGM-free) materials as an alternative to Pt-based electrocatalysts for oxygen reduction reaction (ORR). However, further progress in ORR activity requires evaluation of precursors and synthesis approaches. In response to this challenge, we generated a first of its kind experimental data set of 36 samples using high-throughput synthesis and activity measurements. Several control parameters (e.g., Fe precursor identity, the precursor content, and pyrolysis temperature) were varied. We then developed several state-of-the-art machine learning (ML) based regression models to predict ORR activity, dependent on selected synthesis variables. Through an iterative algorithm, higher prediction accuracy (smaller root-mean-square error) was achieved. We identified that gradient boosting regression (GBR) and support vector regression (SVR), among several methods, work best for this data set. Aided by our ML-based surrogate models, we decided to alter catalyst synthesis conditions, which resulted in a 36% increase in measured ORR activity in comparison to the maximum ORR mass activity value of 21.9 A/g catalyst in the original data set. Overall, this combined experiment and machine learning approach represents a promising path forward toward developing highly efficient next-generation ORR electrocatalysts and, more generally, functional materials.

25 ENERGY STORAGE↗

Selective partitioning and uphill transport enable effective Li/Mg ion separation by negatively charged membranes

Efficient separation of lithium (Li + ) and magnesium (Mg 2+ ) is critical for enhancing sustainable lithium extraction from natural brines, which is vital for battery production and renewable energy technologies. Here we present a method for highly selective Li + /Mg 2+ separation driven by concentration gradients across negatively charged membranes with high charge densities. In contrast to typical electric field-driven transport in negatively charged membranes, where divalent cations generally permeate faster than monovalent cations, Li + ions in our system permeate the membrane at substantially higher rates than Mg 2+ ions. This unexpected selectivity stems from the selective ion partitioning properties of the membrane and the uphill transport of Mg 2+ ions against their external concentration gradient. We demonstrate the efficacy of this separation approach through bench-scale dialysis experiments using a model Atacama brine solution, achieving efficient separation of monovalent and divalent cations. As a result, the high separation efficiency observed in this study suggests a promising approach for monovalent/divalent ion separations, offering higher selectivity compared to current technologies.

36 MATERIALS SCIENCE↗

Simultaneous quantification of uranium( VI ), samarium, nitric acid, and temperature with combined ensemble learning, laser fluorescence, and Raman scattering for real-time monitoring

In this work, laser-induced fluorescence spectroscopy (LIFS), Raman spectroscopy, and a stacked regression ensemble was developed for near real-time quantification of uranium(VI) (1–100 μg mL –1 ), samarium (0–200 μg mL –1 ) and nitric acid (0.1–4 M) with varying temperature (20 °C–45 °C). LIFS applications range from fundamental lab-scale studies to real-time process monitoring at industrial levels, such as nuclear reprocessing applications, provided the phenomena affecting the fluorescence spectrum are accounted for (e.g., absorption, quenching, complexation). Multiple chemometric models were examined and compared to a more traditional multivariate regression approach called partial least squares (PLS). Results obtained on synthetic samples selected using D-optimal experimental design indicated that a stacked regression method, which included ridge regression, random forest, PLS, and an eXtreme gradient boost algorithm, successfully measured uranium(VI) concentrations directly in nitric acid without measuring luminescence lifetimes or standard addition. The top model resulted in percent root-mean-square error of prediction values of 5.2, 1.9, 3.0, and 2.3% for U(VI), Sm 3+ , HNO 3 , and temperature, respectively. The approach may be useful for quantifying fluorescent fission products (e.g., Sm 3+ ) to provide information on burnup of irradiated nuclear fuel. This novel framework reinforces the applicability of LIFS for real-time applications in nuclear fuel cycle applications.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

A benchmark study of electronic excitation energies, transition moments, and excited-state energy gradients on the nicotine molecule

In this work, we report a comparative study of computed excitation energies, oscillator strengths, and excited-state energy gradients of (S)-nicotine, chosen as a test case, using multireference methods, coupled cluster singles and doubles, and methods based on time-dependent density functional theory. This system was chosen because its apparent simplicity hides a complex electronic structure, as several different types of valence excitations are possible, including n-π{sup *}, π-π{sup *}, and charge-transfer states, and in order to simulate its spectrum it is necessary to describe all of them consistently well by the chosen method.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Steady state likelihood ratio sensitivity analysis for stiff kinetic Monte Carlo simulations

Kinetic Monte Carlo simulation is an integral tool in the study of complex physical phenomena present in applications ranging from heterogeneous catalysis to biological systems to crystal growth and atmospheric sciences. Sensitivity analysis is useful for identifying important parameters and rate-determining steps, but the finite-difference application of sensitivity analysis is computationally demanding. Techniques based on the likelihood ratio method reduce the computational cost of sensitivity analysis by obtaining all gradient information in a single run. However, we show that disparity in time scales of microscopic events, which is ubiquitous in real systems, introduces drastic statistical noise into derivative estimates for parameters affecting the fast events. In this work, the steady-state likelihood ratio sensitivity analysis is extended to singularly perturbed systems by invoking partial equilibration for fast reactions, that is, by working on the fast and slow manifolds of the chemistry. Derivatives on each time scale are computed independently and combined to the desired sensitivity coefficients to considerably reduce the noise in derivative estimates for stiff systems. The approach is demonstrated in an analytically solvable linear system.

Núñez, M.↗

High-dimensional multi-fidelity Bayesian optimization for quantum control

Abstract We present the first multi-fidelity Bayesian optimization (BO) approach for solving inverse problems in the quantum control of prototypical quantum systems. Our approach automatically constructs time-dependent control fields that enable transitions between initial and desired final quantum states. Most importantly, our BO approach gives impressive performance in constructing time-dependent control fields, even for cases that are difficult to converge with existing gradient-based approaches. We provide detailed descriptions of our machine learning methods as well as performance metrics for a variety of machine learning algorithms. Taken together, our results demonstrate that BO is a promising approach to efficiently and autonomously design control fields in general quantum dynamical systems.

Computer Science↗

FeRh ground state and martensitic transformation

Cubic B2 FeRh exhibits a metamagnetic transition [(111) antiferromagnet (AFM) to ferromagnet (FM)] around 353 K and remains structurally stable at higher temperatures. However, the calculated zero-Kelvin phonons of AFM FeRh exhibit imaginary modes at M points in the Brillouin zone, indicating a premartensitic instability, which is a precursor to a martensitic transformation at low temperatures. Combining electronic-structure calculations with ab initio molecular dynamics, conjugate gradient relaxation, and the solid-state nudged-elastic band methods, we predict that AFM B2 FeRh becomes unstable at ambient pressure and transforms without a barrier to an AFM(111) orthorhombic (martensitic) ground state below 90±10K. In conclusion, we also consider competing structures, in particular, a tetragonal AFM(100) phase that is not the global ground state, as proposed, but a constrained solution.

36 MATERIALS SCIENCE↗

SU-F-T-570: Comparison of Synthetic Diamond, Microionization Chamber, and Radiochromic Film for Absolute Dosimetry of VMAT Radiosurgery

Purpose: Patient specific quality assurance of stereotactic radiosurgery (SRS) plans is challenging because of small target sizes and high dose gradients. We compared three detectors for dosimetry of VMAT SRS plans. Methods: The dose at the center of seventeen targets was measured using a synthetic diamond detector (2.2 mm diameter, 1 µm thickness), a 0.007 cm{sup 3} ionization chamber, and radiochromic film. Measurements were made in a PMMA phantom in the clinical geometry – all gantry and table angles were delivered as planned. The diamond and chamber positions were offset by 1 cm from the film plane, so the isocenter was shifted accordingly to place the center of the target at the detector of interest. To ensure accurate detector placement, the phantom was positioned using kV images. To account for the shift-induced difference in geometry and differing prescription doses between plans, the measurements were normalized to the expected dose calculated by the treatment planning system. Results: The target sizes ranged from 2.8 mm to 34.8 mm (median 14.8 mm). The mean measurement-to-plan ratios were 1.054, 1.076, and 1.023 for RCF, diamond, and chamber, respectively. The mean difference between the chamber and film was −3.2% and between diamond and film was 2.2%. For targets larger than 15 mm, the mean difference relative to film was −0.8% and 0.1% for chamber and diamond, respectively, whereas for targets smaller than 15 mm, the difference was −5.3% and 4.2% for chamber and diamond, respectively. The difference was significant (p=0.005) using the two-sample Kolmogorov-Smirnov test. Conclusion: The detectors agree for target sizes larger than 15 mm. Relative to film, for smaller targets the diamond detector over-responds, whereas the ionization chamber under-responds. Further work is needed to characterize detector response in modulated SRS fields.

60 APPLIED LIFE SCIENCES↗

Binary Control Pulse Optimization for Quantum Systems

Quantum control aims to manipulate quantum systems toward specific quantum states or desired operations. Designing highly accurate and effective control steps is vitally important to various quantum applications, including energy minimization and circuit compilation. In this paper we focus on discrete binary quantum control problems and apply different optimization algorithms and techniques to improve computational efficiency and solution quality. Specifically, we develop a generic model and extend it in several ways. We introduce a squared L 2 -penalty function to handle additional side constraints, to model requirements such as allowing at most one control to be active. We introduce a total variation (TV) regularizer to reduce the number of switches in the control. We modify the popular gradient ascent pulse engineering (GRAPE) algorithm, develop a new alternating direction method of multipliers (ADMM) algorithm to solve the continuous relaxation of the penalized model, and then apply rounding techniques to obtain binary control solutions. We propose a modified trust-region method to further improve the solutions. Our algorithms can obtain high-quality control results, as demonstrated by numerical studies on diverse quantum control examples.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Sorting process of nanoparticles and applications of same

In one aspect of the present invention, a method for sorting nanoparticles includes preparing a high-viscosity density gradient medium filled in a container, dispersing nanoparticles into an aqueous solution to form a suspension of the nanoparticles, each nanoparticle having one or more cores and a shell encapsulating the one or more cores, layering the suspension of the nanoparticles on the top of the high-viscosity density gradient medium in the container, and centrifugating the layered suspension of the nanoparticles on the top of the high-viscosity density gradient medium in the container at a predetermined speed for a predetermined period of time to form a gradient of fractions of the nanoparticles along the container, where each fraction comprises nanoparticles in a respective one of aggregation states of the nanoparticles.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Visual, Optical and Replica Inspections: Surface Preparation of 650 MHz NB Cavity for PIP-II Linac

Surface preparation of niobium superconducting RF cavities is a critical step for achieving good RF performance under the superconducting state. Surface defect, roughness, and contamination affect the accelerating gradient and quality factor of the cavities. We report surface inspection methods used to control the surface processing of 650 MHz cavities designated for the pre-production and prototype cryomodules for PIP-II linac. The cavity surface was routinely inspected visually, with an optical camera, and by microscopic scanning of surface replicas. This article covers details on the surface inspection methods and surface polishing process used to repair the surface.

43 PARTICLE ACCELERATORS↗

Coupling Finite Element and Finite Volume Simulation Within MOOSE

This work focuses on adding capability to couple finite element (FE) variables into finite volume (FV) physics within INL’s Multiphysics Object-Oriented Simulation Environment (MOOSE). This coupling can allow for improvement of multiphysics simulations where one set of physics is best suited for an FE discretization, while another set is best suited for an FV discretization. Electrohydrodynamics, which govern liquid metal reactor concepts and plasma dynamics, is a fitting example where the electromagnetic field equations are solved with FE and the fluid flow is solved with FV. The new FE to FV coupling method can be summarized as taking the element or face average of the FE variable value or gradient and applying that quantity directly in the FV equation objects.

97 MATHEMATICS AND COMPUTING↗

Neutron Electric Dipole Moments with Clover Fermions

We present preliminary results for the contributions to the neutron EDM arising from the QCD θ-term, the Weinberg three-gluon and the quark chromo-EDM operators from our ongoing lattice calculations using clover valence quarks on the MILC HISQ lattices. We use the gradient-flow technique to smooth the lattices and renormalize the gluonic operators, and use the Schwinger source method to incorporate the quark chromo-EDM interactions in the quark propagator. For the QCD θ-term and the Weinberg three-gluon operator, we report results in the gradient-flow scheme from 8 ensembles at four lattice spacings and three pion masses, including 2 physical pion mass ensembles. For the quark chromo-EDM, unrenormalized results are presented at two lattice spacings, a=0.12 and 0.09 fm, and two pion masses, M π =310 MeV and 220 MeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Feasibility of bit patterned media for HAMR at 5 Tb/in{sup 2}

We have investigated the feasibility of BPM for HAMR via Finite Difference Time Domain and atomistic simulation and we have substantiated the feasibility of 5 Tb/in{sup 2} with two filling factors 25% and 56% even when the maximum on-track bit temperature is below the Curie temperature. The success of this underheated switching is attributed to sufficiently low anisotropy instead of reduction of Curie temperature. The temperature gradient in the cross-track direction is almost doubled if the optical head width is reduced by half, indicating the possibility of higher areal densities. Moreover, contrary to continuous media, we also found that the power absorption peaks at the bottom of the bit patterned FePt when the media is illuminated from above, which is probably due to stronger coupling there between FePt and the surrounding materials.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A NEW APPROACH TO DETERMINE OPTICALLY THICK H{sub 2} COOLING AND ITS EFFECT ON PRIMORDIAL STAR FORMATION

We present a new method for estimating the H{sub 2} cooling rate in the optically thick regime in simulations of primordial star formation. Our new approach is based on the TreeCol algorithm, which projects matter distributions onto a spherical grid to create maps of column densities for each fluid element in the computational domain. We have improved this algorithm by using the relative gas velocities to weight the individual matter contributions with the relative spectral line overlaps, in order to properly account for the Doppler effect. We compare our new method to the widely used Sobolev approximation, which yields an estimate for the column density based on the local velocity gradient and the thermal velocity. This approach generally underestimates the photon escape probability because it neglects the density gradient and the actual shape of the cloud. We present a correction factor for the true line overlap in the Sobolev approximation and a new method based on local quantities, which fits the exact results reasonably well during the collapse of the cloud, with the error in the cooling rates always being less than 10%. Analytical fitting formulae fail at determining the photon escape probability after formation of the first protostar (error of ∼40%) because they are based on the assumption of spherical symmetry and therefore break down once a protostellar accretion disk has formed. Our method yields lower temperatures and hence promotes fragmentation for densities above ∼10{sup 10} cm{sup –3} at a distance of ∼200 AU from the first protostar. Since the overall accretion rates are hardly affected by the cooling implementation, we expect Pop III stars to have lower masses in our simulations, compared to the results of previous simulations that used the Sobolev approximation.

79 ASTRONOMY AND ASTROPHYSICS↗