Search NASA⌕ Search

SEARCH · Search NASA

Results for “Computational Physics”

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 289 records · Page 16

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. In conclusion, we demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

Deep ensembles↗

Extension of the Highly Mobile Membrane Mimetic to Transmembrane Systems through Customized in Silico Solvents

Typically, the mechanics of the protein–lipid interactions of transmembrane proteins are difficult to capture with conventional atomic molecular dynamics, due to the slow lateral diffusion of lipids restricting sampling to states near the initial membrane configuration. The highly mobile membrane mimetic (HMMM) model accelerates lipid dynamics by modeling the acyl tails nearest to the membrane center as a fluid organic solvent while maintaining an atomic description of the lipid headgroups and short acyl tails. The HMMM has been applied to many peripheral protein systems; however, the organic solvent used to date caused deformations in transmembrane proteins by intercalating into the protein and disrupting interactions between individual side chains. We ameliorate the effect of the solvent on transmembrane protein structure through the development of two new in silico Lennard-Jones solvents. The parameters for the new solvents were determined through an extensive parameter search in order to match the bulk properties of alkanes in a highly simplified model. Using these new solvents, we substantially improve the insertion free energy profiles of 10 protein side chain analogues across the entire bilayer. Furthermore, we reduce the intercalation of solvent into transmembrane systems, resulting in native-like transmembrane protein structures from five different topological classes within a HMMM bilayer. The parametrization of the solvents, in addition to their computed physical properties, is discussed. By combining high lipid lateral diffusion with intact transmembrane proteins, we foresee the developed solvents being useful to efficiently identify membrane composition inhomogeneities and lipid binding caused by the presence of membrane proteins.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An error-controlled fast multipole method

We present a two-stage error estimation scheme for the fast multipole method (FMM). This scheme can be applied to any particle system. It incorporates homogeneous as well as inhomogeneous distributions. The FMM error as a consequence of the finite representation of the multipole expansions and the operator error is correlated with an absolute or relative user-requested energy threshold. Such a reliable error control is the basis for making reliable simulations in computational physics. Our FMM program on the basis of the two-stage error estimation scheme is available on request.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Corrected Article: 'An error-controlled fast multipole method' [J. Chem. Phys. 131, 244102 (2009)]

We present a two-stage error estimation scheme for the fast multipole method (FMM). This scheme can be applied to any particle system. It incorporates homogeneous as well as inhomogeneous distributions. The FMM error as a consequence of the finite representation of the multipole expansions and the operator error is correlated with an absolute or relative user-requested energy threshold. Such a reliable error control is the basis for making reliable simulations in computational physics. Our FMM program on the basis of the two-stage error estimation scheme is available on request.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Applications Of Monte Carlo Radiation Transport Simulation Techniques For Predicting Single Event Effects In Microelectronics

MRED (Monte Carlo Radiative Energy Deposition) is Vanderbilt University's Geant4 application for simulating radiation events in semiconductors. Geant4 is comprised of the best available computational physics models for the transport of radiation through matter. In addition to basic radiation transport physics contained in the Geant4 core, MRED has the capability to track energy loss in tetrahedral geometric objects, includes a cross section biasing and track weighting technique for variance reduction, and additional features relevant to semiconductor device applications. The crucial element of predicting Single Event Upset (SEU) parameters using radiation transport software is the creation of a dosimetry model that accurately approximates the net collected charge at transistor contacts as a function of deposited energy. The dosimetry technique described here is the multiple sensitive volume (MSV) model. It is shown to be a reasonable approximation of the charge collection process and its parameters can be calibrated to experimental measurements of SEU cross sections. The MSV model, within the framework of MRED, is examined for heavy ion and high-energy proton SEU measurements of a static random access memory.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Bayesian calibration of coarse-grained forces: Efficiently addressing transferability

Generating and calibrating forces that are transferable across a range of state-points remains a challenging task in coarse-grained (CG) molecular dynamics. In this work, we present a coarse-graining workflow, inspired by ideas from uncertainty quantification and numerical analysis, to address this problem. The key idea behind our approach is to introduce a Bayesian correction algorithm that uses functional derivatives of CG simulations to rapidly and inexpensively recalibrate initial estimates f{sub 0} of forces anchored by standard methods such as force-matching. Taking density-temperature relationships as a running example, we demonstrate that this algorithm, in concert with various interpolation schemes, can be used to efficiently compute physically reasonable force curves on a fine grid of state-points. Importantly, we show that our workflow is robust to several choices available to the modeler, including the interpolation schemes and tools used to construct f{sub 0}. In a related vein, we also demonstrate that our approach can speed up coarse-graining by reducing the number of atomistic simulations needed as inputs to standard methods for generating CG forces.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Path integral molecular dynamics for exact quantum statistics of multi-electronic-state systems

An exact approach to compute physical properties for general multi-electronic-state (MES) systems in thermal equilibrium is presented. The approach is extended from our recent progress on path integral molecular dynamics (PIMD), for quantum statistical mechanics when a single potential energy surface is involved. We first define an effective potential function that is numerically favorable for MES-PIMD and then derive corresponding estimators in MES-PIMD for evaluating various physical properties. Furthermore, its application to several representative one-dimensional and multi-dimensional models demonstrates that MES-PIMD in principle offers a practical tool in either of the diabatic and adiabatic representations for studying exact quantum statistics of complex/large MES systems when the Born-Oppenheimer approximation, Condon approximation, and harmonic bath approximation are broken.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Machine-learning force-field models for dynamical simulations of metallic magnets

We review recent advances in machine-learning (ML) force-field methods for Landau–Lifshitz–Gilbert simulations of itinerant electron magnets, focusing on their scalability and transferability. Built on the principle of locality, a deep neural-network model is developed to efficiently and accurately predict electron-mediated forces governing spin dynamics. Symmetry-aware descriptors constructed through a group-theoretical approach ensure rigorous incorporation of both lattice and spin-rotation symmetries. The framework is demonstrated using the prototypical s-d exchange model widely employed in spintronics. ML-enabled large-scale simulations reveal novel nonequilibrium phenomena, including anomalous coarsening of tetrahedral spin order on the triangular lattice and the freezing of phase-separation dynamics in lightly hole-doped, strong-coupling square-lattice systems. These results establish ML force-field frameworks as scalable, accurate, and versatile tools for modeling nonequilibrium spin dynamics in itinerant magnets.

Artificial neural networks↗

Progress and innovations in the TCV tokamak research programme

Research on the Tokamak à Configuration Variable addresses a wide range of key questions relevant to ITER and future fusion power plants. Over the past two years, highly productive experimental campaigns have led to major advances across several areas: the ITER baseline scenario; pedestal properties in low-collisionality, peeling-limited conditions; and the development of high- β N , non-inductive regimes. Alternative high-confinement scenarios have likewise received significant attention, with remarkable progress in quasi-continuous exhaust operation, X-point radiator plasmas, and negative triangularity configurations. Substantial achievements were also made in the mitigation or benign termination of runaway electron beams, in elucidating fast-ion loss mechanisms, and in improving exhaust behaviour in both conventional and alternative divertor geometries. These experimental results have been strongly supported by advances in modelling and their direct application to the experiment, ranging from gyrokinetic simulations of core and pedestal turbulence to fluid-based studies of scrape-off layer and divertor physics in diverse geometries. Plasma control has taken on an increasingly important role, with model-based and data-driven approaches now closely intertwined with physics studies. This article provides a overview of these recent activities, together with a brief outlook on forthcoming upgrades and next steps.

EPFL↗

Quantum criticality of U(1) gauge theories with fermionic and bosonic matter in two spatial dimensions

We consider relativistic U(1) gauge theories in 2+1 dimensions, with N{sub b} species of complex bosons and N{sub f} species of Dirac fermions at finite temperature. The quantum phase transition between the Higgs and Coulomb phases is described by a conformal field theory (CFT). At large N{sub b} and N{sub f}, but for arbitrary values of the ratio N{sub b}/N{sub f}, we present computations of various critical exponents and universal amplitudes for these CFTs. We make contact with the different spin liquids, charge liquids, and deconfined critical points of quantum magnets that these field theories describe. We compute physical observables that may be measured in experiments or numerical simulations of insulating and doped quantum magnets.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Applying the Scientific Method of Cybersecurity Research

The cyber environment has rapidly evolved from a curiosity to an essential component of the contemporary world. As the cyber environment has expanded and become more complex, so have the nature of adversaries and styles of attacks. Today, cyber incidents are an expected part of life. As a result, cybersecurity research emerged to address adversarial attacks interfering with or preventing normal cyber activities. Historical response to cybersecurity attacks is heavily skewed to tactical responses with an emphasis on rapid recovery. While threat mitigation is important and can be time critical, a knowledge gap exists with respect to developing the science of cybersecurity. Such a science will enable the development and testing of theories that lead to understanding the broad sweep of cyber threats and the ability to assess trade-offs in sustaining network missions while mitigating attacks. The Asymmetric Resilient Cybersecurity Initiative at Pacific Northwest National Laboratory is a multi-year, multi-million dollar investment to develop approaches for shifting the advantage to the defender and sustaining the operability of systems under attack. The initiative established a Science Council to focus attention on the research process for cybersecurity. The Council shares science practices, critiques research plans, and aids in documenting and reporting reproducible research results. The Council members represent ecology, economics, statistics, physics, computational chemistry, microbiology and genetics, and geochemistry. This paper reports the initial work of the Science Council to implement the scientific method in cybersecurity research. The second section describes the scientific method. The third section in this paper discusses scientific practices for cybersecurity research. Section four describes initial impacts of applying the science practices to cybersecurity research.

Tardiff, Mark F.↗

MultiMatTest, Version 1.0

MultiMatTest is a demonstration code to support a talk and paper titled A Comparative Study of Multi-material Data Structures for Computational Physics Applications. The application times basic operations for different representations of multi-material data structures.

Robey, RobertW.↗

Finite Element Solver for Fission Dynamics

FELIX is a physics computer code used to model fission fragment mass distributions in a fully quantum-mechanical, microscopic framework that only relies on our current knowledge of nuclear forces. It is an implementation of the time-dependent generator coordinate method (TDGCM), which simulates the dynamics of a collective quantum wave-packet assuming the motion is adiabatic. In typical applications of the TDGCM, the nuclear collective wavepacket is obtained as a superposition of wavefunctions obtained by solving the Hartree-Fock-Bogoliubov equations of nuclear density functional theory (DFT). The program calculates at each time step the coefficients of that superposition.

Regnier, D.↗

DLIO: A DATA-CENTRIC BENCHMARK FOR DEEP LEARNING APPLICATIONS

SF-22-136 Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize the computational performance of deep learning frameworks. However, the same cannot be said for I/O performance. As deep learning algorithms rely on big-data volume and variety to effectively train neural networks accurately, I/O is a significant bottleneck on large-scale distributed deep learning training. DLIO, is a novel representative benchmark suite built based on the I/O profiling of the selected workloads. DLIO can be utilized to accurately emulate the I/O behavior of modern deep learning applications. Using DLIO, application developers and system software solution architects can identify potential I/O bottlenecks in their applications and guide optimizations to boost the I/O performance leading to lower training times. The storage vendor can also use DLIO as a guide for designing and optimize the storage and filesystem targeting at deep learning application.

ZHENG, HUIHUO↗

Ume: Unstructured Mesh Explorations

Ume is an open-source collection of data structures for unstructured computational meshes and some simple algorithms that operate on them. These algorithms mimic the memory access patterns of a common class of operations found in several of the computational physics simulation codes developed at Los Alamos National Laboratory. The intent is that Ume can be used by hardware vendors to understand the memory traffic created by complex codes in a simplified environment, and to explore new means of optimization for that traffic. Ume is provided as a source-code C++ library and includes several applications that demonstrate the use of that library.

Henning, Paul↗

Toward Quantity-of-Interest Preserving Lossy Compression for Scientific Data

Today's scientific simulations and instruments are producing a large amount of data, leading to difficulties in storing, transmitting, and analyzing these data. While error-controlled lossy compressors are effective in significantly reducing data volumes and efficiently developing databases for multiple scientific applications, they mainly support error controls on raw data, which leaves a significant gap between the data and user's downstream analysis. This may cause unqualified uncertainties in the outcomes of the analysis, a.k.a quantities of interest (QoIs), which are the major concerns of users in adopting lossy compression in practice. In this paper, we propose rigorous mathematical theories to preserve four families of QoIs that are widely used in scientific analysis during lossy compression along with practical implementations. Specifically, we first develop the error control theory for univariate QoIs which are essential for computing physical properties such as kinetic energy, followed by multivariate QoIs that are more commonly used in real-world applications. The proposed method is integrated into a state-of-the-art compression framework in a modular fashion, which could easily adapt to new QoIs and new compression algorithms. Experiments on real-world datasets demonstrate that the proposed method provides faithful error control on important QoIs including kinetic energy, regional average, and isosurface without trials and errors, while offering compression ratios that are up to 4x of the compression ratios provided by state-of-the-art compressors.

Jiao, Pu↗

Delivering Contextual Knowledge and Critical Skills of Disruptive Technologies through Problem-Based Learning in Research Experiences for Undergraduates Setting

The recent development in transportation, such as energy-efficient and autonomous vehicles, defines a condition for the students in transportation engineering. Students in the field of transportation engineering should be ready upon their graduation with new knowledge and skills that are compatible with the need of the industry and sustainable engineering practices.During summers of 2018 and 2019, we developed and implemented an eight-week program to increase the knowledge and skills of students coming from multidisciplinary fields related to autonomous vehicles. Problem of “How much will platooning reduce fuel consumption and emissions per vehicle mile traveled?” was instrumentalized in subsequent activities to introduce the comprehensive knowledge structure of autonomous vehicles.The engineering concept of reducing the cost and sustainability was embedded in the leading research question that helped us to develop and implement activities on an overall knowledge structure in autonomous vehicles. The goal of using problem-based learning activities was not to encourage the students to focus on reaching the solution merely. We aimed to introduce the multidisciplinary knowledge and critical skills aspects of learning about disruptive technologies.In this paper, we will discuss how a multidisciplinary research approach was incorporated into a problem-based learning activity. The students were introduced the subjects related to math, physics, computer science, and biology as the integration of the knowledge structure of autonomous vehicles. We will also present the results on students’ use of critical skills such as machine learning and computer programming.

Comert, Gurcan↗