SEARCH · Engineering Papers

Results for “multiscale inversion”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5

All sources NASA NTRS DOE OSTI

DOE OSTI2015

Atomistic mechanisms for bilayer growth of graphene on metal substrates

Epitaxial growth on metal substrates has been shown to be the most powerful approach in producing large-scale high-quality monolayer graphene, yet it remains a major challenge to realize uniform bilayer graphene growth. Here we carry out a comparative study of the atomistic mechanisms for bilayer graphene growth on the (111) surfaces of Cu and Ni, using multiscale approaches combining first-principles calculations and rate-equation analysis. We first show that the relatively weak graphene-Cu interaction enhances the lateral diffusion and effective nucleation of C atoms underneath the graphene island, thereby making it more feasible to grow bilayer graphene on Cu. In contrast, the stronger graphene-Ni interaction suppresses the lateral mobility and dimerization of C atoms underneath the graphene, making it unlikely to achieve controlled growth of bilayer graphene on Ni. We then determine the critical graphene size beyond which nucleation of the second layer will take place. Intriguingly, the critical size exhibits an effective inverse "Ehrlich-Schwoebel barrier" effect, becoming smaller for faster C migration from the Cu surface to the graphene-Cu interface sites across the graphene edge. Lastly, these findings allow us to propose a novel alternating growth scheme to realize mass production of bilayer graphene.

36 MATERIALS SCIENCE↗

DOE OSTI2014

Review of structure representation and reconstruction on mesoscale and microscale

Structure representation and reconstruction on mesoscale and microscale is critical in material design, advanced manufacturing and multiscale modeling. Microstructure reconstruction has been applied in different areas of materials science and technology, structural materials, energy materials, geology, hydrology, etc. This review summarizes the microstructure descriptors and formulations used to represent and algorithms to reconstruct structures at microscale and mesoscale. In the stochastic methods using correlation function, different optimization approaches have been adapted for objective function minimization. A variety of reconstruction approaches are compared in efficiency and accuracy.

Microstructure reconstruction, inverse design, mic↗

DOE OSTI2023

Machine learning from RANS and LES to inform coarse grid simulations

Nuclear system thermal hydraulic analysis has historically relied on computationally inexpensive 1D codes. However, such tools are unable to capture multiscale multidimensional effects in large nuclear reactor enclosures. On the other hand, simulations with higher fidelity can be too expensive for such purposes. One of the ways to reduce computational cost is to perform simulations on a coarse grid, which, unfortunately, introduces large discretization errors. In this paper, two high-to-low data-driven approaches are investigated: (1) a coarse grid turbulence model to predict eddy viscosity and (2) correction of errors in coarse grid velocity fields. The approaches aim to reduce grid- and turbulence model-induced errors in coarse grid Reynolds-averaged Navier–Stokes (RANS) simulations. Two sources of high-fidelity data, RANS and large eddy simulations (LES), are explored. To extract the eddy viscosity from the LES data, an inverse optimization problem is solved. However, the LES eddy viscosity is shown to be comparable to the RANS eddy viscosity in terms of error reduction. Therefore, the directly available RANS eddy viscosity was used to develop a coarse grid data-driven turbulence model. Additionally, error correction in velocity is used to reduce the remaining uncertainties and bring the results closer to reality. In conclusion, the performance of the frameworks is demonstrated for a scaled upper plenum of a gas-cooled reactor facility.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2025

A multiscale design method using interpretable machine learning for phononic materials with closely interacting scales

Manipulating the dispersive characteristics of vibrational waves is beneficial for many applications, e.g., high-precision instruments. architected hierarchical phononic materials have sparked promise tunability of elastodynamic waves and vibrations over multiple frequency ranges. In this article, hierarchical unit-cells are obtained, where features at each length scale result in a band gap within a targeted frequency range. Our novel approach, the ‘‘hierarchical unit-cell template method,’’ is an interpretable machine-learning approach that uncovers global unit-cell shape/topology patterns corresponding to predefined band-gap objectives. A scale-separation effect is observed where the coarse-scale band-gap objective is mostly unaffected by the fine-scale features despite the closeness of their length scales, thus enabling an efficient hierarchical algorithm. Moreover, the hierarchical patterns revealed are not predefined or self-similar hierarchies as common in current hierarchical phononic materials. Furthermore, our approach offers a flexible and efficient method for the exploration of new regions in the hierarchical design space, extracting minimal effective patterns for inverse design in applications targeting multiple frequency ranges.

Architected materials↗

DOE OSTI2017

Molecular dynamics simulations of stratum corneum lipid mixtures: A multiscale perspective

The lipid matrix of the stratum corneum (SC) layer of skin is essential for human survival; it acts as a barrier to prevent rapid dehydration while keeping potentially hazardous material outside the body. While the composition of the SC lipid matrix is known, the molecular-level details of its organization are difficult to infer experimentally, hindering the discovery of structure-property relationships. To this end, molecular dynamics simulations, which give molecular-level resolution, have begun to play an increasingly important role in understanding these relationships. Furthermore, most simulation studies of SC lipids have focused on preassembled bilayer configurations, which, owing to the slow dynamics of the lipids, may influence the final structure and hence the calculated properties. Self-assembled structures would avoid this dependence on the initial configuration, however, the size and length scales involved make self-assembly impractical to study with atomistic models. Here, we report on the development of coarse-grained models of SC lipids designed to study self-assembly. Building on previous work, we present the interactions between the headgroups of ceramide and free fatty acid developed using the multistate iterative Boltzmann inversion method. Validation of the new interactions is performed with simulations of preassembled bilayers and good agreement between the atomistic and coarse-grained models is found for structural properties. The self-assembly of mixtures of ceramide and free fatty acid is investigated and both bilayer and multilayer structures are found to form. This work therefore represents a necessary step in studying SC lipid systems on multiple time and length scales.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2021

An FFT-based approach for Bloch wave analysis: application to polycrystals

A method based on the Fast Fourier Transform is proposed to obtain the dispersion relation of acoustic waves in heterogeneous periodic media with arbitrary microstructures. The microstructure is explicitly considered using a voxelized Representative Volume Element (RVE). The dispersion diagram is obtained solving an eigenvalue problem for Bloch waves in Fourier space. To this aim, two linear operators representing stiffness and mass are defined through the use of differential operators in Fourier space. The smallest eigenvalues are obtained using the implicitly restarted Lanczos and the subspace iteration methods, and the required inverse of the stiffness operator is done using the conjugate gradient with a preconditioner. The method is used to study the propagation of acoustic waves in elastic polycrystals, showing the strong effect of crystal anistropy and polycrystaline texture on the propagation. It is shown that the method combines the simplicity of classical Fourier series analysis with the versatility of Finite Elements to account for complex geometries proving an efficient and general approach which allows the use of large RVEs in 3D.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Extreme-Scale Bayesian Inference for Uncertainty Quantification of Complex Simulations

Uncertainty quantification (UQ)—that is, quantifying uncertainties in complex mathematical models and their large-scale computational implementations—is widely viewed as one of the outstanding challenges facing the field of CS&E over the coming decade. The EUREKA project set to address the most difficult class of UQ problems: those for which both the underlying PDE model as well as the uncertain parameters are of extreme scale. In the project we worked on these extreme-scale challenges in the following four areas: 1. Scalable parallel algorithms for sampling and characterizing the posterior distribution that exploit the structure of the underlying PDEs and parameter-to-observable map. These include structure-exploiting versions of the randomized maximum likelihood method, which aims to overcome the intractability of employing conventional MCMC methods for solving extreme-scale Bayesian inversion problems by appealing to and adapting ideas from large-scale PDE-constrained optimization, which have been very successful at exploring high-dimensional spaces. 2. Scalable parallel algorithms for construction of prior and likelihood functions based on learning methods and non-parametric density estimation. Constructing problem-specific priors remains a critical challenge in Bayesian inference, and more so in high dimensions. Another challenge is construction of likelihood functions that capture unmodeled couplings between observations and parameters. We will create parallel algorithms for non-parametric density estimation using high dimensional N-body methods and combine them with supervised learning techniques for the construction of priors and likelihood functions. 3. Bayesian inadequacy models, which augment physics models with stochastic models that represent their imperfections. The success of the Bayesian inference framework depends on the ability to represent the uncertainty due to imperfections of the mathematical model of the phenomena of interest. This is a central challenge in UQ, especially for large-scale models. We propose to develop the mathematical tools to address these challenges in the context of extreme-scale problems. 4. Parallel scalable algorithms for Bayesian optimal experimental design (OED). Bayesian inversion yields quantified uncertainties in the model parameters, which can be propagated forward through the model to yield uncertainty in outputs of interest. This opens the way for designing new experiments to reduce the uncertainties in the model parameters and model predictions. Such experimental design problems have been intractable for large-scale problems using conventional methods; we will create OED algorithms that exploit the structure of the PDE model and the parameter-to-output map to overcome these challenges. Parallel algorithms for these four problems were created, analyzed, prototyped, implemented, tuned, and scaled up for leading-edge supercomputers, including UT-Austin’s own 10 petaflops Stampede system, ANL’s Mira system, and ORNL’s Titan system. While our focus is on fundamental mathematical/computational methods and algorithms, we will assess our methods on model problems derived from several DOE mission applications, including multiscale mechanics and ice sheet dynamics.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2018

Changes in Tropical Clouds and Atmospheric Circulation Associated with Rapid Adjustment Induced by Increased Atmospheric CO2 – A Multiscale Modeling Framework Study

The radiative heating increase due to increased CO2 concentration is the primary source for the rapid adjustment of atmospheric circulation and clouds. In this study, we investigate the rapid adjustment resulting from doubling of CO2 and its physical mechanism using a multiscale modeling framework (MMF). The MMF includes an advanced higher-order turbulence closure in its cloud-resolving model component and simulates realistic shallow and deep cloud climatology and boundary layer turbulence. The rapid adjustment over the tropics is characterized by 1) reduced ascent and descent strengths over the ocean, 2) increased lower tropospheric stability (LTS) over the subsidence region, 3) shoaling of planetary boundary layers over the ocean, 4) increased deep convection over lands and shift of cloud coverage from the ocean to lands, and 5) reduced sensible (SH) and latent heat (LH) fluxes over the oceanic deep convective regions. Unlike conventional general circulation models and another MMF, a reduction in the global-mean shortwave cloud radiative cooling is not simulated, due to the increase in low clouds at lower altitudes over the ocean, resulting from reduced cloud-top entrainment due to strengthened inversion. Changes in regional circulation play a key role in cloud changes and shift of cloud coverage to lands. Weaker energy transport resulting from water vapor and cloud CO2 masking effects reduces the upward motion and convective clouds in the oceanic regions. The ocean-land transports are linked to the partitioning of surface SH and LH fluxes that increases humidity over lands and enhances deep convection over the tropical lands.

Xu, Kuan-Man↗

NASA NTRS2007

Progress in Modeling Global Atmospheric CO2 Fluxes and Transport: Results from Simulations with Diurnal Fluxes

Progress in better determining CO2 sources and sinks will almost certainly rely on utilization of more extensive and intensive CO2 and related observations including those from satellite remote sensing. Use of advanced data requires improved modeling and analysis capability. Under NASA Carbon Cycle Science support we seek to develop and integrate improved formulations for 1) atmospheric transport, 2) terrestrial uptake and release, 3) biomass and 4) fossil fuel burning, and 5) observational data analysis including inverse calculations. The transport modeling is based on meteorological data assimilation analysis from the Goddard Modeling and Assimilation Office. Use of assimilated met data enables model comparison to CO2 and other observations across a wide range of scales of variability. In this presentation we focus on the short end of the temporal variability spectrum: hourly to synoptic to seasonal. Using CO2 fluxes at varying temporal resolution from the SIB 2 and CASA biosphere models, we examine the model's ability to simulate CO2 variability in comparison to observations at different times, locations, and altitudes. We find that the model can resolve much of the variability in the observations, although there are limits imposed by vertical resolution of boundary layer processes. The influence of key process representations is inferred. The high degree of fidelity in these simulations leads us to anticipate incorporation of realtime, highly resolved observations into a multiscale carbon cycle analysis system that will begin to bridge the gap between top-down and bottom-up flux estimation, which is a primary focus of NACP.

Collatz, G. James↗

DOE OSTI2013

Numerical schemes for dynamically orthogonal equations of stochastic fluid and ocean flows

The quantification of uncertainties is critical when systems are nonlinear and have uncertain terms in their governing equations or are constrained by limited knowledge of initial and boundary conditions. Such situations are common in multiscale, intermittent and non-homogeneous fluid and ocean flows. The dynamically orthogonal (DO) field equations provide an adaptive methodology to predict the probability density functions of such flows. The present work derives efficient computational schemes for the DO methodology applied to unsteady stochastic Navier-Stokes and Boussinesq equations, and illustrates and studies the numerical aspects of these schemes. Semi-implicit projection methods are developed for the mean and for the DO modes, and time-marching schemes of first to fourth order are used for the stochastic coefficients. Conservative second-order finite-volumes are employed in physical space with new advection schemes based on total variation diminishing methods. Other results include: (i) the definition of pseudo-stochastic pressures to obtain a number of pressure equations that is linear in the subspace size instead of quadratic; (ii) symmetric advection schemes for the stochastic velocities; (iii) the use of generalized inversion to deal with singular subspace covariances or deterministic modes; and (iv) schemes to maintain orthonormal modes at the numerical level. To verify our implementation and study the properties of our schemes and their variations, a set of stochastic flow benchmarks are defined including asymmetric Dirac and symmetric lock-exchange flows, lid-driven cavity flows, and flows past objects in a confined channel. Different Reynolds number and Grashof number regimes are employed to illustrate robustness. Optimal convergence under both time and space refinements is shown as well as the convergence of the probability density functions with the number of stochastic realizations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2016

Multiscale equatorial electrojet turbulence: Energy conservation, coupling, and cascades in a baseline 2-D fluid model

In this paper, progress in understanding the coupling between plasma instabilities in the equatorial electrojet based on a unified fluid model is reported. Simulations with parameters set to various ionospheric background conditions revealed properties of the gradient-drift and Farley-Buneman instabilities. Notably, sharper density gradients increase linear growth rates at all scales, whereas variations in cross-field E × B drift velocity only affect small-scale instabilities. A formalism defining turbulent fluctuation energy for the system is introduced, and the turbulence is analyzed within this framework. This exercise serves as a useful verification test of the numerical simulations and also elucidates the physics underlying the ionospheric turbulence. Various physical mechanisms involved in the energetics are categorized as sources, sinks, nonlinear transfer, and cross-field coupling. The physics of the nonlinear transfer terms is studied to identify their roles in producing energy cascades, which explain the generation of small-scale structures that are stable in the linear regime. The theory of two-step energy cascading to generate the 3 m plasma irregularities in the equatorial electrojet is verified for the first time in the fluid regime. Finally, in addition, the nonlinearity of the system allows the possibility of an inverse energy cascade, potentially responsible for generating large-scale plasma structures at the top of the electrojet as found in different rocket and radar observations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2020

Variations of Stress Parameters in the Southern California Plate Boundary Around the South Central Transverse Ranges

We examine the stress field in Southern California with a focus on the region near the South Central Transverse Ranges (SCTR), using a refined stress inversion methodology. We independently invert declustered and aftershocks focal mechanisms from the years 1981 to 2017. Comparisons between the stress field orientations and the stress ratios from these inversions provide information on local sources of the stress field. Regionally, the SHmax tends to be toward NNE and the stress ratios vary from transtensional stress regime near the Eastern California Shear Zone (ECSZ), to strike-slip faulting near the SCTR, and toward transpression near the Western Transverse Ranges. Detailed analysis of stress parameters near the SCTR indicates deviations from the regional strike-slip faulting including the following: (1) The San Bernardino Mountain area shows S Hmax direction toward NNW and transpressional stress components likely associated with the motions of the San Andreas Fault and ECSZ; (2) the Cajon Pass and San Gorgonio Pass show transpressional stress regime near the bottom of the seismogenic zone likely associated with the elevated topography; and (3) in Crafton Hills, rotation of the principal stress plunges and SHmax direction and transtensional stress regime below ~10 km, along with lower estimated apparent friction coefficient, suggest a weak fault possibly associated with deep creep. The performed multiscale analysis resolves effects of regional and local loadings. The average stress parameters in the analyzed 37 years do not show significant temporal variations of stress field near the SCTR region.

58 GEOSCIENCES↗

DOE OSTI2024

SQMS Quantum R&D in Machine Learning, Optimization and Sensing beyond Fundamental Physics Applications

This newly formed team at SQMS under the Ecosystem Thrust is looking to develop capabilities impacting societal advances outside the core domain of HEP and condensed matter physics. We explicitly leverage the experimental and algorithmic innovations developed across all groups as well as connect to broad-scope external projects of the diverse team of PIs. As the inaugural set of projects, we are studying numerically quantum machine learning models inspired by efficiently trainable echo-state and orthogonal neural networks and developing designs for related experiments to be performed on quantum processors based on SQMS SRF cQED technology and Rigetti s transmon arrays. Investigated models exploit ideas and lessons learned from multiple prior work by SQMS team members in a variety of internal and external activities [R1]. Target initial applications include noisy signal processing, potentially captured by quantum sensors or noisy QPUs, as well as simulation and classification of healthcare data. For instance, image reconstruction of the brain s electrical properties by solving the inverse Maxwell equation problem with uncertainty [R2] through a hybrid quantum-classical physics-informed architecture for time-dependent processes [R3]. The group is also investigating the application and development of novel quantum sensors based on magnetic levitation of a superconducting sphere coupled to a superconducting qubit. This coupling enables high-precision measurements of the position of the sphere, which can be used for sensitive detection of forces, enabling practical applications such as gravimetry for geophysics analysis, or accelerometry for GPS-denied navigation [R4] [R1] Rieffel, Eleanor G., Ata Akbari Asanjan, M. Sohaib Alam, Namit Anand, David E. Bernal Neira, Sophie Block, Lucas T. Brady et al. "Assessing and advancing the potential of quantum computing: A NASA case study." Future Generation Computer Systems (2024). [R2] Yu, X., Serrall s, J.E., Giannakopoulos, I.I., Liu, Z., Daniel, L., Lattanzi, R. and Zhang, Z., 2023. Pifon-ept: Mr-based electrical property tomography using physics-informed fourier networks. IEEE Journal on Multiscale and Multiphysics Computational Techniques. [R3] Wudarski, Filip, Daniel OConnor, Shaun Geaney, Ata Akbari Asanjan, Max Wilson, Elena Strbac, P. Aaron Lott, and Davide Venturelli. "Hybrid quantum-classical reservoir computing for simulating chaotic systems." arXiv preprint arXiv:2311.14105 (2023). [R4] Higgins, Gerard, Saarik Kalia, and Zhen Liu. "Maglev for dark matter: Dark-photon and axion dark matter sensing with levitated superconductors." Physical Review D 109.5 (2024): 055024.

Venturelli, Davide↗

DOE OSTI2024

Impact of Momentum Perturbation on Convective Boundary Layer Turbulence

Mesoscale-to-microscale coupling is an important tool for conducting turbulence-resolving multiscale simulations of realistic atmospheric flows, which are crucial for applications ranging from wind energy to wildfire spread studies. Different techniques are used to facilitate the development of realistic turbulence in the large-eddy simulation (LES) domain while minimizing computational cost. Here, we explore the impact of a simple and computationally efficient Stochastic Cell Perturbation method using momentum perturbation (SCPM-M) to accelerate turbulence generation in boundary-coupled LES simulations using the Weather Research and Forecasting model. We simulate a convective boundary layer (CBL) to characterize the production and dissipation of turbulent kinetic energy (TKE) and the variation of TKE budget terms. Furthermore, we evaluate the impact of applying momentum perturbations of three magnitudes below, up to, and above the CBL on the TKE budget terms. Momentum perturbations greatly reduce the fetch associated with turbulence generation. When applied to half the vertical extent of the boundary layer, momentum perturbations produce an adequate amount of turbulence. However, when applied above the CBL, additional structures are generated at the top of the CBL, near the inversion layer. The magnitudes of the TKE budgets produced by SCPM-M when applied at varying heights and with different perturbation amplitudes are always higher near the surface and inversion layer than those produced by No-SCPM, as are their contributions to the TKE. This study provides a better understanding of how SCPM-M reduces computational costs and how different budget terms contribute to TKE in a boundary-coupled LES simulation.

17 WIND ENERGY↗

DOE OSTI2023

Model data for numerical evaluation of photosensitive tracers as a strategy for separating surface and subsurface transient storage in streams

This model-data archive pertains to a study aimed at the numerical evaluation of the photosensitive tracers as a potential strategy for separating the effects of surface and hyporheic storage zones (SSZs and HSZs, respectively). Separating effects of SSZ and HSZ are important for accurately representing stream function as HSZs and SSZs expose solutes to significantly different biogeochemical conditions. We perform numerical experiments using a multiscale reactive transport model for stream corridors implemented in ATS code, which allows for representing multiple storage zones with their respective travel time distributions and biogeochemistry. For each of the numerical experiment, we provide in this model-data archive python wrapper to drive ATS (forward_model.py), synthetic observation (synthetic_btc.py, BTC_observed.csv), forward model files including ATS input (multiscale_transport.tpl), PFLOTRAN inputs (reactions_channel.tpl, reactions_hz.tpl, reactions_sz.tpl), MCMC (mcmc_run.py), predictive uncertainty (pred_uncert.py, BTCs_simulated_day.csv, BTCs_simulated_night.csv), MCMC outputs (tracer_test_GR.npy, tracer_test_logps.npy, tracer_test_parameters.npy) and Jupyter notebook for post-processing and visualization (post-processing.ipynb). For the denitrification application, ATS input file (denitrification_multisubgrid.tpl), PFLOTRAN input files (denitrification_channel.in, denitrification_hz.in), predictive uncertainty (pred_uncert_denitrification.py, BTCs_simulated_DO.csv, BTCs_simulated_DOC.csv, BTCs_simulated_Nitrate.csv).

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

NeuroSEM: A hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements

Multiphysics problems that are characterized by complex interactions among fluid dynamics, heat transfer, structural mechanics, and electromagnetics, are inherently challenging due to their coupled nature. While experimental data on certain state variables may be available, integrating these data with numerical solvers remains a significant challenge. Physics-informed neural networks (PINNs) have shown promising results in various engineering disciplines, particularly in handling noisy data and solving inverse problems in partial differential equations (PDEs). However, their effectiveness in forecasting nonlinear phenomena in multiphysics regimes, particularly involving turbulence, is yet to be fully established. Here, this study introduces NeuroSEM, a hybrid framework integrating PINNs with the highfidelity Spectral Element Method (SEM) solver, Nektar++. NeuroSEM leverages the strengths of both PINNs and SEM, providing robust solutions for multiphysics problems. PINNs are trained to assimilate data and model physical phenomena in specific subdomains, which are then integrated into the Nektar++ solver. We demonstrate the efficiency and accuracy of NeuroSEM for thermal convection in cavity flow and flow past a cylinder. The framework effectively handles data assimilation by addressing those subdomains and state variables where the data is available. We applied NeuroSEM to the Rayleigh-B´enard convection system, including cases with missing thermal boundary conditions and noisy datasets. Finally, we applied the proposed NeuroSEM framework to real particle image velocimetry (PIV) data to capture flow patterns characterized by horseshoe vortical structures. Our results indicate that NeuroSEM accurately models the physical phenomena and assimilates the data within the specified subdomains. The framework’s plug-and-play nature facilitates its extension to other multiphysics or multiscale problems. Furthermore, NeuroSEM is optimized for efficient execution on emerging integrated GPU-CPU architectures. This hybrid approach enhances the accuracy and efficiency of simulations, making it a powerful tool for tackling complex engineering challenges in various scientific domains.

42 ENGINEERING↗

NASA NTRS2020

Characteristics of Escaping Magnetospheric Ions Associated with Magnetic Field Fluctuations

The four Magnetospheric Multiscale (MMS) spacecraft observed a hot ow anomaly (HFA) at the boundary of the quasi-parallel and quasi-perpendicular bow shock and energetic (E > 50 keV) ion bursts exhibitinginverse dispersion in the magnetosheath on 28 December 2015. We consider the possibility that the ions seen both upstream in the foreshock and down-stream in the magnetosheath originate from the magnetosphere. The ion composition ratios, flux levels, and the spectral slopes of the energetic ion energy spectra observed in the region upstream from the bow shock and in themagnetosheath are similar to those in the outer magnetosphere but significantly differ from those seen further upstream from the bow shock at ACE. We can exclude an explanation of the particle source in terms of HFA acceleration. A simulation shows that escaping magnetospheric ions can be scattered and transported across the magnetosheath. Ground magnetometer observations indicate that a solar wind pressure decrease subsequently allows the bow shock to move outward past the MMS spacecraft placing them in the magnetosheath. This was followed by a pressure increase in the magnetosphere. We posit that the enhanced pressure applied to the magnetosphere accelerates the escaping magnetospheric ions (the betatron acceleration), resulting in the inverse dispersion.

S. H. Lee↗

DOE OSTI2022

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗