Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Matrix Models”

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 847 records · Page 47
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Physics-constrained Deep Recurrent Neural Models of Building Thermal Dynamics

We develop physics-constrained and control-oriented predictive deep learning models for the thermal dynamics of a real-world commercial office building. The proposed method is based on the systematic encoding of physics-based prior knowledge into a structured recurrent neural architecture. Specifically, our model mimics the structure of the building thermal dynamics model and leverages penalty methods to model inequality constraints. Additionally, we use constrained matrix parameterization based on the Perron-Frobenius theorem to bound the eigenvalues of the learned network weights. We interpret the stable eigenvalues as dissipativeness of the learned building thermal model. We demonstrate the effectiveness of the proposed approach on a dataset obtained from an office building with $20$ thermal zones.

Building Energy, structured neural networks↗
DOE OSTI2014

The Coal-Seq III Consortium. Advancing the Science of CO 2 Sequestration in Coal Seam and Gas Shale Reservoirs

The Coal-Seq consortium is a government-industry collaborative that was initially launched in 2000 as a U.S. Department of Energy sponsored investigation into CO2 sequestration in deep, unmineable coal seams. The consortium’s objective aimed to advancing industry’s understanding of complex coalbed methane and gas shale reservoir behavior in the presence of multi-component gases via laboratory experiments, theoretical model development and field validation studies. Research from this collaborative effort was utilized to produce modules to enhance reservoir simulation and modeling capabilities to assess the technical and economic potential for CO2 storage and enhanced coalbed methane recovery in coal basins. Coal-Seq Phase 3 expands upon the learnings garnered from Phase 1 & 2, which has led to further investigation into refined model development related to multicomponent equations-of-state, sorption and diffusion behavior, geomechanical and permeability studies, technical and economic feasibility studies for major international coal basins the extension of the work to gas shale reservoirs, and continued global technology exchange. The first research objective assesses changes in coal and shale properties with exposure to CO2 under field replicated conditions. Results indicate that no significant weakening occurs when coal and shale were exposed to CO2, therefore, there was no need to account for mechanical weakening of coal due to the injection of CO2 for modeling. The second major research objective evaluates cleat, Cp, and matrix, Cm, swelling/shrinkage compressibility under field replicated conditions. The experimental studies found that both Cp and Cm vary due to changes in reservoir pressure during injection and depletion under field replicated conditions. Using laboratory data from this study, a compressibility model was developed to predict the pore-volume compressibility, Cp, and the matrix compressibility, Cm, of coal and shale, which was applied to modeling software to enhance model robustness. Research was also conducted to improve algorithms and generalized adsorption models to facilitate realistic simulation of CO2 sequestration in coal seams and shale gas reservoirs. The interaction among water and the adsorbed gases, carbon dioxide (CO2), methane (CH4), and nitrogen (N2) in coalbeds is examined using experimental in situ laboratory techniques to comprehensively model CBM production and CO2 sequestration in coals. An equation of state (EOS) module was developed which is capable of predicting the density of pure components and mixtures involving the wet CBM gases CH4, CO2, and N2 at typical reservoir condition, and is used to inform CO2 injection models. The final research objective examined the effects adsorbed CO2 has on coal strength and permeability. This research studied the weakening or failure of coal by the adsorption of CO2 from empirically derived gas production data to develop models for advanced modeling of permeability changes during CO2 sequestration. The results of this research effort have been used to construct a new and improved model for assessing changes in permeability of coal reservoirs due CO2 injection. The modules developed from these studies and knowledge learned are applied to field validation and basin assessment studies. These data were used to assess the flow and storage of CO2 in a shale reservoir, test newly developed code against large-scale projects, and conduct a basin-oriented review of coal storage potential in the San Juan Basin. The storage potential and flow of CO2 was modeled for shale sequestration of a proprietary Marcellus Shale horizontal gas production well using COMET3 simulation software. Simulation results from five model runs indicate that stored CO2 quantities are linked to the duration of primary production preceding injection. Matrix CO2 saturation is observed to increase in each shale zone after injection with an increase in primary production, and the size of the CO2 plume is also observed to increase in size the longer initial production is sustained. The simulation modules developed around the Coal-Seq experimental work are also incorporated into a pre-existing large-scale numerical simulation model of the Pump Canyon CO2-ECBM pilot in the San Juan Basin. The new model was applied to re-history match the data set to explore the improvements made in permeability prediction against previously published data sets and to validate this module. The assessment of the new data, however, indicates that the impact of the variable Cp is negligible on the overall behavior of the coal for CO2 storage purposes. Applying these new modules, the San Juan Basin and the Marcellus Shale are assessed for their technical ECBM/AGR and CO2 storage potential and the economic potential of these operations. The San Juan Basin was divided into 4 unique geographic zones based on production history, and the Marcellus was divided into nine. Each was assessed based upon each zone’s properties, and simulations were run to assess the potential of full Basin development. Models of a fully developed San Juan Basin suggest the potential for up to 104 Tcf of CO2 storage, and 12.3 Tcf of methane recovery. The Marcellus models suggest 1,248 Tcf of CO2 storage and 924 Tcf of AGR. The economics are deemed favorable where credits cover the cost of CO2 in the San Juan Basin, and in many cases in the Marcellus, but to maximize storage potential, credits need to extend to pay the operator to store CO2.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2024

A flexible class of priors for orthonormal matrices with basis function-specific structure

Statistical modeling of high-dimensional matrix-valued data motivates the use of a low-rank representation that simultaneously summarizes key characteristics of the data and enables dimension reduction. Low-rank representations commonly factor the original data into the product of orthonormal basis functions and weights, where each basis function represents an independent feature of the data. However, the basis functions in these factorizations are typically computed using algorithmic methods that cannot quantify uncertainty or account for basis function correlation structure a priori. While there exist Bayesian methods that allow for a common correlation structure across basis functions, empirical examples motivate the need for basis function-specific dependence structure. We propose a prior distribution for orthonormal matrices that can explicitly model basis function-specific structure. The prior is used within a general probabilistic model for singular value decomposition to conduct posterior inference on the basis functions while accounting for measurement error and fixed effects. We discuss how the prior specification can be used for various scenarios and demonstrate favorable model properties through synthetic data examples. Finally, we apply our method to two-meter air temperature data from the Pacific Northwest, enhancing our understanding of the Earth system’s internal variability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Finite-temperature density matrix embedding theory

We describe a formulation of the density matrix embedding theory at finite temperature. We present a generalization of the ground-state bath orbital construction that embeds a mean-field finite-temperature density matrix up to a given order in the Hamiltonian, or the Hamiltonian up to a given order in the density matrix. In this work, we assess the performance of the finite-temperature density matrix embedding on the one-dimensional Hubbard model both at half-filling and away from it, and the two-dimensional Hubbard model at half-filling, comparing to exact data where available, as well as results from finite-temperature density matrix renormalization group, dynamical mean-field theory, and dynamical cluster approximations. The accuracy of finite-temperature density matrix embedding appears comparable to that of the ground-state theory, with, at most, a modest increase in bath size, and competitive with that of cluster dynamical mean-field theory.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Machine-learning prediction for quasiparton distribution function matrix elements

There have been rapid developments in the direct calculation in lattice QCD (LQCD) of the Bjorken-x dependence of hadron structure through large-momentum effective theory (LaMET). LaMET overcomes the previous limitation of LQCD to moments (that is, integrals over Bjorken x) of hadron structure, allowing LQCD to directly provide the kinematic regions where the experimental values are least known. LaMET requires large-momentum hadron states to minimize its systematics and allow us to reach small- x reliably. This means that very fine lattice spacing to minimize lattice artifacts at order $(P_za)^n$ will become crucial for next-generation LaMET-like structure calculations. Furthermore, such calculations require operators with long Wilson-link displacements, especially in finer lattice units, increasing the communication costs relative to that of the propagator inversion. In this work, we explore whether machine-learning algorithms can make predictions of correlators to reduce the computational cost of these LQCD calculations. We consider two algorithms, gradient-boosting decision tree and linear models, applied to LaMET data, the matrix elements needed to determine the kaon and $η_s$ unpolarized parton distribution functions (PDFs), meson distribution amplitude (DA), and the nucleon gluon PDF. We find that both algorithms can reliably predict the target observables with different prediction accuracy and systematic errors. The predictions from smaller displacement z to larger ones work better than those for momentum p due to the higher correlation among the data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Finite-size subthermal regime in disordered SU ( N ) -symmetric Heisenberg chains

SU⁡(N) symmetry is incompatible with the many-body localized (MBL) phase, even when strong disorder is present. However, recent studies have shown that finite-size SU⁡(2) systems exhibit nonergodic, subthermal behavior, characterized by the breakdown of the eigenstate thermalization hypothesis, and by the excited eigenstates entanglement entropy that is intermediate between area and volume law. In this paper, we extend previous studies of the SU⁡(2)-symmetric disordered Heisenberg model to larger systems, using the time-dependent density matrix renormalization group (tDMRG) method. We simulate quench dynamics from weakly entangled initial states up to long times, finding robust subthermal behavior at stronger disorder. Although we find an increased tendency towards thermalization at larger system sizes, the subthermal regime persists at intermediate time scales, nevertheless, and therefore should be accessible experimentally. At weaker disorder, we observe signatures of thermalization; however, entanglement entropy exhibits slow sublinear growth, in contrast to conventional thermalizing systems. Furthermore, we study dynamics of the SU⁡(3)-symmetric disordered Heisenberg model. Similarly, strong disorder drives the system into subthermal regime, albeit thermalizing phase is broader compared to the SU⁡(2) case. Finally, our findings demonstrate the robustness of the subthermal regime in spin chains with non-Abelian continuous symmetry, and are consistent with eventual thermalization at large system sizes and long time scales, suggested by previous studies.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2014

Renormalization of quantum discord and Bell nonlocality in the XXZ model with Dzyaloshinskii–Moriya interaction

In this paper, we have investigated the dynamical behaviors of the two important quantum correlation witnesses, i.e. geometric quantum discord (GQD) and Bell–CHSH inequality in the XXZ model with DM interaction by employing the quantum renormalization group (QRG) method. The results have shown that the anisotropy suppresses the quantum correlations while the DM interaction can enhance them. Meanwhile, using the QRG method we have studied the quantum phase transition of GQD and obtained two saturated values, which are associated with two different phases: spin-fluid phase and the Néel phase. It is worth mentioning that the block–block correlation is not strong enough to violate the Bell–CHSH inequality in the whole iteration steps. Moreover, the nonanalytic phenomenon and scaling behavior of Bell inequality are discussed in detail. As a byproduct, the conjecture that the exact lower and upper bounds of Bell inequality versus GQD can always be established for this spin system although the given density matrix is a general X state.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

Molecular dynamics-based multiscale damage initiation model for CNT/epoxy nanopolymers

This study presents a methodology that accurately simulates the brittle behavior of epoxy polymers initiating at the molecular level due to bond elongation and subsequent bond dissociation. The system investigated in this study comprises a combination of crystalline carbon nanotubes (CNTs) dispersed in epoxy polymer molecules. Molecular dynamics (MD) simulations are performed with an appropriate bond order-based force field to capture deformation-induced bond dissociation between atoms within the simulation volume. During deformation, the thermal vibration of molecules causes the elongated bonds to re-equilibrate; thus, the effect of mechanical deformation on bond elongation and scission cannot be captured effectively. This issue is overcome by deforming the simulation volume at zero temperature—a technique adopted from the concept of quasi-continuum and demonstrated successfully in the authors’ previous work. Results showed that a combination of MD deformation tests with ultra-high strain rates at near-zero temperatures provides a computationally efficient alternative for the study of bond dissociation phenomenon in amorphous epoxy polymer. In this paper, the ultra-high strain rate deformation approach is extended to the CNT-epoxy system at various CNT weight fractions and the corresponding bond disassociation energy extracted from the simulation volume is used as input to a low-fidelity continuum damage mechanics (CDM) model to demonstrate the bridging of length scales and to study matrix failure at the microscale. The material parameters for the classical CDM model are directly obtained from physics-based atomistic simulations, thus improving the accuracy of the multiscale approach.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Accelerating population balance-Monte Carlo simulation for coagulation dynamics from the Markov jump model, stochastic algorithm and GPU parallel computing

This paper proposes a comprehensive framework for accelerating population balance-Monte Carlo (PBMC) simulation of particle coagulation dynamics. By combining Markov jump model, weighted majorant kernel and GPU (graphics processing unit) parallel computing, a significant gain in computational efficiency is achieved. The Markov jump model constructs a coagulation-rule matrix of differentially-weighted simulation particles, so as to capture the time evolution of particle size distribution with low statistical noise over the full size range and as far as possible to reduce the number of time loopings. Here three coagulation rules are highlighted and it is found that constructing appropriate coagulation rule provides a route to attain the compromise between accuracy and cost of PBMC methods. Further, in order to avoid double looping over all simulation particles when considering the two-particle events (typically, particle coagulation), the weighted majorant kernel is introduced to estimate the maximum coagulation rates being used for acceptance–rejection processes by single-looping over all particles, and meanwhile the mean time-step of coagulation event is estimated by summing the coagulation kernels of rejected and accepted particle pairs. The computational load of these fast differentially-weighted PBMC simulations (based on the Markov jump model) is reduced greatly to be proportional to the number of simulation particles in a zero-dimensional system (single cell). Finally, for a spatially inhomogeneous multi-dimensional (multi-cell) simulation, the proposed fast PBMC is performed in each cell, and multiple cells are parallel processed by multi-cores on a GPU that can implement the massively threaded data-parallel tasks to obtain remarkable speedup ratio (comparing with CPU computation, the speedup ratio of GPU parallel computing is as high as 200 in a case of 100 cells with 10 000 simulation particles per cell). These accelerating approaches of PBMC are demonstrated in a physically realistic Brownian coagulation case. The computational accuracy is validated with benchmark solution of discrete-sectional method. The simulation results show that the comprehensive approach can attain very favorable improvement in cost without sacrificing computational accuracy.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

A cavitation and dynamic void growth model for a general class of strain-softening amorphous materials

In this work, we derive a homogenized framework for cavitation and dynamic void growth in a general class of strain-softening materials with particular emphasis on amorphous materials. Analytic solutions are derived by considering an assemblage of spherical shells subject to dynamic hydrostatic tensile pressure. The isochoric response of the matrix material is captured by a free volume theory based amorphous viscoplasticity model. The framework accounts for the coupled rate-dependent, strain-softening behavior typically exhibited in amorphous materials. Validation of the homogenized theory is carried out against numerous direct numerical simulations with excellent agreement. Furthermore, we find that a reduction in cavitation strength and significant acceleration of dynamic void growth rates is induced as a consequence of strain-softening in the amorphous matrix material. Lastly, the model is utilized to interpret some non-intuitive experimental observations of amorphous materials subject to shock compression and subsequent spall failure.

36 MATERIALS SCIENCE↗
DOE OSTI2024

CFT and lattice correlators near an RG domain wall between minimal models

Conformal interfaces separating two conformal field theories (CFTs) provide maps between different CFTs, and naturally exist in nature as domain walls between different phases. One particularly interesting construction of a conformal interface is the renormalization group (RG) domain wall between CFTs. For a given Virasoro minimal model \mathcal{M}_{k+3,k+2} ℳ k + 3 , k + 2 , an RG domain wall can be generated by a specific deformation which triggers an RG flow towards its adjacent Virasoro minimal model \mathcal{M}_{k+2,k+1} ℳ k + 2 , k + 1 with the deformation turned on over part of the space. An algebraic construction of this domain wall was proposed by Gaiotto in [J. High Energy Phys. 12, 103 (2012)]. In this paper, we will provide a study of this RG domain wall for the minimal case k=2 k = 2 , which can be thought of as a nonperturbative check of the construction. In this case the wall is separating the Tricritical Ising Model (TIM) CFT and the Ising Model (IM) CFT. We will check the analytical results of correlation functions from the RG brane construction with the numerical density matrix renormalization group (DMRG) calculation using a lattice model proposed in [arXiv.1206.1332, Science 344, 280 (2014)], and find a perfect agreement. We comment on possible experimental realizations of this RG domain wall.

Cogburn, Cameron V.↗
DOE OSTI2021

Random circuit block-encoded matrix and a proposal of quantum LINPACK benchmark

The LINPACK benchmark reports the performance of a computer for solving a system of linear equations with dense random matrices. Although this task was not designed with a real application directly in mind, the LINPACK benchmark has been used to define the list of TOP500 supercomputers since the debut of the list in 1993. We propose that a similar benchmark, called the quantum LINPACK benchmark, could be used to measure the whole machine performance of quantum computers. The success of the quantum LINPACK benchmark should be viewed as the minimal requirement for a quantum computer to perform a useful task of solving linear algebra problems, such as linear systems of equations. We propose an input model called the Random Circuit Block-Encoded Matrix (RACBEM), which is a proper generalization of a dense random matrix in the quantum setting. The RACBEM model is efficient to be implemented on a quantum computer and can be designed to optimally adapt to any given quantum architecture, with relying on a black-box quantum compiler. Besides solving linear systems, the RACBEM model can be used to perform a variety of linear algebra tasks relevant to many physical applications, such as computing spectral measures, time series generated by a Hamiltonian simulation, and thermal averages of the energy. We implement these linear algebra operations on IBM Q quantum devices as well as quantum virtual machines, and demonstrate their performance in solving scientific computing problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Development of Nepheline Models for Hanford HLW Glass

The work included the design, fabrication, and characterization of a matrix of 35 new highAl HLW glasses and assessment and development of new models for nepheline formation.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2017

Simulating cosmologies beyond ΛCDM with PINOCCHIO

We present a method that extends the capabilities of the PINpointing Orbit-Crossing Collapsed HIerarchical Objects (PINOCCHIO) code, allowing it to generate accurate dark matter halo mock catalogues in cosmological models where the linear growth factor and the growth rate depend on scale. Such cosmologies comprise, among others, models with massive neutrinos and some classes of modified gravity theories. We validate the code by comparing the halo properties from PINOCCHIO against N-body simulations, focusing on cosmologies with massive neutrinos: νΛCDM. We analyse the halo mass function, halo two-point correlation function and halo power spectrum, showing that PINOCCHIO reproduces the results from simulations with the same level of precision as the original code (∼ 5–10%). We demonstrate that the abundance of halos in cosmologies with massless and massive neutrinos from PINOCCHIO matches very well the outcome of simulations, and point out that PINOCCHIO can reproduce the Ω{sub ν}–σ{sub 8} degeneracy that affects the halo mass function. We finally show that the clustering properties of the halos from PINOCCHIO matches accurately those from simulations both in real and redshift-space, in the latter case up to k = 0.3 h Mpc{sup −1}. We emphasize that the computational time required by PINOCCHIO to generate mock halo catalogues is orders of magnitude lower than the one needed for N-body simulations. This makes this tool ideal for applications like covariance matrix studies within the standard ΛCDM model but also in cosmologies with massive neutrinos or some modified gravity theories.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Pair Density Wave in the Doped t - J Model with Ring Exchange on a Triangular Lattice

In our previous study, we found a quantum spin liquid phase with a spinon Fermi surface in the two dimensional spin-1/2 Heisenberg model with four-spin ring exchange on a triangular lattice. In this work we dope the spinon Fermi surface phase by studying the $t - J$ model with four-spin ring exchange. We perform density matrix renormalization group calculations on four-leg cylinders of a triangular lattice and find that the dominant pair correlation function is that of a pair density wave; i.e., it is oscillatory while decaying with distance with a power law. The doping dependence of the period is studied. This is the first example where a pair density wave is the dominant pairing in a generic strongly interacting system where the pair density wave cannot be explained as a composite order and no special symmetry is required.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Modification of transparent materials with ultrashort laser pulses: What is energetically and mechanically meaningful?

A comprehensive analysis of laser-induced modification of bulk glass by single ultrashort laser pulses is presented which is based on combination of optical Maxwell-based modeling with thermoelastoplastic simulations of post-irradiation behavior of matter. A controversial question on free electron density generated inside bulk glass by ultrashort laser pulses in modification regimes is addressed on energy balance grounds. Spatiotemporal dynamics of laser beam propagation in fused silica have been elucidated for the regimes used for direct laser writing in bulk glass. 3D thermoelastoplastic modeling of material relocation dynamics under laser-induced stresses has been performed up to the microsecond timescale when all motions in the material decay. The final modification structure is found to be imprinted into material matrix already at sub-nanosecond timescale. Modeling results agree well with available experimental data on laser light transmission through the sample and the final modification structure.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Quantum Phases of Transition Metal Dichalcogenide Moiré Systems

Moiré systems provide a rich platform for studies of strong correlation physics. Recent experiments on heterobilayer transition metal dichalcogenide Moiré systems are exciting in that they manifest a relatively simple model system of an extended Hubbard model on a triangular lattice. Inspired by the prospect of the hetero-transition metal dichalcogenide Moiré system’s potential as a solid-state-based quantum simulator, we explore the extended Hubbard model on the triangular lattice using the density matrix renormalization group. Specifically, we explore the two-dimensional phase space spanned by the key tuning parameters in the extended Hubbard model, namely, the kinetic energy strength and the further-range Coulomb interaction strengths. We find competition between Fermi fluid, chiral spin liquid, spin density wave, and charge order. In particular, our finding of the optimal further-range interaction for the chiral correlation presents a tantalizing possibility.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Skyrmion superconductivity: DMRG evidence for a topological route to superconductivity

It was recently suggested that the topology of magic-angle twisted bilayer graphene's (MATBG) flat bands could provide a novel mechanism for superconductivity distinct from both weakly coupled BCS theory and the d-wave phenomenology of the high-T c cuprates. In this work, we examine this possibility using a density matrix renormalization group (DMRG) study of a model which captures the essential features of MATBG's symmetry and topology. Here, using large-scale cylinder-DMRG calculations to obtain the ground state and its excitations as a function of the electron doping, we find clear evidence for superconductivity driven by the binding of electrons into charge-2e skyrmions. Remarkably, this binding is observed even in the regime where the unscreened Coulomb repulsion is by far the largest energy scale, demonstrating the robustness of this topological, all-electronic pairing mechanism.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗