Search NASA⌕ Search

SEARCH · Search NASA

Results for “COMPUTING PROBLEMS”

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 1,423 records · Page 79

Learning energy-based representations of quantum many-body states

Efficient representation of quantum many-body states on classical computers is a problem of practical importance. An ideal representation of a quantum state combines a succinct characterization informed by the structure and symmetries of the system along with the ability to predict the physical observables of interest. Several machine-learning approaches have been recently used to construct such classical representations, which enable predictions of observables and account for physical symmetries. However, the structure of a quantum state typically gets lost unless a specialized is employed based on prior knowledge of the system. Moreover, most such approaches give no information about what states are easier to learn in comparison with others. Here, we propose a generative energy-based representation of quantum many-body states derived from Gibbs distributions used for modeling the thermal states of classical spin systems. Based on the prior information on a family of quantum states, the energy function can be specified by a small number of parameters using an explicit low-degree polynomial or a generic parametric family such as neural nets and can naturally include the known symmetries of the system. Our results show that such a representation can be efficiently learned from data using exact algorithms in a form that enables the prediction of expectation values of physical observables. Importantly, the structure of the learned energy function provides a natural explanation for the difficulty of learning an energy-based representation of a given class of quantum states when measured in a certain basis. Published by the American Physical Society 2024

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A general solution to the hidden-line problem

The requirements for computer-generated perspective projections of three dimensional objects has escalated. A general solution was developed. The theoretical solution to this problem is presented. The method is very efficient as it minimizes the selection of points and comparison of line segments and hence avoids the devastation of square-law growth.

Hedgley, D. R., Jr.↗

Typical motions in multiple systems

In very old times, people counted - one, two, many. The author wants to show that they were right. Consider the motions of isolated bodies: (1) N = 1 - simple motion; (2) N = 2 - Keplerian orbits; and (3) N = 3 - this is the difficult problem. In general, this problem can be studied only by computer simulations. The author studied this problem over many years (see, e.g., Agekian and Anosova, 1967; Anosova, 1986, 1989 a,b). The principal result is that two basic types of dynamics take place in triple systems. The first special type is the stable hierarchical systems with two almost Keplerian orbits. The second general type is the unstable triple systems with complicated motions of the bodies. By random choice of the initial conditions, by the Monte-Carlo method, the stable systems comprised about approx. 10% of the examined cases; the unstable systems comprised the other approx. 90% of cases under consideration. In N greater than 3, the studies of dynamics of such systems by computer simulations show that we have in general also the motions roughly as at the cases 1 - 3 with the relative negative or positive energies of the bodies. In the author's picture, the typical trajectories of the bodies in unstable triple systems of the general type of dynamics are seen. Such systems are disrupted always after close triple approaches of the bodies. These approaches play a role like the gravitational slingshot. Often, the velocities of escapers are very large. On the other hand, the movie also shows the dynamical processes of a formation, dynamical evolution and disruption of the temporary wide binaries in triples and a formation of final hard massive binaries in the final evolution of triples.

Anosova, Joanna P.↗

Accelerating Bilevel Optimization With Hierarchical Many-Threaded Parallel Differential Evolution

Bilevel optimization is encountered in many relevant real-world applications. The main feature of this type of problem is that an upper-level optimization problem is constrained by a nested lower-level optimization problem. Because of this nested structure, bilevel problems (BLPs) are usually computationally expensive to solve. Differential evolution (DE) has demonstrated promising results in solving BLPs of relatively small scales. As the problem scale increases, the decision space becomes intrinsically larger, requiring a growing number of function evaluations for the method to work properly. In this context, heavy parallelization and high-performance computing techniques are indispensable to enable the resolution of more complex and challenging optimization problems. Hence, we propose a hierarchical many-threaded parallel DE approach for BLPs, where both levels are parallelized. The computational experiments demonstrate that the parallel implementation achieved runtime speeds ranging from 44 to 2559 times faster than the sequential version on a well-known scalable SMD benchmark test problem when executed on an NVIDIA A100 GPU. The findings indicate that the algorithm’s convergence is strongly influenced by the number of both upper- and lower-level generations. Moreover, the success of experiments with large-scale problems is closely linked to the choice of small population sizes.

Dufek, Amanda S↗

A Computational Intelligence (CI) Approach to the Precision Mars Lander Problem

A Mars precision landing requires a landed footprint of no more than 100 meters. Obstacles to reducing the landed footprint include trajectory dispersions due to initial atmospheric entry conditions such as entry angle, parachute deployment height, environment parameters such as wind, atmospheric density, parachute deployment dynamics, unavoidable injection error or propagated error from launch, etc. Computational Intelligence (CI) techniques such as Artificial Neural Nets and Particle Swarm Optimization have been shown to have great success with other control problems. The research period extended previous work on investigating applicability of the computational intelligent approaches. The focus of this investigation was on Particle Swarm Optimization and basic Neural Net architectures. The research investigating these issues was performed for the grant cycle from 5/15/01 to 5/15/02. Matlab 5.1 and 6.0 along with NASA's POST were the primary computational tools.

Birge, Brian↗

Accurate computation and continuation of homoclinic and heteroclinic orbits for singular perturbation problems

In earlier papers, Doedel and the authors have developed a numerical method and derived error estimates for the computation of branches of heteroclinic orbits for a system of autonomous ordinary differential equations in R(exp n). The idea of the method is to reduce a boundary value problem on the real line to a boundary value problem on a finite interval by using a local (linear or higher order) approximation of the stable and unstable manifolds. A practical limitation for the computation of homoclinic and heteroclinic orbits has been the difficulty in obtaining starting orbits. Typically these were obtained from a closed form solution or via a homotopy from a known solution. Here we consider extensions of our algorithm which allow us to obtain starting orbits on the continuation branch in a more systematic way as well as make the continuation algorithm more flexible. In applications, we use the continuation software package AUTO in combination with some initial value software. The examples considered include computation of homoclinic orbits in a singular perturbation problem and in a turbulent fluid boundary layer in the wall region problem.

Vaughan, William W.↗

An Incremental Gradient Method for Optimization Problems With Variational Inequality Constraints

We consider minimizing a sum of agent-specific nondifferentiable merely convex functions over the solution set of a variational inequality (VI) problem in that each agent is associated with a local monotone mapping. This problem finds an application in computation of the best equilibrium in nonlinear complementarity problems arising in transportation networks. We develop an iteratively regularized incremental gradient method where at each iteration, agents communicate over a directed cycle graph to update their solution iterates using their local information about the objective and the mapping. The proposed method is single-timescale in the sense that it does not involve any excessive hard-to-project computation per iteration. We derive nonasymptotic agent-wise convergence rates for the suboptimality of the global objective function and infeasibility of the VI constraints measured by a suitably defined dual gap function. Finally, the proposed method appears to be the first fully iterative scheme equipped with iteration complexity that can address distributed optimization problems with VI constraints over cycle graphs.

convergence↗

French Meteor Network for High Precision Orbits of Meteoroids

There is a lack of precise meteoroids orbit from video observations as most of the meteor stations use off-the-shelf CCD cameras. Few meteoroids orbit with precise semi-major axis are available using film photographic method. Precise orbits are necessary to compute the dust flux in the Earth s vicinity, and to estimate the ejection time of the meteoroids accurately by comparing them with the theoretical evolution model. We investigate the use of large CCD sensors to observe multi-station meteors and to compute precise orbit of these meteoroids. An ideal spatial and temporal resolution to get an accuracy to those similar of photographic plates are discussed. Various problems faced due to the use of large CCD, such as increasing the spatial and the temporal resolution at the same time and computational problems in finding the meteor position are illustrated.

Atreya, P.↗

High-resolution methods for preserving the sum of mass fractions: improved χ-scheme and an alternative

When high resolution convection schemes are used for discretizing chemical species mass balance equations, the mass fractions are not guaranteed to add to one. We show that a proposed remedy called χ-scheme (Darwish and Moukalled, Comput.Methods Appl.Mech. Engrg. 192 (2003): 1711) will degrade to a diffusive first-order scheme when a chemical species vanishes from the mixture, for example, because of chemical reactions. We propose an improvement to the χ-scheme to overcome this problem. Furthermore, a computationally efficient alternative scheme is proposed and evaluated with several examples, to quantify the improvements in the accuracy and the computational time.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A dual-mesh method for efficient thermal stress analysis of large-scale welded structures

Transient thermo-mechanical analysis of welding problem requires tremendous computation cost. To accelerate the thermal analysis of large-scale welded structures, an efficient computation scheme based on heat transfer localization and dual meshes was proposed. The computation accuracy is guaranteed by a local fine mesh model with size determined by a theoretical solution and a global coarse mesh model with equivalent heat input. The validity and accuracy of the dual-mesh method were verified using an experimental bead-on-plate model. By extending the weld length, the computation time of the proposed method was proved to be almost linearly dependent on the model scale. The thermal analysis of fillet welding of a large panel structure with 6-m-long weld was accelerated by 10 times over conventional finite element analysis and 2.2 times over adaptive mesh method. Meanwhile, the physical memory consumption was also greatly reduced by the dual-mesh method. Such efficient computation method enables fast evaluation of welding stress and distortion which are vital for manufacturing process and structure performance.

36 MATERIALS SCIENCE↗

Energy dissipation via acoustic emission in ductile crack initiation

In this article, we present a modeling approach to estimate the energy release due to ductile crack initiation in conjunction to the energy dissipation associated with the formation and propagation of transient stress waves typically referred to as acoustic emission. To achieve this goal, a ductile fracture problem is investigated computationally using the finite element method based on a compact tension geometry under Mode I loading conditions. To quantify the energy dissipation associated with acoustic emission, a crack increment is produced given a pre-determined notch size in a 3D cohesive-based extended finite element model. The computational modeling methodology consists of defining a damage initiation state from static simulations and linking such state to a dynamic formulation used to evaluate wave propagation and related energy redistribution effects. The model relies on a custom traction separation law constructed using full field deformation measurements obtained experimentally using the digital image correlation method. The amount of energy release due to the investigated first crack increment is evaluated through three different approaches both for verification purposes and to produce an estimate of the portion of the energy that radiates away from the crack source in the form of transient waves. The results presented herein propose an upper bound for the energy dissipation associated to acoustic emission, which could assist the interpretation and implementation of relevant nondestructive evaluation methods and the further enrichment of the understanding of effects associated with fracture.

42 ENGINEERING↗

Reduced-order modeling of near-field THMC coupled processes for nuclear waste repositories in shale

Performance assessment (PA) of geologic radioactive waste repositories requires three-dimensional simulation of highly nonlinear, thermo-hydro-mechanical-chemical (THMC), multiphase flow and transport processes across many kilometers and over tens to hundreds of thousands of years. Integrating the effects of a near-field geomechanical process (i.e. buffer swelling) into coupled THC simulations through reduced-order modeling, rather than through fully coupled geomechanics, can reduce the dimensionality of the problem and improve computational efficiency. In this study, PFLOTRAN simulations model a single waste package in a shale host rock repository, where re-saturation of a bentonite buffer causes the buffer to swell and exert stress on a highly fractured disturbed rock zone (DRZ). Three types of stress-dependent permeability functions (exponential, modified cubic, and Two-part Hooke’s law models) are implemented to describe mechanical characteristics of the system. Overall, our modeling study suggests that compressing fractures reduces DRZ permeability, which could influence the rate of radionuclide transport and exchange with corrosive species in host rock groundwater that could accelerate waste package degradation. Less permeable shale host rock delays buffer swelling, consequently retarding DRZ permeability reduction as well as chemical transport within the barrier system.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A derivation and scalable implementation of the synchronous parallel kinetic Monte Carlo method for simulating long-time dynamics

Kinetic Monte Carlo (KMC) simulations are used to study long-time dynamics of a wide variety of systems. Unfortunately, the conventional KMC algorithm is not scalable to larger systems, since its time scale is inversely proportional to the simulated system size. A promising approach to resolving this issue is the synchronous parallel KMC (SPKMC) algorithm, which makes the time scale size-independent. This paper introduces a formal derivation of the SPKMC algorithm based on local transition-state and time-dependent Hartree approximations, as well as its scalable parallel implementation based on a dual linked-list cell method. The resulting algorithm has achieved a weak-scaling parallel efficiency of 0.935 on 1024 Intel Xeon processors for simulating biological electron transfer dynamics in a 4.2 billion-heme system, as well as decent strong-scaling parallel efficiency. The parallel code has been used to simulate a lattice of cytochrome complexes on a bacterial-membrane nanowire, and it is broadly applicable to other problems such as computational synthesis of new materials.

97 MATHEMATICS AND COMPUTING↗

A Physics-Constrained Deep Learning Model for Simulating Multiphase Flow in 3D Heterogeneous Porous Media

Physics-based simulators for multiphase flow in porous media emulate nonlinear processes with coupled physics, and usually require extensive computational resources for software development, maintenance and simulation execution. As a result, a huge demand exists for fast modeling of coupled processes in a wide range of subsurface applications including geological sequestration, hydrocarbon recovery and geothermal energy extraction. In this work, an efficient physics-constrained deep learning model is developed for solving multiphase flow in 3-Dimensional (3D) heterogeneous porous media. The model fully leverages the spatial topology predictive capability of convolutional neural networks, specifically U-Net with successive contracting and expansive steps, and is coupled with an efficient continuity-based smoother to predict flow responses that need spatial continuity. Furthermore, the transient regions are penalized to steer the training process such that the model can accurately capture flow in these regions. The model takes inputs including properties of porous media, fluid properties and well controls, and predicts the temporal-spatial evolution of the state variables (pressure and saturation). While maintaining the continuity of fluid flow, the 3D spatial domain is decomposed into 2D images for reducing training cost, and the decomposition results in an increased number of training data samples and better training efficiency. Additionally, a surrogate model is separately constructed as a postprocessor to calculate well flow rate based on the predictions of state variables from the deep learning model. We use the example of CO 2 injection into saline aquifers, and apply the physics-constrained deep learning model that is trained from physics-based simulation data and emulates the physics process. The model performs prediction with a speedup of ~ 1400 times compared to physics-based simulations, and the average temporal errors of predicted pressure and saturation plumes are 0.27% and 0.099% respectively. Furthermore, water production rate is efficiently predicted by a surrogate model for well flow rate, with a mean error less than 5%. Therefore, with its unique scheme to cope with the fidelity in fluid flow in porous media, the physics-constrained deep learning model can become an efficient predictive model for computationally demanding inverse problems or other coupled processes.

58 GEOSCIENCES↗

Nonlinear, real-time optimization for actuator management in tokamaks

Experiments in DIII-D have been carried out to test a novel actuator management approach in tokamaks. Here, the actuator management scheme is posed as a nonlinear-optimization problem in which the actuator commands are calculated in real time according to the changing control priorities, plasma state, and actuator availability. Such optimization problem is solved using the augmented Lagrangian method, combined with a gradient projection method and a conjugate-gradient iteration algorithm. The algorithmic approach followed in this work does not depend on the particular control objectives or actuators considered, which facilitates its integration with other independently-designed control components within a plasma-control system. In addition, the actuator-management algorithm is able to handle the optimization problem in a computationally efficient manner, making it suitable for real-time implementations. Initial DIII-D results in the steady-state high-q min scenario have demonstrated the capabilities of the actuator manager to perform both simultaneous multiple mission and repurposing sharing, which will be required in ITER.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Chemical thermodynamic data. 1. The concept of links to the chemical elements and the historical development of key thermodynamic data [plus Supplementary Electronic Annex 2]

Chemical thermodynamic data remain a keystone for geochemical modeling and reactive transport simulation as applied to an increasing number of applications in the earth sciences, as well as applications in other areas including metallurgy, material science, and industrial process design. The last century has seen the development of a large body of thermodynamic data and a number of major compilations. The past several decades have seen the development of thermodynamic databases in digital form designed to support computer calculations. However, problems with thermodynamic data appear to be persistent. One problem pertains to the use of inconsistent primary key reference data. Such data pertain to elemental reference forms and key, stoichiometrically simple chemical species including metal oxides, CO 2 , water, and aqueous species such as Na + and Cl - . A consistent set of primary key data (standard Gibbs energies, standard enthalpies, and standard entropies for key chemical species) for 298.15K and 1 bar pressure is essential. Thermochemical convention is to define the standard Gibbs energy and the standard enthalpy of an individual chemical species in terms of formation from reference forms of the constituent chemical elements. We propose a formal concept of “links” to the elemental reference forms. This concept involves a documented understanding of all reactions and calculations leading to values for a formation property (standard Gibbs energy or enthalpy). A valid link consists of two parts: (a) the path of reactions and corrections and (b) the associated data, which are key data. Such a link differs from a bare “key” or “reference” datum in that it requires additional information. Some or all of its associated data may also be key data. In evaluating a reported thermodynamic datum, one should identify the links to the chemical elements, a process which can be time-consuming and which may lead to a dead end (an incomplete link). The use of two or more inconsistent links to the same elemental reference form in a thermodynamic database will result in an inconsistency in the database. Thus, in constructing a database, it is important to establish a set of reliable links (generally resulting in a set of primary reference data) and then correct all data adopted subsequently for consistency with that set. Recommended values of key data have not been constant through history. We review some of this history through the lens of major compilations and other influential reports, and note a number of problem areas. Finally, we illustrate the concepts developed in this paper by applying them to some key species of geochemical interest, including liquid water; quartz and aqueous silica; and gibbsite, corundum, and the aqueous aluminum ion.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Residual Monte Carlo high-order solver for Moment-Based Accelerated Thermal Radiative Transfer equations

In this article we explore the possibility of replacing Standard Monte Carlo (SMC) transport sweeps within a Moment-Based Accelerated Thermal Radiative Transfer (TRT) algorithm with a Residual Monte Carlo (RMC) formulation. Previous Moment-Based Accelerated TRT implementations have encountered trouble when stochastic noise from SMC transport sweeps accumulates over several iterations and pollutes the low-order system. With RMC we hope to significantly lower the build-up of statistical error at a much lower cost. First, we display encouraging results for a zero-dimensional test problem. Then, we demonstrate that we can achieve a lower degree of error in two one-dimensional test problems by employing an RMC transport sweep with multiple orders of magnitude fewer particles per sweep. We find that by reformulating the high-order problem, we can compute more accurate solutions at a fraction of the cost.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗