Search NASA⌕ Search

SEARCH · Search NASA

Results for “discrete optimization”

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 1,495 records · Page 83

Improving reproducibility in synchrotron tomography using implementation-adapted filters

For reconstructing large tomographic datasets fast, filtered backprojection-type or Fourier-based algorithms are still the method of choice, as they have been for decades. These robust and computationally efficient algorithms have been integrated in a broad range of software packages. The continuous mathematical formulas used for image reconstruction in such algorithms are unambiguous. However, variations in discretization and interpolation result in quantitative differences between reconstructed images, and corresponding segmentations, obtained from different software. This hinders reproducibility of experimental results, making it difficult to ensure that results and conclusions from experiments can be reproduced at different facilities or using different software. In this paper, a way to reduce such differences by optimizing the filter used in analytical algorithms is proposed. These filters can be computed using a wrapper routine around a black-box implementation of a reconstruction algorithm, and lead to quantitatively similar reconstructions. Use cases for this approach are demonstrated by computing implementation-adapted filters for several open-source implementations and applying them to simulated phantoms and real-world data acquired at the synchrotron. Our contribution to a reproducible reconstruction step forms a building block towards a fully reproducible synchrotron tomography data processing pipeline.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

High-Order Mesh r-Adaptivity with Tangential Relaxation and Guaranteed Mesh Validity

High-order meshes are crucial for achieving optimal convergence rates in curvilinear domains, preserving symmetry, and aligning with key flow features in moving mesh simulations [1], but their quality is challenging to control. In prior work, we have developed techniques based on Target-Matrix Optimization Paradigm (TMOP) to adapt a given high-order mesh to the geometry and solution of the partial differential equation (PDE) [2, 3]. Here, we extend this framework to address two key gaps in the literature for highorder mesh 𝑟-adaptivity. First, we introduce tangential relaxation on curved surfaces using solely the discrete mesh representation, eliminating the need for access to underlying geometry (e.g., CAD model). Second, we ensure a continuously positive Jacobian determinant throughout the domain. This determinant positivity is essential for using the high-order mesh resulting from 𝑟-adaptivity with arbitrary quadrature schemes in simulations. The proposed approach is demonstrated to be robust using a variety of numerical experiments.

Mathematics and Computing↗

Control optimization for parametric Hamiltonians by pulse reconstruction

Optimal control techniques provide a means to tailor the control pulses required to generate customized quantum gates, which helps to improve the resilience of quantum simulations to gate errors and device noise. However, the significant amount of (classical) computation required to generate customized gates can quickly undermine the effectiveness of this approach, especially when pulse optimization needs to be iterated. We propose a method to reduce the computational time required to generate the control pulse for a Hamiltonian that is parametrically dependent on a time-varying quantity. We use simple interpolation schemes to accurately reconstruct the control pulses from a set of pulses obtained in advance for a discrete set of predetermined parameter values. We obtain a reconstruction with very high fidelity and a significant reduction in computational effort. We report the results of the application of the proposed method to device-level quantum simulations of the unitary (real) time evolution of two interacting neutrons based on superconducting qubits.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Automatic Differentiation of C++ Codes on Emerging Manycore Architectures with Sacado

Automatic differentiation (AD) is a well-known technique for evaluating analytic derivatives of calculations implemented on a computer, with numerous software tools available for incorporating AD technology into complex applications. However, a growing challenge for AD is the efficient differentiation of parallel computations implemented on emerging manycore computing architectures such as multicore CPUs, GPUs, and accelerators as these devices become more pervasive. In this work, we explore forward mode, operator overloading-based differentiation of C++ codes on these architectures using the widely available Sacado AD software package. In particular, we leverage Kokkos, a C++ tool providing APIs for implementing parallel computations that is portable to a wide variety of emerging architectures. Here we describe the challenges that arise when differentiating code for these architectures using Kokkos, and two approaches for overcoming them that ensure optimal memory access patterns as well as expose additional dimensions of fine-grained parallelism in the derivative calculation. We describe the results of several computational experiments that demonstrate the performance of the approach on a few contemporary CPU and GPU architectures. We then conclude with applications of these techniques to the simulation of discretized systems of partial differential equations.

97 MATHEMATICS AND COMPUTING↗

Generic Discretization Library

The GenDiL library is a collection of C++ software abstractions designed to discretize and solve partial differential equations (PDEs) for high-performance computing (HPC) applications. Its primary focus is on modern C++ generic programming, which helps ensure portability across various hardware architectures. The central idea behind the library is to provide building blocks for numerical algorithms-such as discretization methods and iteration patterns-so that domain experts can focus on the math, rather than the low-level details of hardware or implementation. By defining abstractions for data types, iteration over computational grids, and scheduling of operations, the library isolates the high-level PDE algorithms from the platform-specific optimizations needed to achieve efficient performance.

Dudouit, Yohann [Lawrence Livermore National Labor↗

Discrete sequence prediction and its applications

Learning from experience to predict sequences of discrete symbols is a fundamental problem in machine learning with many applications. We apply sequence prediction using a simple and practical sequence-prediction algorithm, called TDAG. The TDAG algorithm is first tested by comparing its performance with some common data compression algorithms. Then it is adapted to the detailed requirements of dynamic program optimization, with excellent results.

Laird, Philip↗

Individual differences in strategic flight management and scheduling

A group of 30 instrument-rated pilots was made to fly simulator approaches to three airports under conditions of low, medium, and high workload conditions. An analysis is presently conducted of the difference in discrete task scheduling between the group of 10 highest and 10 lowest performing pilots in the sample; this categorization was based on the mean of various flight-profile measures. The two groups were found to differ from each other only in terms of the time when specific events were conducted, and of the optimality of scheduling for certain high-priority tasks. These results are assessed in view of the relative independence of task-management skills from aircraft-control skills.

Wickens, Christopher D.↗

The suitability of selected multidisciplinary design and optimization techniques to conceptual aerospace vehicle design

Four methods for preliminary aerospace vehicle design are reviewed. The first three methods (classical optimization, system decomposition, and system sensitivity analysis (SSA)) employ numerical optimization techniques and numerical gradients to feed back changes in the design variables. The optimum solution is determined by stepping through a series of designs toward a final solution. Of these three, SSA is argued to be the most applicable to a large-scale highly coupled vehicle design where an accurate minimum of an objective function is required. With SSA, several tasks can be performed in parallel. The techniques of classical optimization and decomposition can be included in SSA, resulting in a very powerful design method. The Taguchi method is more of a 'smart' parametric design method that analyzes variable trends and interactions over designer specified ranges with a minimum of experimental analysis runs. Its advantages are its relative ease of use, ability to handle discrete variables, and ability to characterize the entire design space with a minimum of analysis runs.

Olds, John R.↗

Stochastic search in structural optimization - Genetic algorithms and simulated annealing

An account is given of illustrative applications of genetic algorithms and simulated annealing methods in structural optimization. The advantages of such stochastic search methods over traditional mathematical programming strategies are emphasized; it is noted that these methods offer a significantly higher probability of locating the global optimum in a multimodal design space. Both genetic-search and simulated annealing can be effectively used in problems with a mix of continuous, discrete, and integer design variables.

Hajela, Prabhat↗

Micellar Mimicry of Intermetallic C14 and C15 Laves Phases by Aqueous Lyotropic Self-Assembly

Concentration-dependent supramolecular self-assembly of amphiphilic molecules in water furnishes a variety of nanostructured lyotropic liquid crystals (LLCs), which typically display high symmetry bicontinuous network and discontinuous micellar morphologies. Aqueous dispersions of soft spherical micelles derived from small molecule amphiphile hydration typically pack into exemplary body-centered cubic and closest-packed LLCs. However, investigations of hydrated mixtures of the ionic surfactant tetramethylammonium decanoate loaded with 40 wt % n-decane (TMADec-40) revealed the formation of a high symmetry bicontinuous double diamond LLC, as well as cubic C15 and hexagonal C14 Laves LLC phases that mirror the MgCu 2 and MgZn 2 intermetallic structure types, respectively. Detailed small-angle X-ray scattering analyses demonstrate that the complex C15 and C14 LLCs exhibit large unit cells, in which 12 or more ~3–4 nm diameter micelles of multiple discrete sizes arrange into tetrahedral close packing arrangements with exceptional long-range translational order. The symmetry breaking that drives self-assembly into these low-symmetry LLC phases is rationalized in terms of a frustrated balance between maximizing counterion-mediated micellar cohesion within the ensemble of oil-swollen particles, while simultaneously optimizing local spherical particle symmetry to minimize molecular-level variations in surfactant solvation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hypothesis tests on Rayleigh wave radiation pattern shapes: a theoretical assessment of idealized source screening

SUMMARY Shallow seismic sources excite Rayleigh wave ground motion with azimuthally dependent radiation patterns. We place binary hypothesis tests on theoretical models of such radiation patterns to screen cylindrically symmetric sources (like explosions) from non-symmetric sources (like non-vertical dip-slip or non-VDS faults). These models for data include sources with several unknown parameters, contaminated by Gaussian noise and embedded in a layered half-space. The generalized maximum likelihood ratio tests that we derive from these data models produce screening statistics and decision rules that depend on measured, noisy ground motion at discrete sensor locations. We explicitly quantify how the screening power of these statistics increase with the size of any dip-slip and strike-slip components of the source, relative to noise (faulting signal strength) and how they vary with network geometry. As applications of our theory, we apply these tests to (1) find optimal sensor locations that maximize the probability of screening non-circular radiation patterns and (2) invert for the largest non-VDS faulting signal that could be mistakenly attributed to an explosion with damage, at a particular attribution probability. Finally, we quantify how certain errors that are sourced by opening cracks increase screening rate errors. While such theoretical solutions are ideal and require future validation, they remain important in underground explosion monitoring scenarios because they provide fundamental physical limits on the discrimination power of tests that screen explosive from non-VDS faulting sources.

58 GEOSCIENCES↗

Ground-Based Remote Retrievals of Cumulus Entrainment Rates

While fractional entrainment rates for cumulus clouds have typically been derived from airborne observations, this limits the size and scope of available data sets. To increase the number of continental cumulus entrainment rate observations available for study, an algorithm for retrieving them from ground-based remote sensing observations has been developed. This algorithm, called the Entrainment Rate In Cumulus Algorithm (ERICA), uses the suite of instruments at the Southern Great Plains (SGP) site of the United States Department of Energy's Atmospheric Radiation Measurement (ARM) Climate Research Facility as inputs into a Gauss-Newton optimal estimation scheme, in which an assumed guess of the entrainment rate is iteratively adjusted through intercomparison of modeled liquid water path and cloud droplet effective radius to their observed counterparts. The forward model in this algorithm is the Explicit Mixing Parcel Model (EMPM), a cloud parcel model that treats entrainment as a series of discrete entrainment events. A quantified value for measurement uncertainty is also returned as part of the retrieval. Sensitivity testing and information content analysis demonstrate the robust nature of this method for retrieving accurate observations of the entrainment rate without the drawbacks of airborne sampling. Results from a test of ERICA on three months of shallow cumulus cloud events show significant variability of the entrainment rate of clouds in a single day and from one day to the next. The mean value of 1.06 km-¹ for the entrainment rate in this dataset corresponds well with prior observations and simulations of the entrainment rate in cumulus clouds.

Wagner, Timothy J.↗

Closed loop models for analyzing engineering requirements for simulators

A closed loop analytic model, incorporating a model for the human pilot, (namely, the optimal control model) that would allow certain simulation design tradeoffs to be evaluated quantitatively was developed. This model was applied to a realistic flight control problem. The resulting model is used to analyze both overall simulation effects and the effects of individual elements. The results show that, as compared to an ideal continuous simulation, the discrete simulation can result in significant performance and/or workload penalties.

Baron, S.↗

Model reduction for the dynamics and control of large structural systems via neutral network processing direct numerical optimization

Three neural network processing approaches in a direct numerical optimization model reduction scheme are proposed and investigated. Large structural systems, such as large space structures, offer new challenges to both structural dynamicists and control engineers. One such challenge is that of dimensionality. Indeed these distributed parameter systems can be modeled either by infinite dimensional mathematical models (typically partial differential equations) or by high dimensional discrete models (typically finite element models) often exhibiting thousands of vibrational modes usually closely spaced and with little, if any, damping. Clearly, some form of model reduction is in order, especially for the control engineer who can actively control but a few of the modes using system identification based on a limited number of sensors. Inasmuch as the amount of 'control spillover' (in which the control inputs excite the neglected dynamics) and/or 'observation spillover' (where neglected dynamics affect system identification) is to a large extent determined by the choice of particular reduced model (RM), the way in which this model reduction is carried out is often critical.

Becus, Georges A.↗

Parallelization of an Object-Oriented Unstructured Aeroacoustics Solver

A computational aeroacoustics code based on the discontinuous Galerkin method is ported to several parallel platforms using MPI. The discontinuous Galerkin method is a compact high-order method that retains its accuracy and robustness on non-smooth unstructured meshes. In its semi-discrete form, the discontinuous Galerkin method can be combined with explicit time marching methods making it well suited to time accurate computations. The compact nature of the discontinuous Galerkin method also makes it well suited for distributed memory parallel platforms. The original serial code was written using an object-oriented approach and was previously optimized for cache-based machines. The port to parallel platforms was achieved simply by treating partition boundaries as a type of boundary condition. Code modifications were minimal because boundary conditions were abstractions in the original program. Scalability results are presented for the SCI Origin, IBM SP2, and clusters of SGI and Sun workstations. Slightly superlinear speedup is achieved on a fixed-size problem on the Origin, due to cache effects.

Baggag, Abdelkader↗

Microbatteries for Combinatorial Studies of Conventional Lithium-Ion Batteries

Integrated arrays of microscopic solid-state batteries have been demonstrated in a continuing effort to develop microscopic sources of power and of voltage reference circuits to be incorporated into low-power integrated circuits. Perhaps even more importantly, arrays of microscopic batteries can be fabricated and tested in combinatorial experiments directed toward optimization and discovery of battery materials. The value of the combinatorial approach to optimization and discovery has been proven in the optoelectronic, pharmaceutical, and bioengineering industries. Depending on the specific application, the combinatorial approach can involve the investigation of hundreds or even thousands of different combinations; hence, it is time-consuming and expensive to attempt to implement the combinatorial approach by building and testing full-size, discrete cells and batteries. The conception of microbattery arrays makes it practical to bring the advantages of the combinatorial approach to the development of batteries.

West, William↗

Optimizing conformance control for gas injection EOR in unconventional reservoirs

Field observations have shown that interference between wells caused by connected fractures commonly exists in unconventional reservoirs. The well interference effect and conformance issues could lead to minimal enhanced oil recovery response (EOR) in actual operations if they are not properly controlled. A series of data analysis, modeling, and experimental activities were performed in this study to investigate the gas EOR strategies in the Bakken Formation. Employing the embedded discrete fracture modeling technology, a multiple-well, multiple-fracture model was developed to capture the well interference and conformance issues in the Bakken. The model made it possible to investigate the complex flow behavior between wells in the EOR (huff ‘n’ puff) processes. Alternative EOR and conformance control strategies using water and surfactant injections were studied to improve oil production performance based on experimental observations. Results showed that water injection could be used for conformance control in the Bakken because of its relatively low mobility compared to gas; however, water could also block oil flow around the conformance control wells because of the high interfacial tension between oil and water. Adding surfactant to the injection water is more effective for improving gas EOR performance since surfactant reduces the oil–water interfacial tension and contact angle effectively. Up to 9.4% more oil could be produced from the huff ‘n’ puff wells in 2 years of EOR operations when rich gas was injected at only 3 MMscf/d with 6000 psi of surfactant injection in the offset wells for conformance control.

42 ENGINEERING↗

Kernel learning backward SDE filter for data assimilation

In this paper, we develop a kernel learning backward SDE filter method to estimate the state of a stochastic dynamical system based on its partial noisy observations. A system of forward backward stochastic differential equations is used to propagate the state of the target dynamical model, and Bayesian inference is applied to incorporate the observational information. Further, to characterize the dynamical model in the entire state space, we introduce a kernel learning method to learn a continuous global approximation for the conditional probability density function of the target state by using discrete approximated density values as training data. Numerical experiments demonstrate that the kernel learning backward SDE is highly effective.

97 MATHEMATICS AND COMPUTING↗