Search NASA⌕ Search

SEARCH · Search NASA

Results for “algorithmic differentiation”

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 1,801 records · Page 100

The DESC stellarator code suite Part 3: Quasi-symmetry optimization

The DESC stellarator optimization code takes advantage of advanced numerical methods to search the full parameter space much faster than conventional tools. Only a single equilibrium solution is needed at each optimization step thanks to automatic differentiation, which efficiently provides exact derivative information. A Gauss–Newton trust-region optimization method uses second-order derivative information to take large steps in parameter space and converges rapidly. With just-in-time compilation and GPU portability, high-dimensional stellarator optimization runs take orders of magnitude less computation time with DESC compared to other approaches. This paper presents the theory of the DESC fixed-boundary local optimization algorithm along with demonstrations of how to easily implement it in the code. Example quasi-symmetry optimizations are shown and compared to results from conventional tools. Three different forms of quasi-symmetry objectives are available in DESC, and their relative advantages are discussed in detail. In the examples presented, the triple product formulation yields the best optimization results in terms of minimized computation time and particle transport. This paper concludes with an explanation of how the modular code suite can be extended to accommodate other types of optimization problems.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

An extended trajectory-mechanics approach for calculating two-phase flow paths

A technique originating in quantum dynamics is used to derive a trajectory-based, semi-analytical solution for two-phase flow. The partial differential equation governing the evolution of the aqueous phase is equivalent to a family of ordinary differential equations defined along a path through the porous medium. The trajectories may be found by solving the differential equations directly or by post-processing the output of a numerical solution to the full set of governing equations. The trajectories, which differ from conventional streamlines, are found to bend downward in response to gravitational forces. The curvature is more pronounced as the dip of the porous layer containing the flow increases. Subtle changes in the relative permeability curve can lead to significant variations in the trajectories. The ordinary differential equation for the trajectory provides an expression for the travel time along the path. The expression produces a semi-analytical approximation to the model parameter sensitivities, the partial derivatives of the travel times with respect to changes in the permeability model. The semi-analytical trajectory-based sensitivities generally agree with those computed using a numerical reservoir simulator and a perturbation approach. The sensitivities are useful in tomographic imaging algorithms designed to estimate the spatial variation in permeability within a porous medium using multiphase observations.

58 GEOSCIENCES↗

Convex relaxation for Fokker–Planck equation

We propose an approach to directly estimate the moments or marginals for a high-dimensional equilibrium distribution in statistical mechanics by solving the high-dimensional Fokker–Planck equation in terms of low-order cluster moments or marginals. With this approach, we bypass the exponential complexity of estimating the full high-dimensional distribution and directly solve the simplified partial differential equations for low-order moments/marginals. Moreover, the proposed moment/marginal relaxation is fully convex and can be solved via off-the-shelf solvers. We further propose a time-dependent version of the convex programs to study non-equilibrium dynamics. In a specific setting, we show the proposed method can recover a mean-field-type equilibrium density. Numerical results are provided to demonstrate the performance of the proposed algorithm for high-dimensional systems.

Chen, Yian↗

Stochastic basis adaptation and spatial domain decomposition for partial differential equations with random coefficients

We present a novel approach for high-dimensional stochastic partial differential equations that decomposes the computational domain into smaller disjoint subdomains and reduces the random dimension in each subdomain to save computational costs. The proposed approach combines spatial domain decomposition and stochastic basis adaptation in each subdomain. We adapt the stochastic basis in each subdomain such that the local solution has a low-dimensional random space representation. To compute the stochastic solution in each subdomain, we use the Neumann-Neumann algorithm, where we first estimate the interface solution and then evaluate the interior solution in each subdomain using the interface solution as a boundary condition. The interior solutions in each subdomain are computed independently of each other, which reduces the operation count from $\approx O(N^\alpha)$ to $\approx O(M^\alpha),$ where $N$ is the total number of degrees of freedom, $M$ is the number of degrees of freedom in each subdomain, and the exponent $\alpha>1$ depends on the uncertainty quantification method used. In addition, the localized nature of solutions makes the proposed approach highly parallelizable. We illustrate the accuracy and efficiency of the approach for a steady-state diffusion equation with a random space-dependent diffusion coefficient.

uncertainty quantification, basis adaptation↗

Numerical computation of shock wave-turbulent boundary layer interaction in transonic flow over an axisymmetric curved hill

A control-volume based finite difference computation of a turbulent transonic flow over an axisymmetric curved hill is presented. The numerical method is based on the SIMPLE algorithm, and hence the conservation of mass equation is replaced by a pressure correction equation for compressible flows. The turbulence is described by a k-epsilon turbulence model supplemented by a near-wall turbulence model. In the method, the dissipation rate in the region very close to the wall is obtained from an algebraic equation and that for the rest of the flow domain is obtained by solving a partial differential equation for the dissipation rate. The other flow equations are integrated up to the wall. It is shown that the present turbulence model yields the correct location of the compression shock. The other computational results are also in good agreement with experimental data.

Kim, S.-W.↗

A Preliminary Analysis of Precipitation Properties and Processes during NASA GPM IFloodS

The Iowa Flood Studies (IFloodS) is a NASA Global Precipitation Measurement (GPM) ground measurement campaign, which took place in eastern Iowa from May 1 to June 15, 2013. The goals of the field campaign were to collect detailed measurements of surface precipitation using ground instruments and advanced weather radars while simultaneously collecting data from satellites passing overhead. Data collected by the radars and other ground instruments, such as disdrometers and rain gauges, will be used to characterize precipitation properties throughout the vertical column, including the precipitation type (e.g., rain, graupel, hail, aggregates, ice crystals), precipitation amounts (e.g., rain rate), and the size and shape of raindrops. The impact of physical processes, such as aggregation, melting, breakup and coalescence on the measured liquid and ice precipitation properties will be investigated. These ground observations will ultimately be used to improve rainfall estimates from satellites and in particular the algorithms that interpret raw data for the upcoming GPM mission's Core Observatory satellite, which launches in 2014. The various precipitation data collected will eventually be used as input to flood forecasting models in an effort to improve capabilities and test the utility and limitations of satellite precipitation data for flood forecasting. In this preliminary study, the focus will be on analysis of NASA NPOL (S‐band, polarimetric) radar (e.g., radar reflectivity, differential reflectivity, differential phase, correlation coefficient) and NASA 2D Video Disdrometers (2DVDs) measurements. Quality control and processing of the radar and disdrometer data sets will be outlined. In analyzing preliminary cases, particular emphasis will be placed on 1) documenting the evolution of the rain drop size distribution (DSD) as a function of column melting processes and 2) assessing the impact of range on ground‐based polarimetric radar estimates of DSD properties.

Carey, Lawrence↗

A vectorized algorithm for 3D dynamics of a tethered satellite

Equations of motion characterizing the three dimensional motion of a tethered satellite during the retrieval phase are studied. The mathematical model involves an arbitrary number of point masses connected by weightless cords. Motion occurs in a gravity gradient field. The formulation presented accounts for general functions describing support point motion, rate of tether retrieval, and arbitrary forces applied to the point masses. The matrix oriented program language MATLAB is used to produce an efficient vectorized formulation for computing natural frequencies and mode shapes for small oscillations about the static equilibrium configuration; and for integrating the nonlinear differential equations governing large amplitude motions. An example of time response pertaining to the skip rope effect is investigated.

Wilson, Howard B.↗

SMACK: A New Algorithm for Modeling Collisions and Dynamics of Planetesimals in Debris Disks

We present the Superparticle Model/Algorithm for Collisions in Kuiper belts and debris disks (SMACK), a new method for simultaneously modeling, in 3-D, the collisional and dynamical evolution of planetesimals in a debris disk with planets. SMACK can simulate azimuthal asymmetries and how these asymmetries evolve over time. We show that SMACK is stable to numerical viscosity and numerical heating over 10(exp 7) yr, and that it can reproduce analytic models of disk evolution. We use SMACK to model the evolution of a debris ring containing a planet on an eccentric orbit. Differential precession creates a spiral structure as the ring evolves, but collisions subsequently break up the spiral, leaving a narrower eccentric ring.

circumstellar matter↗

Momentum informed muon scattering tomography for monitoring spent nuclear fuels in dry storage cask

Development of an effective monitoring method for spent nuclear fuel (SNF) in a dry storage cask (DSC) is important to meet the increasing demand for dry storage investigations. The DSC investigation should provide information about the quantity of stored SNF, and quality assurance of materials should be possible without opening the cask. However, traditional nondestructive examination (NDE) methods such as x-rays are difficult to deploy for DSC investigation because a typical DSC is intentionally designed to shield against radiation. To address this challenge, cosmic ray muons (CRMs) are used as an alternative NDE radiation probe because they can easily penetrate an entire DSC system; however, a wide application of muons is often hindered due to the naturally low CRM flux (~10 4 muons/m 2 /min). This paper introduces a newly proposed imaging algorithm, momentum-informed muon scattering tomography (MMST) , and presents how a limitation of the current muon scattering tomography technique has been addressed by measuring muon momentum. To demonstrate its functionality, a commercial DSC with 24 pressurized light water reactor fuel assemblies (FAs) and the MMST system were designed in GEANT4. Three noticeable improvements were observed for MMST system as a DSC investigation tool: (1) a signal stabilization, (2) an enhanced capability to differentiate various materials, and (3) statistically increased precision to identify and locate missing FAs. The results show that MMST improves the investigation accuracy from 79 to 98% when one FA is missing and 51% to 88% when one-half FA is missing. The advancement of the NDE technique using CRM for DSC verification is expected to resolve long-standing problems in increasing demand for DSC inspections and nuclear security.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Two-dimensional mesh embedding for Galerkin B-spline methods

A number of advantages result from using B-splines as basis functions in a Galerkin method for solving partial differential equations. Among them are arbitrary order of accuracy and high resolution similar to that of compact schemes but without the aliasing error. This work develops another property, namely, the ability to treat semi-structured embedded or zonal meshes for two-dimensional geometries. This can drastically reduce the number of grid points in many applications. Both integer and non-integer refinement ratios are allowed. The report begins by developing an algorithm for choosing basis functions that yield the desired mesh resolution. These functions are suitable products of one-dimensional B-splines. Finally, test cases for linear scalar equations such as the Poisson and advection equation are presented. The scheme is conservative and has uniformly high order of accuracy throughout the domain.

Shariff, Karim↗

A Walsh Function Module Users' Manual

The solution of partial differential equations (PDEs) with Walsh functions offers new opportunities to simulate many challenging problems in mathematical physics. The approach was developed to better simulate hypersonic flows with shocks on unstructured grids. It is unique in that integrals and derivatives are computed using simple matrix multiplication of series representations of functions without the need for divided differences. The product of any two Walsh functions is another Walsh function - a feature that radically changes an algorithm for solving PDEs. A FORTRAN module for supporting Walsh function simulations is documented. A FORTRAN code is also documented with options for solving time-dependent problems: an advection equation, a Burgers equation, and a Riemann problem. The sample problems demonstrate the usage of the Walsh function module including such features as operator overloading, Fast Walsh Transforms in multi-dimensions, and a Fast Walsh reciprocal.

Gnoffo, Peter A.↗

MedNAS: Multiscale Training-Free Neural Architecture Search for Medical Image Analysis

Deep neural networks have demonstrated impressive results in medical image analysis, but designing suitable architectures for each specific task is expertise dependent and time consuming. Neural architecture search (NAS) offers an effective means of discovering architectures. It has been highly successful in numerous applications, particularly in natural image classification. Yet, medical images possess unique characteristics, such as small regions and a wide variety of lesion sizes, that differentiate them from natural images. Furthermore, most current NAS methods struggle with high computational costs, especially when dealing with high-resolution image datasets. In this article, we present a novel evolutionary NAS method called multiscale training-free neural architecture search (MSTF-NAS) to address these challenges. Specifically, to accommodate the broad range of lesion region sizes in disease diagnosis, we develop a new reduction cell search space that enables the search algorithm to explicitly identify the optimal scale combination for multiscale feature extraction. Further, to overcome the issue of high computational costs, we utilize training-free indicators as performance measures for candidate architectures, which allows us to search for the optimal architecture more efficiently. More specifically, by considering the capability and simplicity of various networks, we formulate a multiobjective optimization problem that involves two training-free indicators and model complexity for candidate architectures. Extensive experiments on a large medical image benchmark and a publicly available breast cancer detection dataset are conducted. The empirical results demonstrate that our MSTF-NAS outperforms both human-designed architectures and current state-of-the-art NAS algorithms on both datasets, indicating the effectiveness of our proposed method.

60 APPLIED LIFE SCIENCES↗

Computer-aided design and distributed system technology development for large space structures

Proposed large space structures have many characteristics that make them difficult to analyze and control. They are highly flexible, with components mathematically modeled by partial differential equations or very large systems of ordinary differential equations. They have many resonant frequencies, typically low and closely spaced. Natural damping may be low and/or improperly modeled. Coupled with stringent operational requirements of orientation, shape control, and vibration suppression, and the inability to perform adequate ground testing, these characteristics present an unconventional identification and control design problem to the systems theorist. Some of the research underway within Langley's Spacecraft Control Branch, Guidance and Control Division aimed at developing theory and algorithms to treat large space structures systems identification and control problems is described. The research areas to be considered are computer-aided design algorithms, and systems identification and control of distributed systems.

Armstrong, Ernest S.↗

Application-driven Privacy-preserving Data Publishing with Correlated Attributes

Recent advances in computing have allowed for the possibility to collect large amounts of data on personal activities and private living spaces. To address the privacy concerns of users in this environment, we propose a novel framework called PR-GAN that offers privacy-preserving mechanism using generative adversarial networks. Given a target application, PR-GAN automatically modifies the data to hide sensitive attributes - which may be hidden and can be inferred by machine learning algorithms - while preserving the data utility in the target application. Unlike prior works, the public's possible knowledge of the correlation between the target application and sensitive attributes is built into our modeling. We formulate our problem as an optimization problem, show that an optimal solution exists and use generative adversarial networks (GAN) to create perturbations. We further show that our method provides privacy guarantees under the Pufferfish framework, an elegant generalization of the differential privacy that allows for the modeling of prior knowledge on data and correlations. Through experiments, we show that our method outperforms conventional methods in effectively hiding the sensitive attributes while guaranteeing high performance in the target application, for both property inference and training purposes. Finally, we demonstrate through further experiments that once our model learns a privacy-preserving task, such as hiding subjects' identity, on a group of individuals, it can perform the same task on a separate group with minimal performance drops.

97 MATHEMATICS AND COMPUTING↗

Solving high-dimensional partial integral differential equations: The finite expression method

Partial integro-differential equations (PIDEs) have broad applications in the sciences, from electro-magnetism to options pricing. Here, in this paper, we introduce a new finite expression method (FEX) to solve PIDEs. This approach builds upon the original FEX and its inherent advantages with new advances: 1) A novel method of parameter grouping is proposed to reduce the number of coefficients in high-dimensional function approximation; 2) A Taylor series approximation method is implemented to significantly improve the computational efficiency and accuracy of the evaluation of the integral terms of PIDEs. The new FEX based method, denoted FEX-PG to indicate the addition of the parameter grouping (PG) step to the algorithm, provides both high accuracy and interpretable numerical solutions, with the outcome being an explicit equation that facilitates intuitive understanding of the underlying solution structures. These features are often absent in traditional methods, such as finite element methods (FEM) and finite difference methods, as well as in deep learning-based approaches. To benchmark our method against recent advances, we apply the new FEX-PG to solve benchmark PIDEs in the literature. In high-dimensional settings, FEX-PG exhibits strong and robust performance, achieving relative errors on the order of single precision machine epsilon, significantly outperforming existing approaches based on neural networks.

Combinatorial optimization↗

Characterizing 4-string contact interaction using machine learning

Abstract The geometry of 4-string contact interaction of closed string field theory is characterized using machine learning. We obtain Strebel quadratic differentials on 4-punctured spheres as a neural network by performing unsupervised learning with a custom-built loss function. This allows us to solve for local coordinates and compute their associated mapping radii numerically. We also train a neural network distinguishing vertex from Feynman region. As a check, 4-tachyon contact term in the tachyon potential is computed and a good agreement with the results in the literature is observed. We argue that our algorithm is manifestly independent of number of punctures and scaling it to characterize the geometry ofn-string contact interaction is feasible.

Physics↗

Modeling of wave processes in blocky media with porous and fluid-saturated interlayers

The wave processes in blocky media are analyzed by applying different mathematical models, wherein the elastic blocks interact with each other via pliant interlayers with the complex mechanical properties. Four versions of constitutive equations are considered. In the first version, an elastic interaction between the blocks is simulated within the framework of linear elasticity theory, and the model of elastic–plastic interlayers is constructed to take into account the appearance of irreversible deformation of interlayers at short time intervals. In the second one, the effects of viscoelastic shear in the interblock interlayers are taken into the consideration using the Poynting–Thomson rheological scheme. In the third option, the model of an elastic porous material is used in the interlayers, where the pores collapse if an abrupt compressive stress is applied. In the fourth case, the model of a fluid-saturated material with open pores is examined based on Biot's equations. The collapse of pores is modeled by the generalized rheological approach, wherein the mechanical properties of a material are simulated using four rheological elements. Three of them are the traditional elastic, viscous and plastic elements, the fourth element is the so-called rigid contact, which is used to describe the behavior of materials with the different resistance to tension and compression. It was shown that the thermodynamically consistent model is provided, which means that the energy balance equation is fulfilled for an entire blocky structure, where the kinetic and potential energy of the system is the sum of the kinetic and potential energies of the blocks and interlayers. Under numerical implementation of the interlayers models, the dissipationless finite difference Ivanov's method was used. The splitting method by spatial variables in the combination with the Godunov gap decay scheme was applied in the blocks. As a result, robust and stable computational algorithms are built and tested. Using MPI technology, the parallel software was designed for the modeling of wave processes in 2D setting. The numerical results are presented, discussed and future studies are outlined. - Highlights: • We model the propagation of waves in a blocky medium with compliant interlayers. • Models of porous and fluid saturated interlayers are proposed using Biot's approach. • A parallel computational algorithm is developed for supercomputers of cluster type. • Problems of instant rotation of central and boundary blocks are solved numerically. • Wave motions are compared in the cases of elastic and fluid-saturated interlayers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DoE Phase II SBIR: Spectrally-Assisted Vehicle Tracking

The goal of this Phase II SBIR is to develop a prototype software package to demonstrate spectrally-aided vehicle tracking performance. The primary application is to demonstrate improved target vehicle tracking performance in complex environments where traditional spatial tracker systems may show reduced performance. Example scenarios in Figure 1 include a) the target vehicle obscured by a large structure for an extended period of time, or b), the target engaging in extreme maneuvers amongst other civilian vehicles. The target information derived from spatial processing is unable to differentiate between the green versus the red vehicle. Spectral signature exploitation enables comparison of new candidate targets with existing track signatures. The ambiguity in this confusing scenario is resolved by folding spectral analysis results into each target nomination and association processes. Figure 3 shows a number of example spectral signatures from a variety of natural and man-made materials. The work performed over the two-year effort was divided into three general areas: algorithm refinement, software prototype development, and prototype performance demonstration. The tasks performed under this Phase II to accomplish the program goals were as follows: 1. Acquire relevant vehicle target datasets to support prototype. 2. Refine algorithms for target spectral feature exploitation. 3. Implement a prototype multi-hypothesis target tracking software package. 4. Demonstrate and quantify tracking performance using relevant data.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Refine your search to explore more results.