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 469 records · Page 26

ZEUS: An Efficient GPU Optimization Method Integrating PSO, BFGS, and Automatic Differentiation

We introduce a novel, efficient computational method, ZEUS, for numerical optimization, and provide an open-source implementation. It has four key ingredients: (1) particle swarm optimization (PSO), (2) the use of the Broyden-Fletcher-Goldfarb-Shanno (BFGS) method, (3) automatic differentiation (AD), and (4) GPUs. Our approach addresses the computational challenges inherent in high-dimensional, non-convex optimization problems. In the first phase of the algorithm, we get a potentially good set of starting points using PSO. Thereafter, we run BFGS independently in parallel from these starting points. BFGS is one of the best-performing algorithms for numerical optimization. However, it requires the gradient of the function being optimized. ZEUS integrates automatic differentiation into BFGS thus avoiding the need for the user to calculate derivatives explicitly. The use of GPUs allows ZEUS to speed up the calculations substantially. We carry out systematic studies to explore the trade-offs between the number of PSO iterations taken, starting points, and BFGS iteration depth. We show that a handful of iterations of PSO can improve global convergence when combined with BFGS. We also present performance studies using common test functions. The source code can be found at https://github.com/fnal-numerics/global-optimizer-gpu.

Soos, Dominik [Old Dominion U.]↗

Learning Stochastic Parametric Differentiable Predictive Control Policies

We present a scalable unsupervised learning-based method for obtaining explicit control policies for model predictive control problems for stochastic linear systems with additive uncertainties subject to nonlinear chance constraints. We call the proposed method stochastic parametric differentiable predictive control (SP-DPC), which extends the recently proposed deterministic DPC policy optimization algorithm. We formulate the SP-DPC as a deterministic approximation to the stochastic parametric constrained optimal control problem via independent sampling of the problem's parameters and uncertainties. This formulation allows us to directly compute the policy gradients via automatic differentiation of the problem's value function, evaluated over sampled parameters and uncertainties. In particular, the computed expectation of the problem's value function is backpropagated through the finite-time closed-loop system rollouts parametrized by a known nominal system dynamics model and neural control policy. We also provide theoretical probabilistic guarantees on closed-loop stability and chance constraints satisfaction for systems controlled by learned neural policies. We demonstrate the computational efficiency and scalability of the proposed policy optimization algorithm in three numerical examples, including systems with a large number of states or subject to nonlinear constraints.

Drgona, Jan↗

A dynamically bi-orthogonal method for time-dependent stochastic partial differential equations II: Adaptivity and generalizations

This is part II of our paper in which we propose and develop a dynamically bi-orthogonal method (DyBO) to study a class of time-dependent stochastic partial differential equations (SPDEs) whose solutions enjoy a low-dimensional structure. In part I of our paper [9], we derived the DyBO formulation and proposed numerical algorithms based on this formulation. Some important theoretical results regarding consistency and bi-orthogonality preservation were also established in the first part along with a range of numerical examples to illustrate the effectiveness of the DyBO method. In this paper, we focus on the computational complexity analysis and develop an effective adaptivity strategy to add or remove modes dynamically. Our complexity analysis shows that the ratio of computational complexities between the DyBO method and a generalized polynomial chaos method (gPC) is roughly of order O((m/N{sub p}){sup 3}) for a quadratic nonlinear SPDE, where m is the number of mode pairs used in the DyBO method and N{sub p} is the number of elements in the polynomial basis in gPC. The effective dimensions of the stochastic solutions have been found to be small in many applications, so we can expect m is much smaller than N{sub p} and computational savings of our DyBO method against gPC are dramatic. The adaptive strategy plays an essential role for the DyBO method to be effective in solving some challenging problems. Another important contribution of this paper is the generalization of the DyBO formulation for a system of time-dependent SPDEs. Several numerical examples are provided to demonstrate the effectiveness of our method, including the Navier–Stokes equations and the Boussinesq approximation with Brownian forcing.

97 MATHEMATICS AND COMPUTING↗

Verification of Bison fission product species conservation under TRISO reactor conditions

When assessing the reliability and predictive capabilities of a simulation tool, code verification is used to ensure that the implemented numerical algorithm is a faithful representation of its underlying mathematical model, including partial differential or integral equations, initial and boundary conditions, and auxiliary relationships. During this process, numerical results in a discrete solution are compared to the analytical solution of the mathematical model. Here, in this paper, the code verification process is applied to one-dimensional spatiotemporal problems that exercise partial differential equation governing the conservation of fission product species (or mass diffusion). Numerical experiments were performed in the Bison fuel performance code to evaluate its predictive capability under various TRISO reactor conditions such as base irradiation and safety heating test conditions for either short- or long-lived fission product species, as well as a case concerning evaporation from the outer surface of a particle. The code predictions were compared with the expected exact results obtained from the analytical expressions, and the fact that they demonstrate the correct analytical behavior provides strong evidence of proper numerical algorithm implementation.

07 ISOTOPE AND RADIATION SOURCES↗

A WAVELET-GALERKIN ALGORITHM OF THE E/B DECOMPOSITION OF COSMIC MICROWAVE BACKGROUND POLARIZATION MAPS

We develop an algorithm of separating the E and B modes of the cosmic microwave background (CMB) polarization from the noisy and discretized maps of Stokes parameters Q and U in a finite area. A key step of the algorithm is to take a wavelet-Galerkin discretization of the differential relation between the E, B and Q, U fields. This discretization allows derivative operator to be represented by a matrix, which is exactly diagonal in scale space, and narrowly banded in spatial space. We show that the effect of boundary can be eliminated by dropping a few discrete wavelet transform modes, located on or nearby the boundary. This method reveals that the derivative operators will cause large errors in the E and B power spectra on small scales if the Q and U maps contain Gaussian noise. It also reveals that if the Q and U maps are random, these fields lead to the mixing of E and B modes. Consequently, the B mode will be contaminated if the powers of E modes are much larger than that of B modes. Nevertheless, numerical tests show that the power spectra of both E and B on scales larger than the finest scale by a factor of 4 and higher can reasonably be recovered, even when the power ratio of E to B modes is as large as about 10{sup 2}, and the signal-to-noise ratio is equal to 10 and higher. This is because the Galerkin discretization is free of false correlations and keeps the contamination under control. As wavelet variables contain information of both spatial and scale spaces, the developed method is also effective to recover the spatial structures of the E and B mode fields.

79 ASTRONOMY AND ASTROPHYSICS↗

PHURBAS: AN ADAPTIVE, LAGRANGIAN, MESHLESS, MAGNETOHYDRODYNAMICS CODE. I. ALGORITHM

We present an algorithm for simulating the equations of ideal magnetohydrodynamics and other systems of differential equations on an unstructured set of points represented by sample particles. Local, third-order, least-squares, polynomial interpolations (Moving Least Squares interpolations) are calculated from the field values of neighboring particles to obtain field values and spatial derivatives at the particle position. Field values and particle positions are advanced in time with a second-order predictor-corrector scheme. The particles move with the fluid, so the time step is not limited by the Eulerian Courant-Friedrichs-Lewy condition. Full spatial adaptivity is implemented to ensure the particles fill the computational volume, which gives the algorithm substantial flexibility and power. A target resolution is specified for each point in space, with particles being added and deleted as needed to meet this target. Particle addition and deletion is based on a local void and clump detection algorithm. Dynamic artificial viscosity fields provide stability to the integration. The resulting algorithm provides a robust solution for modeling flows that require Lagrangian or adaptive discretizations to resolve. This paper derives and documents the Phurbas algorithm as implemented in Phurbas version 1.1. A following paper presents the implementation and test problem results.

79 ASTRONOMY AND ASTROPHYSICS↗

Sampling-Based Motion Planning Algorithms for Replanning and Spatial Load Balancing

The common theme of this dissertation is sampling-based motion planning with the two key contributions being in the area of replanning and spatial load balancing for robotic systems. Here, we begin by recalling two sampling-based motion planners: the asymptotically optimal rapidly-exploring random tree (RRT*), and the asymptotically optimal probabilistic roadmap (PRM*). We also provide a brief background on collision cones and the Distributed Reactive Collision Avoidance (DRCA) algorithm. The next four chapters detail novel contributions for motion replanning in environments with unexpected static obstacles, for multi-agent collision avoidance, and spatial load balancing. First, we show improved performance of the RRT* when using the proposed Grandparent-Connection (GP) or Focused-Refinement (FR) algorithms. Next, the Goal Tree algorithm for replanning with unexpected static obstacles is detailed and proven to be asymptotically optimal. A multi-agent collision avoidance problem in obstacle environments is approached via the RRT*, leading to the novel Sampling-Based Collision Avoidance (SBCA) algorithm. The SBCA algorithm is proven to guarantee collision free trajectories for all of the agents, even when subject to uncertainties in the knowledge of the other agents’ positions and velocities. Given that a solution exists, we prove that livelocks and deadlock will lead to the cost to the goal being decreased. We introduce a new deconfliction maneuver that decreases the cost-to-come at each step. This new maneuver removes the possibility of livelocks and allows a result to be formed that proves convergence to the goal configurations. Finally, we present a limited range Graph-based Spatial Load Balancing (GSLB) algorithm which fairly divides a non-convex space among multiple agents that are subject to differential constraints and have a limited travel distance. The GSLB is proven to converge to a solution when maximizing the area covered by the agents. The analysis for each of the above mentioned algorithms is confirmed in simulations.

97 MATHEMATICS AND COMPUTING↗

Solving the Bernstein-Vazirani problem using Majorana-based topological quantum algorithms

Executing quantum algorithms using Majorana zero modes—a major milestone for the field of topological quantum computing—requires a platform that can be scaled to large quantum registers, can be controlled in real time and space, and a braiding protocol that uses the unique properties of these exotic particles. Here, we demonstrate the first successful simulation of a Majorana-based, fault-tolerant quantum algorithm to solve the Bernstein-Vazirani problem in two-dimensional magnet-superconductor hybrid structures from initialization to read-out of the final many-body state. Utilizing the Majorana zero modes’ topological properties, we introduce an optimized braiding protocol for the algorithm and a scalable architecture for its implementation with an arbitrary number of qubits. We visualize the algorithm protocol in real time and space by computing the non-equilibrium density of states, which is proportional to the time-dependent differential conductance, and the non-equilibrium charge density, which assigns a unique signature to each final state of the algorithm.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

A stochastic multiscale framework for modeling flow through random heterogeneous porous media

Flow through porous media is ubiquitous, occurring from large geological scales down to the microscopic scales. Several critical engineering phenomena like contaminant spread, nuclear waste disposal and oil recovery rely on accurate analysis and prediction of these multiscale phenomena. Such analysis is complicated by inherent uncertainties as well as the limited information available to characterize the system. Any realistic modeling of these transport phenomena has to resolve two key issues: (i) the multi-length scale variations in permeability that these systems exhibit, and (ii) the inherently limited information available to quantify these property variations that necessitates posing these phenomena as stochastic processes. A stochastic variational multiscale formulation is developed to incorporate uncertain multiscale features. A stochastic analogue to a mixed multiscale finite element framework is used to formulate the physical stochastic multiscale process. Recent developments in linear and non-linear model reduction techniques are used to convert the limited information available about the permeability variation into a viable stochastic input model. An adaptive sparse grid collocation strategy is used to efficiently solve the resulting stochastic partial differential equations (SPDEs). The framework is applied to analyze flow through random heterogeneous media when only limited statistics about the permeability variation are given.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Weak Convergence of a Mass-Structured Individual-Based Model

We propose a model of chemostat where the bacterial population is individually-based, each bacterium is explicitly represented and has a mass evolving continuously over time. The substrate concentration is represented as a conventional ordinary differential equation. These two components are coupled with the bacterial consumption. Mechanisms acting on the bacteria are explicitly described (growth, division and washout). Bacteria interact via consumption. We set the exact Monte Carlo simulation algorithm of this model and its mathematical representation as a stochastic process. We prove the convergence of this process to the solution of an integro-differential equation when the population size tends to infinity. Finally, we propose several numerical simulations.

60 APPLIED LIFE SCIENCES↗

Attenuation correction in emission tomography using the emission data—A review

The problem of attenuation correction (AC) for quantitative positron emission tomography (PET) had been considered solved to a large extent after the commercial availability of devices combining PET with computed tomography (CT) in 2001; single photon emission computed tomography (SPECT) has seen a similar development. However, stimulated in particular by technical advances toward clinical systems combining PET and magnetic resonance imaging (MRI), research interest in alternative approaches for PET AC has grown substantially in the last years. In this comprehensive literature review, the authors first present theoretical results with relevance to simultaneous reconstruction of attenuation and activity. The authors then look back at the early history of this research area especially in PET; since this history is closely interwoven with that of similar approaches in SPECT, these will also be covered. We then review algorithmic advances in PET, including analytic and iterative algorithms. The analytic approaches are either based on the Helgason–Ludwig data consistency conditions of the Radon transform, or generalizations of John’s partial differential equation; with respect to iterative methods, we discuss maximum likelihood reconstruction of attenuation and activity (MLAA), the maximum likelihood attenuation correction factors (MLACF) algorithm, and their offspring. The description of methods is followed by a structured account of applications for simultaneous reconstruction techniques: this discussion covers organ-specific applications, applications specific to PET/MRI, applications using supplemental transmission information, and motion-aware applications. After briefly summarizing SPECT applications, we consider recent developments using emission data other than unscattered photons. In summary, developments using time-of-flight (TOF) PET emission data for AC have shown promising advances and open a wide range of applications. These techniques may both remedy deficiencies of purely MRI-based AC approaches in PET/MRI and improve standalone PET imaging.

61 RADIATION PROTECTION AND DOSIMETRY↗

Structural characterization of an intrinsically disordered protein complex using integrated small-angle neutron scattering and computing

Characterizing structural ensembles of intrinsically disordered proteins (IDPs) and intrinsically disordered regions (IDRs) of proteins is essential for studying structure–function relationships. Due to the different neutron scattering lengths of hydrogen and deuterium, selective labeling and contrast matching in small-angle neutron scattering (SANS) becomes an effective tool to study dynamic structures of disordered systems. However, experimental timescales typically capture measurements averaged over multiple conformations, leaving complex SANS data for disentanglement. We hereby demonstrate an integrated method to elucidate the structural ensemble of a complex formed by two IDRs. We use data from both full contrast and contrast matching with residue-specific deuterium labeling SANS experiments, microsecond all-atom molecular dynamics (MD) simulations with four molecular mechanics force fields, and an autoencoder-based deep learning (DL) algorithm. From our combined approach, we show that selective deuteration provides additional information that helps characterize structural ensembles. We find that among the four force fields, a99SB-disp and CHARMM36m show the strongest agreement with SANS and NMR experiments. In addition, our DL algorithm not only complements conventional structural analysis methods but also successfully differentiates NMR and MD structures which are indistinguishable on the free energy surface. Finally, we present an ensemble that describes experimental SANS and NMR data better than MD ensembles generated by one single force field and reveal three clusters of distinct conformations. Our results demonstrate a new integrated approach for characterizing structural ensembles of IDPs.

59 BASIC BIOLOGICAL SCIENCES↗

Approximating the linear response of physical chaos

Abstract Parametric derivatives of statistics are highly desired quantities in prediction, design optimization and uncertainty quantification. In the presence of chaos, the rigorous computation of these quantities is certainly possible, but mathematically complicated and computationally expensive. Based on Ruelle’s formalism, this paper shows that the sophisticated linear response algorithm can be dramatically simplified in higher-dimensional systems featuring statistical homogeneity in the physical space. We argue that the contribution of the SRB (Sinai–Ruelle–Bowen) measure gradient, which is an integral yet the most cumbersome part of the full algorithm, is negligible if the objective function is appropriately aligned with unstable manifolds. This abstract condition could potentially be satisfied by a vast family of real-world chaotic systems, regardless of the physical meaning and mathematical form of the objective function and perturbed parameter. We demonstrate several numerical examples that support these conclusions and that present the use and performance of a simplified linear response algorithm. In the numerical experiments, we consider physical models described by differential equations, including Lorenz 96 and Kuramoto–Sivashinsky.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Reduced Common-Mode Voltage PWM Scheme for Full-SiC Three-Level Uninterruptible Power Supply With Small DC-Link Capacitors

In this article, a pulsewidth modulation (PWM) scheme for three-level full-SiC uninterruptible power supplies is developed to achieve a high power density. Two key passive components are selected for size reduction of the ac-ac stage: common-mode (CM) EMI filter, and dc-link capacitors. To reduce the CM noise, a new vector combination is proposed based on synchronous switching among three-phases. The proposed combinations align CM voltage (CMV) to be a single pulse per switching period. Owing to the simple shape, CMV cancellation between a three-level rectifier and inverter can be maximally utilized. A transition between the three combinations can control the drift of neutral point voltage. An equivalent carrier-based implementation is developed. Second, a simple algorithm to compensate neutral point voltage fluctuation is proposed both for differential mode (DM) and CM output voltage. Low-order harmonics on three-phase currents and an additional high-frequency CM noise by misaligned switching instant can be eliminated. The proposed compensation can be implemented by a simple correction on carrier slopes and injected zero-sequence voltage. The proposed PWM scheme is verified with 20-kW full-SiC UPS switching at 60 kHz with 140 μF dc-link capacitors.

42 ENGINEERING↗

Noise in x-ray grating-based phase-contrast imaging

Purpose: Grating-based x-ray phase-contrast imaging is a fast developing new modality not only for medical imaging, but as well for other fields such as material sciences. While these many possible applications arise, the knowledge of the noise behavior is essential. Methods: In this work, the authors used a least squares fitting algorithm to calculate the noise behavior of the three quantities absorption, differential phase, and dark-field image. Further, the calculated error formula of the differential phase image was verified by measurements. Therefore, a Talbot interferometer was setup, using a microfocus x-ray tube as source and a Timepix detector for photon counting. Additionally, simulations regarding this topic were performed. Results: It turned out that the variance of the reconstructed phase is only dependent of the total number of photons used to generate the phase image and the visibility of the experimental setup. These results could be evaluated in measurements as well as in simulations. Furthermore, the correlation between absorption and dark-field image was calculated. Conclusions: These results provide the understanding of the noise characteristics of grating-based phase-contrast imaging and will help to improve image quality.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Measurement of jet shapes in top-quark pair events at s = 7 TeV using the ATLAS detector

CERN-LHC. Measurement of the jet shapes in top-quark pair events from proton-proton collisions at a centre-of-mass energy of 7 TeV. The analysis uses a data sample of integrated luminosity 1.8 fb-1. Jets are reconstructed using the anti-kT algorithm with jet radius R=0.4. Jet shapes are presented as differential and integrated defined as: Differential (RHO) - scalar sum of jet PT in an annulus of thickness deltaR around radius r / scalar sum of jet PT in the cone of radius R. Integrated (PSI) - scalar sum of jet PT in cone radius r / scalar sum of jet PT in cone radius R. Data are shown separated into b-jets and light-jet events.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The orbit and transit prospects for β pictoris b constrained with one milliarcsecond astrometry

A principal scientific goal of the Gemini Planet Imager (GPI) is obtaining milliarcsecond astrometry to constrain exoplanet orbits. However, astrometry of directly imaged exoplanets is subject to biases, systematic errors, and speckle noise. Here, we describe an analytical procedure to forward model the signal of an exoplanet that accounts for both the observing strategy (angular and spectral differential imaging) and the data reduction method (Karhunen–Loève Image Projection algorithm). We use this forward model to measure the position of an exoplanet in a Bayesian framework employing Gaussian processes and Markov-chain Monte Carlo to account for correlated noise. In the case of GPI data on β Pic b, this technique, which we call Bayesian KLIP-FM Astrometry (BKA), outperforms previous techniques and yields 1σ errors at or below the one milliarcsecond level. We validate BKA by fitting a Keplerian orbit to 12 GPI observations along with previous astrometry from other instruments. The statistical properties of the residuals confirm that BKA is accurate and correctly estimates astrometric errors. Our constraints on the orbit of β Pic b firmly rule out the possibility of a transit of the planet at 10-σ significance. However, we confirm that the Hill sphere of β Pic b will transit, giving us a rare chance to probe the circumplanetary environment of a young, evolving exoplanet. As a result, we provide an ephemeris for photometric monitoring of the Hill sphere transit event, which will begin at the start of April in 2017 and finish at the end of January in 2018.

79 ASTRONOMY AND ASTROPHYSICS↗

Technical Note: An efficient method for accelerating the spin-up process for process-based biogeochemistry models

To better understand the role of terrestrial ecosystems in the global carbon cycle and their feedbacks to the global climate system,process-based biogeochemistry models need to be improved with respect tomodel parameterization and model structure. To achieve these improvements,the spin-up time for those differential equation-based models needs to be shortened. Here, an algorithm for a fast spin-up was developed by finding the exact solution of a linearized system representing the cyclo-stationary state of a model and implemented in a biogeochemistry model, the Terrestrial Ecosystem Model (TEM). With the new spin-up algorithm, we showed that the model reacheda steady state in less than 10 years of computing time, while the original method requires more than 200 years on average of model run. For the test sites with five different plant functional types, the new method saves over 90 % of the original spin-up time in site-level simulations. In North American simulations, average spin-up time savings for all grid cells is 85 % for either the daily or monthly version of TEM. Thus, the developed spin-up method shall be used for future quantification of carbon dynamics at fine spatial and temporal scales.

58 GEOSCIENCES↗