SEARCH · Search NASA
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.
The design of iterative schemes for sparse matrix computations often leads to constrained polynomial approximation problems on sets in the complex plane. For the case of ellipses, we introduce a new class of complex polynomials which are in general very good approximations to the best polynomials and even optimal in most cases.
Here, we propose a new algorithm for the fast solution of large, sparse, symmetric positive-definite linear systems, spaND (sparsified Nested Dissection). It is based on nested dissection, sparsification, and low-rank compression. After eliminating all interiors at a given level of the elimination tree, the algorithm sparsifies all separators corresponding to the interiors. This operation reduces the size of the separators by eliminating some degrees of freedom but without introducing any fill-in. This is done at the expense of a small and controllable approximation error. The result is an approximate factorization that can be used as an efficient preconditioner. We then perform several numerical experiments to evaluate this algorithm. We demonstrate that a version using orthogonal factorization and block-diagonal scaling takes fewer CG iterations to converge than previous similar algorithms on various kinds of problems. Furthermore, this algorithm is provably guaranteed to never break down and the matrix stays symmetric positive-definite throughout the process. We evaluate the algorithm on some large problems show it exhibits near-linear scaling. The factorization time is roughly $\mathcal{O}$(N), and the number of iterations grows slowly with N.
This paper develops an approximate steady state Fokker Planck theory for energetic electron transport in a spherical laser fusion target. First, we apply the theory to only a small population of electrons, which is specified at the outset. Then, one determines the nonlocal electron energy flux of these electrons at a particular position. These energetic electrons may either come from the tail of the thermal distribution function or be generated by an instability. This paper develops two approximate methods of solution, which we call the “characteristic method” and “sparse eigenfunction.” The former works only in planar geometry and the latter in both planar and spherical geometry. Comparison of the two methods in planar geometry shows that even though the approximations are very different, they give about the same result, increasing their credibility. For the example we have chosen, spherical effects are not important for electrons from the tail of the distribution function but may well be for instability generated electrons, which have much higher energy. Comparing planar to spherical, one finds an additional spherical barrier protecting the fuel. It turns out that the associated fuel preheat a Fokker Planck model predicts is considerably less than that predicted by the Krook models as developed at both NRL and other places.
We investigate the solution of low-rank matrix approximation problems using the truncated singular value decomposition (SVD). For this purpose, we develop and optimize graphics processing unit (GPU) implementations for the randomized SVD and a blocked variant of the Lanczos approach. Our work takes advantage of the fact that the two methods are composed of very similar linear algebra building blocks, which can be assembled using numerical kernels from existing high-performance linear algebra libraries. Furthermore, the experiments with several sparse matrices arising in representative real-world applications and synthetic dense test matrices reveal a performance advantage of the block Lanczos algorithm when targeting the same approximation accuracy.
The Kepler Science Operations Center stores pixel values on approximately six million pixels collected every 30-minutes, as well as data products that are generated as a result of running the Kepler science processing pipeline. The Kepler Database (Kepler DB) management system was created to act as the repository of this information. After one year of ight usage, Kepler DB is managing 3 TiB of data and is expected to grow to over 10 TiB over the course of the mission. Kepler DB is a non-relational, transactional database where data are represented as one dimensional arrays, sparse arrays or binary large objects. We will discuss Kepler DB's APIs, implementation, usage and deployment at the Kepler Science Operations Center.
Here we present the PI-CKL-NN method for parameter estimation in differential equation (DE) models given sparse measurements of the parameters and states. In the proposed approach, the space- or time-dependent parameters are approximated by Karhunen-Loeve (KL) expansions that are conditioned on the parameters’ measurements, and the states are approximated by deep neural networks (DNNs). The unknown weights in the KL expansions and DNNs are found my minimizing the cost function that enforces the measurements of the states the DE constraint. Regularization is achieved by adding the l2 norm of the conditional KL coefficients into the loss function. Our approach assumes that the parameter fields are correlated in space or time and enforces the statistical knowledge (the mean and the covariance function) in addition to the DE constraints and measurements as opposed to the physics-informed neural network (PINN) and other similar physics-informed machine learning methods where only DE constraints and data are used for parameter estimation. We use the PI-CKL-NN method for parameter estimation in an ordinary differential equation with an unknown time-dependent parameter and the one- and two-dimensional partial differential diffusion equations with unknown space-dependent diffusion coefficients. We also demonstrate that PI-CKL-NN is more accurate than the PINN method, especially when the observations of the parameters are very sparse
Monthly composite images from the global coastal zone color scanner (CZCS) data set are used to provide an initial illustration and comparison of seasonal and interannual variability of phytoplankton pigment concentration along the western coasts of South and North America in the Peru Current system (PCS) and California Current system (CCS). The analysis utilizes the entire time series of available data (November 1978 to June 1986) to form a mean annual cycle and an index of interannual variability for a series of both latitudinal and cross-shelf regions within each current system. Within 100 km of the coast, the strongest seasonal cycles in the CCS are in two regions, one between 34 deg and 45 deg N and the second between 24 deg and 29 deg N, each with maximum concentrations (greater than 3.0 mg m(exp-3)) in May-June. Weaker seasonal variability is present north of 45 deg N and in the Southern California Bight region (32 deg N). Within the PCS, in the same 100-km-wide coastal region, highest (greater than 45 deg S) and lowest (less than 20 deg S) latitude regions have a similar seasonal cycle with maximum concentrations (greater than 1.5 mg m(exp -3)) during the austral spring, summer, and fall, matching that evident throughout the CCS. Between these regions, off northern and central Chile, the seasonal maximum occurs during July-August (austral winter), contrary to the influence of upwelling favorable winds. Within the CCS, the dominant feature of interannual variability in the 8-year time series is a strong negative concentration anomaly in 1983, an El Nino year. The relative value of this negative anomaly is strongest off central California and is followed by an even stronger negative anomaly is strongest off central California and is followed by an even stronger negative anomaly in 1984 off Baja, California. In the PCS, strong negative anomalies during the 1982-1983 El Nino period are evident only off the Peruvian coast and are evident there only in the regions 100 km or more from the coast. Although negative anomalies associated with the El Nino were not present at higher latitudes (more than approximately 20 deg S) in the PCS, the extremely sparse sampling weakens our confidence in the results of the interannual analysis in this region. An upper estimate of the systematic winter bias remaining in the global CZCS data after reprocessing with the multiple scattering algorithm is given in the appendix.
An incremental iterative formulation together with the well-known spatially split approximate-factorization algorithm, is presented for solving the large, sparse systems of linear equations that are associated with aerodynamic sensitivity analysis. This formulation is also known as the 'delta' or 'correction' form. For the smaller two dimensional problems, a direct method can be applied to solve these linear equations in either the standard or the incremental form, in which case the two are equivalent. However, iterative methods are needed for larger two-dimensional and three dimensional applications because direct methods require more computer memory than is currently available. Iterative methods for solving these equations in the standard form are generally unsatisfactory due to an ill-conditioned coefficient matrix; this problem is overcome when these equations are cast in the incremental form. The methodology is successfully implemented and tested using an upwind cell-centered finite-volume formulation applied in two dimensions to the thin-layer Navier-Stokes equations for external flow over an airfoil. In three dimensions this methodology is demonstrated with a marching-solution algorithm for the Euler equations to calculate supersonic flow over the High-Speed Civil Transport configuration (HSCT 24E). The sensitivity derivatives obtained with the incremental iterative method from a marching Euler code are used in a design-improvement study of the HSCT configuration that involves thickness. camber, and planform design variables.
An algorithm that performs sparse linear discriminant analysis (Sparse-LDA) finds near-optimal solutions in far less time than the prior art when specialized to binary classification (of 2 classes). Sparse-LDA is a type of feature- or variable- selection problem with numerous applications in statistics, machine learning, computer vision, computational finance, operations research, and bio-informatics. Because of its combinatorial nature, feature- or variable-selection problems are NP-hard or computationally intractable in cases involving more than 30 variables or features. Therefore, one typically seeks approximate solutions by means of greedy search algorithms. The prior Sparse-LDA algorithm was a greedy algorithm that considered the best variable or feature to add/ delete to/ from its subsets in order to maximally discriminate between multiple classes of data. The present algorithm is designed for the special but prevalent case of 2-class or binary classification (e.g. 1 vs. 0, functioning vs. malfunctioning, or change versus no change). The present algorithm provides near-optimal solutions on large real-world datasets having hundreds or even thousands of variables or features (e.g. selecting the fewest wavelength bands in a hyperspectral sensor to do terrain classification) and does so in typical computation times of minutes as compared to days or weeks as taken by the prior art. Sparse LDA requires solving generalized eigenvalue problems for a large number of variable subsets (represented by the submatrices of the input within-class and between-class covariance matrices). In the general (fullrank) case, the amount of computation scales at least cubically with the number of variables and thus the size of the problems that can be solved is limited accordingly. However, in binary classification, the principal eigenvalues can be found using a special analytic formula, without resorting to costly iterative techniques. The present algorithm exploits this analytic form along with the inherent sequential nature of greedy search itself. Together this enables the use of highly-efficient partitioned-matrix-inverse techniques that result in large speedups of computation in both the forward-selection and backward-elimination stages of greedy algorithms in general.
This software combines nonlinear-manifold reduced order models (NM-ROMs) with domain decomposition (DD) techniques. NM-ROMs, which utilize a shallow, sparse autoencoder trained with full order model (FOM) snapshot data, approximate the FOM state on a nonlinear manifold. These models offer advantages over linear-subspace ROMs (LS-ROMs) particularly in scenarios with slowly decaying Kolmogorov n-width. However, the training of NM-ROMs involves a number of parameters that scale with the size of the FOM, and storing high-dimensional FOM snapshots can significantly increase the cost of ROM training for extreme-scale problems. To mitigate these costs, the software employs DD to partition the FOM into smaller subdomains, computes NM-ROMs for each, and then integrates these to form a global NM-ROM. This strategy offers multiple benefits: it enables parallel training of subdomain NM-ROMs, reduces the number of parameters needed, decreases the dimensional requirements of subdomain FOM training data, and allows for customization to the unique characteristics of each FOM subdomain. The use of a shallow, sparse autoencoder architecture in each subdomain NM-ROM facilitates the application of hyper-reduction (HR), simplifying the nonlinear complexities and enhancing computational speed. This software marks the inaugural application of NM-ROM combined with HR to a DD problem. It features an algebraic DD reformulation of the FOM, training of NM-ROMs with HR for each subdomain, and employs a sequential quadratic programming (SQP) solver for the evaluation of the coupled global NMROM. The effectiveness of the DD NM-ROM with HR is numerically demonstrated on the 2D steady-state Burgers' equation, showing an order of magnitude improvement in accuracy over the DD LS-ROM with HR.
Light-sheet microscopy has revolutionized bioimaging by enabling approximately an order of magnitude reduction in specimen irradiance compared to confocal imaging. Here, we introduce a light-sheet imaging system that enables an additional order of magnitude reduction in specimen irradiance by operating at the Poisson limit. To operate at this limit, we integrated classical illumination with single-photon detection and wavelet-based image reconstruction. This integration enabled brightness quantification and object recognition from fewer than one detected photon per image pixel, corresponding to more than 10-fold lower irradiance levels than modern systems. We demonstrate how such photon-sparse imaging can eradicate photobleaching and enable both dim and bright object imaging, thus, further enhancing the related gains of light-sheet microscopy.
In this study involving advanced fluid flow codes, an incremental iterative formulation (also known as the delta or correction form) together with the well-known spatially-split approximate factorization algorithm, is presented for solving the very large sparse systems of linear equations which are associated with aerodynamic sensitivity analysis. For smaller 2D problems, a direct method can be applied to solve these linear equations in either the standard or the incremental form, in which case the two are equivalent. Iterative methods are needed for larger 2D and future 3D applications, however, because direct methods require much more computer memory than is currently available. Iterative methods for solving these equations in the standard form are generally unsatisfactory due to an ill-conditioning of the coefficient matrix; this problem can be overcome when these equations are cast in the incremental form. These and other benefits are discussed. The methodology is successfully implemented and tested in 2D using an upwind, cell-centered, finite volume formulation applied to the thin-layer Navier-Stokes equations. Results are presented for two sample airfoil problems: (1) subsonic low Reynolds number laminar flow; and (2) transonic high Reynolds number turbulent flow.
Successive approximations formed in Fourier space. Algorithm extracts periodic signals from sparse, irregularly sampled sets of measurement data. Pertains to data processed via fast Fourier transforms (FFTs). Data represents signal components with initially unknown frequencies spanning large spectral range and includes frequencies not integer multiples of minimum FFT frequency.
We present a method for dimensionally adaptive sparse trigonometric interpolation of multidimensional periodic functions belonging to a smoothness class of finite order. This method targets applications where periodicity must be preserved and the precise anisotropy is not known a priori. To the authors' knowledge, this is the first instance of a dimensionally adaptive sparse interpolation algorithm that uses a trigonometric interpolation basis. The motivating application behind this work is the adaptive approximation of a multi-input model for a molecular potential energy surface (PES) where each input represents an angle of rotation. Our method is based on an anisotropic quasi-optimal estimate for the decay rate of the Fourier coefficients of the model; a least-squares fit to the coefficients of the interpolant is used to estimate the anisotropy. Thus, our adaptive approximation strategy begins with a coarse isotropic interpolant, which is gradually refined using the estimated anisotropic rates. The procedure takes several iterations where ever-more accurate interpolants are used to generate ever-improving anisotropy rates. We present several numerical examples of our algorithm where the adaptive procedure successfully recovers the theoretical “best” convergence rate, including an application to a periodic PES approximation. An open-source implementation of our algorithm resides in the Tasmanian UQ library developed at Oak Ridge National Laboratory.
A simplified procedure is presented for analysis of ducts with discretely nonuniform properties. The analysis uses basis functions as the generalized coordinates. The duct eigenfunctions are approximated by finite series of these functions. The emphasis is on solution of the resulting large sparse set of linear equations. Characteristics of sparse matrix algorithms are outlined and some criteria for application are established. Analogies with structural methods are used to illustrate variations which can increase efficiency in generating values for design optimization routines. The effects of basis function selection, number of eigenfunctions and identification and ordering of equations on the sparsity and solution stability are included.
This is the AmeriFlux Management Project (AMP) created FLUXNET-1F version of the carbon flux data for the site US-Me2 Metolius mature ponderosa pine. This is the FLUXNET version of the carbon flux data for the site US-Me2 Metolius mature ponderosa pine produced by applying the standard ONEFlux (1F) software. Site Description - Site Description before Fire (January, 2002 - August, 2020): The mean stand age is 71 years old and the stand age of the oldest 10% of trees is about 108 years old. This site is one of the Metolius core cluster sites with different age and disturbance classes and part of the AmeriFlux network. The overstory is almost exclusively composed of ponderosa pine trees (Pinus ponderosa Doug. Ex P. Laws) with a few scattered incense cedars (Calocedrus decurrens (Torr.) Florin) and has a peak leaf area index (LAI) of 2.1 m2 m-2. Tree height is relatively homogeneous at about 18 m, and the mean tree density is approximately 339 trees ha-1 (Irvine et al., 2008). The understory is sparse with an LAI of 0.2 m2 m-2 and primarily composed of bitterbrush (Purshia tridentata (Push) DC.) and greenleaf manzanita (Arctostaphylos patula Greene). Soils at the site are sandy (69%/24%/7% sand/silt/clay at 0–0.2 m depth and 66%/27%/7% at 0.2–0.5 m depth, and 54%/ 35%/11% at 0.5–1.0 m depth), freely draining with a soil depth of approximately 1.5 m (Irvine et al., 2008; Law et al., 2001b; Schwarz et al., 2004). Green Ridge Fire: On August 20, 2020, the Green Ridge Fire burned through Us-Me2. The fire was ignited by lightning on August 16th, and grew rapidly to the east over the first few days driven by strong, downslope (westerly) afternoon winds. Fire behavior and observed fire effects were highly heterogeneous due to the localized wind pattern carrying the flaming head of the fire forward, and the efforts being made by suppression resources to contain the fire. The site experienced the full range of fire effects, from <1 m high surface fire that charred litter and duff and only consumed shrubs and herbaceous material to full tree (>15 m) crown fire that consumed 100% of needles, small limbs, and surface fuels at high intensity, leaving only ash and bare soil post-fire. Salvage Logging: From late March to late April 2021, salvage logging by the landowner occurred at the site. Almost all trees within the flux footprint were logged except a small area with lower burn severity, where sap flow and automatic soil respiration measurements are continued since the fire. In August 2022 the primary flux system was moved from the top of the damaged tall tower to a nearby shorter tower centering on the salvage/regenerating footprint.
The emergent practical applicability of the Quantum Approximate Optimization Algorithm (QAOA) for approximate combinatorial optimization is a subject of considerable interest. One of the primary limitations of QAOA is the task of finding a set of good parameters, which is usually done using a variational optimization loop. Parameter transfer, or parameter concentration, is a phenomenon where QAOA angles trained on problem instances that are self-similar tend to perform well for other problem instances from that similar class. This suggests a potentially highly efficient and scalable non-variational learning method for QAOA angle finding. In this work, we systematically study QAOA parameter transferability from small problem sizes (16 and 27 decision variables) onto large problem instances (up to 156 qubits) for heavy-hex graph Ising models with geometrically local higher order terms using the Julia based QAOA simulation tool \texttt{JuliQAOA} to perform classical angle finding for up to $49$ QAOA layers ($p$). Parameter transfer of the fixed angles is validated using a combination of full statevector, Projected Entangled Pair States (PEPS), Matrix Product State (MPS), and LOWESA numerical simulations. We find that the QAOA parameter transfer from single instances applied to other (unseen) problem instances does not in general provide monotonically improving performance as a function of $p$ - there are many cases where the performance temporarily decreases as a function of $p$ - but despite this the transferred angles have a general trend of improved expectation value as the QAOA depth increases, in many cases converging close to the true ground-state energy of the $100+$ qubit instances. We also sample the hardware-compatible Ising models using the ensemble of transfer-learned QAOA parameters on several superconducting qubit IBM Quantum processors with 127, 133, and 156 qubits. We find continuous solution quality improvement of the hardware-compatible QAOA circuits run on the IBM NISQ processors up to $p=5$ on \texttt{ibm\_fez}, up to $p=9$ on \texttt{ibm\_torino}, and up to $p=10$ on \texttt{ibm\_pittsburgh}.