SEARCH · Engineering Papers

Results for “matrix factorization”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15

All sources NASA NTRS DOE OSTI

NASA NTRS2010

Partial Least Squares and Neural Networks for Quantitative Calibration of Laser-induced Breakdown Spectroscopy (LIBs) of Geologic Samples

The ChemCam instrument [1] on the Mars Science Laboratory (MSL) rover will be used to obtain the chemical composition of surface targets within 7 m of the rover using Laser Induced Breakdown Spectroscopy (LIBS). ChemCam analyzes atomic emission spectra (240-800 nm) from a plasma created by a pulsed Nd:KGW 1067 nm laser. The LIBS spectra can be used in a semiquantitative way to rapidly classify targets (e.g., basalt, andesite, carbonate, sulfate, etc.) and in a quantitative way to estimate their major and minor element chemical compositions. Quantitative chemical analysis from LIBS spectra is complicated by a number of factors, including chemical matrix effects [2]. Recent work has shown promising results using multivariate techniques such as partial least squares (PLS) regression and artificial neural networks (ANN) to predict elemental abundances in samples [e.g. 2-6]. To develop, refine, and evaluate analysis schemes for LIBS spectra of geologic materials, we collected spectra of a diverse set of well-characterized natural geologic samples and are comparing the predictive abilities of PLS, cascade correlation ANN (CC-ANN) and multilayer perceptron ANN (MLP-ANN) analysis procedures.

Anderson, R. B.↗

NASA NTRS2000

A Study of Influencing Factors on the Tensile Response of a Titanium Matrix Composite With Weak Interfacial Bonding

The generalized method of cells micromechanics model is utilized to analyze the tensile stress-strain response of a representative titanium matrix composite with weak interfacial bonding. The fiber/matrix interface is modeled through application of a displacement discontinuity between the fiber and matrix once a critical debonding stress has been exceeded. Unidirectional composites with loading parallel and perpendicular to the fibers are examined, as well as a cross-ply laminate. For each of the laminates studied, analytically obtained results are compared to experimental data. The application of residual stresses through a cool-down process was found to have a significant effect on the tensile response. For the unidirectional laminate with loading applied perpendicular to the fibers, fiber packing and fiber shape were shown to have a significant effect on the predicted tensile response. Furthermore, the interface was characterized through the use of semi-emperical parameters including an interfacial compliance and a "debond stress;" defined as the stress level across the interface which activates fiber/matrix debonding. The results in this paper demonstrate that if architectural factors are correctly accounted for and the interface is appropriately characterized, the macro-level composite behavior can be correctly predicted without modifying any of the fiber or matrix constituent properties.

Goldberg, Robert K.↗

NASA NTRS1987

Inverse of polynomial matrices in the irreducible form

An algorithm is developed for finding the inverse of polynomial matrices in the irreducible form. The computational method involves the use of the left (right) matrix division method and the determination of linearly dependent vectors of the remainders. The obtained transfer function matrix has no nontrivial common factor between the elements of the numerator polynomial matrix and the denominator polynomial.

Chang, Fan R.↗

NASA NTRS1994

Method and apparatus for the collection, storage, and real time analysis of blood and other bodily fluids

The present invention provides a simple, portable, relatively inexpensive apparatus for accurately and efficiently collecting, separating, testing, and even storing between about 1-20 ml, preferably about 1-10 ml, of blood or other bodily fluid in situ. The apparatus includes a collection chamber bounded on its sides by an opening in a sheet of material, preferably clear plastic, abutting a filter card. The filter card is made of fibrous material, preferably less than about a millimeter thick, having an average pore size of less than about 3 microns. Preferably, the fibers are glass and the fibrous material has an average pore size of about 1 micron. The fibrous material is treated with a carbohydrate/protein mixture which contains between about 1-40 percent wt/vol carbohydrate and about 0.1-15 percent wt/vol nonspecific protein, preferably between about 10-20 percent carbohydrate and about 5-8 percent protein. A preferred carbohydrate/protein mixture comprises about 10 percent mannitol and about 6 percent albumin. The blood or other fluid moves through the filter card by capillary action aided by an absorbent matrix with a high Klemm factor which abuts the filter card. The absorbent matrix and/or filter card can be treated with a wide spectrum of test reagents. The speed, cleanliness, and efficiency of the separation process can be altered by: (a) changing the absolute concentration of the carbohydrate/protein mixture; (b) applying positive or negative pressure to one side of the filter; and/or (c) varying the relative density and pore size of the filter card and absorbent matrix.

Whitson, Peggy A.↗

NASA NTRS1983

Damping mechanisms in fiber reinforced laminates

Low strain damping in fiber reinforced composite materials is due to material loss factors in both fibers and matrix materials. The high modulus of the fibers makes up for the low damping factor. Strains high enough to cause transverse layer cracking in laminates with organic matrix materials causes a large permanent increase in the damping factor. This increase is not due to the transverse cracks but rather to short microcracks in the high shear strain regions. Other damping mechanisms at high strains are also discussed.

Plunkett, R.↗

NASA NTRS1991

Extracellular matrix and cell shape: potential control points for inhibition of angiogenesis

Capillary endothelial (CE) cells require two extracellular signals in order to switch from quiescence to growth and back to differentiation during angiogenesis: soluble angiogenic factors and insoluble extracellular matrix (ECM) molecules. Soluble endothelial mitogens, such as basic fibroblast growth factor (FGF), act over large distances to trigger capillary growth, whereas ECM molecules act locally to modulate cell responsiveness to these soluble cues. Recent studies reveal that ECM molecules regulate CE cell growth and differentiation by modulating cell shape and by activating intracellular chemical signaling pathways inside the cell. Recognition of the importance of ECM and cell shape during capillary morphogenesis has led to the identification of a series of new angiogenesis inhibitors. Elucidation of the molecular mechanism of capillary regulation may result in development of even more potent angiogenesis modulators in the future.

Non-NASA Center↗

NASA NTRS1974

Matrix isolation technique for the study of some factors affecting the partitioning of trace elements

The factors that affect the preferred positions of cations in ionic solid solutions were investigated utilizing vibrational spectroscopy. Solid solutions of the sulfate and chromate ions codoped with La(+3) and Ca(+2) in a KBr host lattice were examined as a function of the polyvalent cation concentration. The cation-anion pairing process was found to be random for Ca(+2), whereas the formation of La(+3)-SO4(-2) ion pairs with a C2 sub v bonding geometry is highly preferential to any type of La(+3)-CrO4(-2) ion pair formation. The relative populations of ion pair site configurations are discussed in terms of an energy-entropy competition model which can be applied to the partition of trace elements during magmatic processes.

Grzybowski, J. M.↗

NASA NTRS1991

Unified formulation of dynamics for serial rigid multibody systems

There has been a growing interest in the development of new and efficient algorithms for multibody dynamics in recent years. Serial rigid multibody systems form the basic subcomponents of general multibody systems, and a variety of algorithms to solve the serial chain forward dynamics problem have been proposed. In this paper, the economy of representation and analysis tools provided by the spatial operator algebra are used to clarify the inherent structure of these algorithms, to identify those that are similar, and to study the relationships among the ones that are distinct. For the purposes of this study, the algorithms are categorized into three classes: algorithms that require the explicit computation of the mass matrix, algorithms that are completely recursive in nature, and algorithms of intermediate complexity. In addition, alternative factorizations for the mass matrix and closed form expressions for its inverse are derived. These results provide a unifying perspective, within which these diverse dynamics algorithms arise naturally as a consequence of a progressive exploitation of the structure of the mass matrix.

Jain, Abhinandan↗

NASA NTRS1992

A performance study of sparse Cholesky factorization on INTEL iPSC/860

The problem of Cholesky factorization of a sparse matrix has been very well investigated on sequential machines. A number of efficient codes exist for factorizing large unstructured sparse matrices. However, there is a lack of such efficient codes on parallel machines in general, and distributed machines in particular. Some of the issues that are critical to the implementation of sparse Cholesky factorization on a distributed memory parallel machine are ordering, partitioning and mapping, load balancing, and ordering of various tasks within a processor. Here, we focus on the effect of various partitioning schemes on the performance of sparse Cholesky factorization on the Intel iPSC/860. Also, a new partitioning heuristic for structured as well as unstructured sparse matrices is proposed, and its performance is compared with other schemes.

Zubair, M.↗

NASA NTRS1989

Recursive form of the eigensystem realization algorithm for system identification

An algorithm is developed for recursively calculating the minimum realization of a linear system from sampled impulse response data. The Gram-Schmidt orthonormalization technique is used to generate an orthonormal basis for factorization of the data matrix. The system matrix thus identified is in upper Hessenberg form, which has advantages for the identification of modal parameters including damping coefficients, frequencies, mode shapes, and modal participation factors. It also has the property that once an element of the system matrix is computed, it is never altered as the dimension of the model is increased in the recursive process. Numerical examples are presented for comparison of the recursive and nonrecursive forms of the eigensystem realization algorithm.

Longman, Richard W.↗

DOE OSTI2023

Measurement of differential distributions of B → D*$\ell\bar{v}_{\ell}$ and implications on | V cb |

We present a measurement of the differential shapes of exclusive B → D*$\ell\bar{v}_{\ell}$ (B = B - , $\bar{B}$ 0 and $\ell$ = e, μ) decays with hadronic tag-side reconstruction for the full 711 fb -1 Belle dataset. We extract the Caprini-Lellouch-Neubert (CLN) and Boyd-Grinstein-Lebed (BGL) form factor parameters and use an external input for the absolute branching fractions to determine the Cabibbo-Kobayashi-Maskawa matrix element and find |V cb | CLN = (40.2 ± 0.9) x 10 -3 and |V cb | BGL = (40.7 ± 1.0) x 10 -3 with the zero-recoil lattice QCD point $\mathcal{F}$(1) = 0.906 ± 0.013. We also perform a study of the impact of beyond zero-recoil lattice QCD calculations on the |V cb | determinations. Additionally, we present the lepton-flavor universality ratio R eμ = $\mathcal{B}$(B → D*e$\bar{v}$ e )/$\mathcal{B}$(B → D*μ$\bar{v}$ μ ) = 0.993 ± 0.023 ± 0.023, the electron and muon forwardbackward asymmetry and their difference ΔA FB = 0.028 ± 0.028 ± 0.008, and the electron and muon D* longitudinal polarization fraction and their difference ΔF$^{D*}_L$ = 0.030 ± 0.025 ± 0.007. The uncertainties quoted correspond to the statistical and systematic uncertainties, respectively.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2008

Direct Solve of Electrically Large Integral Equations for Problem Sizes to 1M Unknowns

Matrix methods for solving integral equations via direct solve LU factorization are presently limited to weeks to months of very expensive supercomputer time for problems sizes of several hundred thousand unknowns. This report presents matrix LU factor solutions for electromagnetic scattering problems for problem sizes to one million unknowns with thousands of right hand sides that run in mere days on PC level hardware. This EM solution is accomplished by utilizing the numerical low rank nature of spatially blocked unknowns using the Adaptive Cross Approximation for compressing the rank deficient blocks of the system Z matrix, the L and U factors, the right hand side forcing function and the final current solution. This compressed matrix solution is applied to a frequency domain EM solution of Maxwell's equations using standard Method of Moments approach. Compressed matrix storage and operations count leads to orders of magnitude reduction in memory and run time.

Shaeffer, John↗

DOE OSTI2020

Quantum Fourier transform revisited

Summary The fast Fourier transform (FFT) is one of the most successful numerical algorithms of the 20th century and has found numerous applications in many branches of computational science and engineering. The FFT algorithm can be derived from a particular matrix decomposition of the discrete Fourier transform (DFT) matrix. In this paper, we show that the quantum Fourier transform (QFT) can be derived by further decomposing the diagonal factors of the FFT matrix decomposition into products of matrices with Kronecker product structure. We analyze the implication of this Kronecker product structure on the discrete Fourier transform of rank‐1 tensors on a classical computer. We also explain why such a structure can take advantage of an important quantum computer feature that enables the QFT algorithm to attain an exponential speedup on a quantum computer over the FFT algorithm on a classical computer. Further, the connection between the matrix decomposition of the DFT matrix and a quantum circuit is made. We also discuss a natural extension of a radix‐2 QFT decomposition to a radix‐ d QFT decomposition. No prior knowledge of quantum computing is required to understand what is presented in this paper. Yet, we believe this paper may help readers to gain some rudimentary understanding of the nature of quantum computing from a matrix computation point of view.

Camps, Daan↗

NASA NTRS1992

Extracellular matrix as a solid-state regulator in angiogenesis: identification of new targets for anti-cancer therapy

Angiogenesis, the growth of blood capillaries, is regulated by soluble growth factors and insoluble extracellular matrix (ECM) molecules. Soluble angiogenic mitogens act over large distances to initiate capillary growth whereas changes in ECM govern whether individual cells will grow, differentiate, or involute in response to these stimuli in the local tissue microenvironment. Analysis of this local control mechanism has revealed that ECM molecules switch capillary endothelial cells between differentiation and growth by both binding specific transmembrane integrin receptors and physically resisting cell-generated mechanical loads that are applied to these receptors. Control of capillary endothelial cell form and function therefore may be exerted by altering the mechanical properties of the ECM as well as its chemical composition. Understanding of this mechanochemical control mechanism has led to the development of new angiogenesis inhibitors that may be useful for the treatment of cancer.

Review↗

NASA NTRS1977

On the factorization and fitting of molecular scattering information

The reported analysis is based on the factored IOS T-matrix. It is shown that line shape measurements may be used over a range of temperatures to evaluate inelastic scattering cross sections. Basic factorization or parameterization relations are derived by considering the wavefunction equations. The parameterization of cross sections is considered, taking into account the differential scattering amplitude and cross section, integral cross sections, phenomenological cross sections for general relaxation processes, and viscosity and diffusion cross sections. Thermal averages and rates are discussed, giving attention to integral cross sections and rates, and general phenomenological cross sections. The results of computational studies are also presented.

Goldflam, R.↗

NASA NTRS1993

Preconditioned conjugate-gradient methods for low-speed flow calculations

An investigation is conducted into the viability of using a generalized Conjugate Gradient-like method as an iterative solver to obtain steady-state solutions of very low-speed fluid flow problems. Low-speed flow at Mach 0.1 over a backward-facing step is chosen as a representative test problem. The unsteady form of the two dimensional, compressible Navier-Stokes equations is integrated in time using discrete time-steps. The Navier-Stokes equations are cast in an implicit, upwind finite-volume, flux split formulation. The new iterative solver is used to solve a linear system of equations at each step of the time-integration. Preconditioning techniques are used with the new solver to enhance the stability and convergence rate of the solver and are found to be critical to the overall success of the solver. A study of various preconditioners reveals that a preconditioner based on the Lower-Upper Successive Symmetric Over-Relaxation iterative scheme is more efficient than a preconditioner based on Incomplete L-U factorizations of the iteration matrix. The performance of the new preconditioned solver is compared with a conventional Line Gauss-Seidel Relaxation (LGSR) solver. Overall speed-up factors of 28 (in terms of global time-steps required to converge to a steady-state solution) and 20 (in terms of total CPU time on one processor of a CRAY-YMP) are found in favor of the new preconditioned solver, when compared with the LGSR solver.

Ajmani, Kumud↗

NASA NTRS1993

Preconditioned Conjugate Gradient methods for low speed flow calculations

An investigation is conducted into the viability of using a generalized Conjugate Gradient-like method as an iterative solver to obtain steady-state solutions of very low-speed fluid flow problems. Low-speed flow at Mach 0.1 over a backward-facing step is chosen as a representative test problem. The unsteady form of the two dimensional, compressible Navier-Stokes equations are integrated in time using discrete time-steps. The Navier-Stokes equations are cast in an implicit, upwind finite-volume, flux split formulation. The new iterative solver is used to solve a linear system of equations at each step of the time-integration. Preconditioning techniques are used with the new solver to enhance the stability and the convergence rate of the solver and are found to be critical to the overall success of the solver. A study of various preconditioners reveals that a preconditioner based on the lower-upper (L-U)-successive symmetric over-relaxation iterative scheme is more efficient than a preconditioner based on incomplete L-U factorizations of the iteration matrix. The performance of the new preconditioned solver is compared with a conventional line Gauss-Seidel relaxation (LGSR) solver. Overall speed-up factors of 28 (in terms of global time-steps required to converge to a steady-state solution) and 20 (in terms of total CPU time on one processor of a CRAY-YMP) are found in favor of the new preconditioned solver, when compared with the LGSR solver.

Ajmani, Kumud↗

DOE OSTI2022

HyKKT: a hybrid direct-iterative method for solving KKT linear systems

Here, we propose a solution strategy for the large indefinite linear systems arising in interior methods for nonlinear optimization. The method is suitable for implementation on hardware accelerators such as graphical processing units (GPUs). The current gold standard for sparse indefinite systems is the LBLT factorization where L is a lower triangular matrix and B is 1×1 or 2×2 block diagonal. However, this requires pivoting, which substantially increases communication cost and degrades performance on GPUs. Our approach solves a large indefinite system by solving multiple smaller positive definite systems, using an iterative solver on the Schur complement and an inner direct solve (via Cholesky factorization) within each iteration. Cholesky is stable without pivoting, thereby reducing communication and allowing reuse of the symbolic factorization. We demonstrate the practicality of our approach on large optimal power flow problems and show that it can efficiently utilize GPUs and outperform LBL T factorization of the full system.

97 MATHEMATICS AND COMPUTING↗