Search NASA⌕ Search

SEARCH · Search NASA

Results for “projected gradient”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Linear ion-scale microstability analysis of high and low-collisionality NSTX discharges and NSTX-U projections

Linear gyrokinetic simulations were conducted to investigate ion-gyroradius-scale micro-instability predictions for high-beta NSTX discharges and NSTX-U projections that span over an order of magnitude variation in collisionality. A complex mix of microtearing modes and hybrid trapped electron modes/kinetic ballooning modes (TEM/KBM) is predicted for all experimental or projected conditions. Ion temperature gradient (ITG) instabilities are typically stable in the NSTX discharges investigated, consistent with the observed neoclassical ion thermal transport. ITG thresholds inferred from the simulations are typically much higher than the experimental NSTX gradients, as well as the projected gradients in the NSTX-U scenario, which assumed ion temperatures limited by neoclassical transport only. The analysis suggests ITG instabilities are unlikely to contribute significant anomalous thermal losses in high-beta, lower collisionality NSTX-U scenarios. On the other hand, the NSTX experimental profiles and NSTX-U projections are predicted to be very close to the predicted onset of unstable KBM at most radii investigated. The proximity of the various discharges to the KBM instability threshold implies it may play an important role in setting profile shapes and limiting global energy confinement. Here, it remains to be understood and predicted how KBM contributes to multi-channel transport (thermal and particle transport, for both ions and electrons) in a way that is consistent with experimental inferences.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Uncertainty quantification of graph convolution neural network models of evolving processes

The application of neural network models to scientific machine learning tasks has proliferated in recent years. In particular, neural networks have proved to be adept at modeling processes with spatial–temporal complexity. Nevertheless, these highly parameterized models have garnered skepticism in their ability to produce outputs with quantified error bounds over the regimes of interest. Hence there is a need to find uncertainty quantification methods that are suitable for neural networks. In this work we present comparisons of the parametric uncertainty quantification of neural networks modeling complex spatial–temporal processes with Hamiltonian Monte Carlo and Stein variational gradient descent and its projected variant. Specifically we apply these methods to graph convolutional neural network models of evolving systems modeled with recurrent neural network and neural ordinary differential equations architectures. We show that Stein variational inference is a viable alternative to Monte Carlo methods with some clear advantages for complex neural network models. For our exemplars, Stein variational interference gave similar pushed forward uncertainty profiles through time compared to Hamiltonian Monte Carlo, albeit with generally more generous variance. As a result, projected Stein variational gradient descent also produced similar uncertainty profiles to the non-projected counterpart, but large reductions in the active weight space were confounded by the stability of the neural network predictions and the convoluted likelihood landscape.

36 MATERIALS SCIENCE↗
NASA NTRS1980

ACCESS 3. Approximation concepts code for efficient structural synthesis: User's guide

A user's guide is presented for ACCESS-3, a research oriented program which combines dual methods and a collection of approximation concepts to achieve excellent efficiency in structural synthesis. The finite element method is used for structural analysis and dual algorithms of mathematical programming are applied in the design optimization procedure. This program retains all of the ACCESS-2 capabilities and the data preparation formats are fully compatible. Four distinct optimizer options were added: interior point penalty function method (NEWSUMT); second order primal projection method (PRIMAL2); second order Newton-type dual method (DUAL2); and first order gradient projection-type dual method (DUAL1). A pure discrete and mixed continuous-discrete design variable capability, and zero order approximation of the stress constraints are also included.

Fleury, C.↗
NASA NTRS1994

Local performance optimization for a class of redundant eight-degree-of-freedom manipulators

Local performance optimization for joint limit avoidance and manipulability maximization (singularity avoidance) is obtained by using the Jacobian matrix pseudoinverse and by projecting the gradient of an objective function into the Jacobian null space. Real-time redundancy optimization control is achieved for an eight-joint redundant manipulator having a three-axis spherical shoulder, a single elbow joint, and a four-axis spherical wrist. Symbolic solutions are used for both full-Jacobian and wrist-partitioned pseudoinverses, partitioned null-space projection matrices, and all objective function gradients. A kinematic limitation of this class of manipulators and the limitation's effect on redundancy resolution are discussed. Results obtained with graphical simulation are presented to demonstrate the effectiveness of local redundant manipulator performance optimization. Actual hardware experiments performed to verify the simulated results are also discussed. A major result is that the partitioned solution is desirable because of low computation requirements. The partitioned solution is suboptimal compared with the full solution because translational and rotational terms are optimized separately; however, the results show that the difference is not significant. Singularity analysis reveals that no algorithmic singularities exist for the partitioned solution. The partitioned and full solutions share the same physical manipulator singular conditions. When compared with the full solution, the partitioned solution is shown to be ill-conditioned in smaller neighborhoods of the shared singularities.

Williams, Robert L., II↗
NASA NTRS1982

Nonlinear optimization with linear constraints using a projection method

Nonlinear optimization problems that are encountered in science and industry are examined. A method of projecting the gradient vector onto a set of linear contraints is developed, and a program that uses this method is presented. The algorithm that generates this projection matrix is based on the Gram-Schmidt method and overcomes some of the objections to the Rosen projection method.

Fox, T.↗
DOE OSTI2021

Increasing temperatures reduce invertebrate abundance and slow decomposition

Decomposition is an essential ecosystem service driven by interacting biotic and abiotic factors. Increasing temperatures due to climate change can affect soil moisture, soil fauna, and subsequently, decomposition. Understanding how projected climate change scenarios will affect decomposition is of vital importance for predicting nutrient cycling and ecosystem health. In this study, we experimentally addressed the question of how the early stages of decomposition would vary along a gradient of projected climate change scenarios. Given the importance of biodiversity for ecosystem service provisioning, we measured the effect of invertebrate exclusion on red maple ( Acer rubrum ) leaf litter breakdown along a temperature gradient using litterbags in warming chambers over a period of five weeks. Leaf litter decomposed more slowly in the warmer chambers and in the litterbag treatment that minimized invertebrate access. Moreover, increasing air temperature reduced invertebrate abundance and richness, and altered the community composition, independent of exclusion treatment. Using structural equation models, we were able to disentangle the effects of average air temperature on leaf litter loss, finding a direct negative effect of warming on the early stages of decomposition, independent of invertebrate abundance. This result indicates that not only can climate change affect the invertebrate community, but may also directly influence how the remaining organisms interact with their environment and their effectiveness at provisioning ecosystem services. Overall, our study highlights the role of biodiversity in maintaining ecosystem services and contributes to our understanding of how climate change could disrupt nutrient cycling.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Multi‐Material Gradient Printing Using Meniscus‐enabled Projection Stereolithography (MAPS)

Light‐based additive manufacturing methods are widely used to print high‐resolution 3D structures for applications in tissue engineering, soft robotics, photonics, and microfluidics, among others. Despite this progress, multi‐material printing with these methods remains challenging due to constraints associated with hardware modifications, control systems, cross‐contamination, waste, and resin properties. Here, a new printing platform coined Meniscus‐enabled Projection Stereolithography (MAPS) is reported, a vat‐free method that relies on generating and maintaining a resin meniscus between a crosslinked structure and bottom window to print lateral, vertical, discrete, or gradient multi‐material 3D structures with no waste and user‐defined mixing between layers. MAPS is compatible with a wide range of resins shown and can print complex multi‐material 3D structures without requiring specialized hardware, software, or complex washing protocols. MAPS's ability to print structures with microscale variations in mechanical stiffness, opacity, surface energy, cell densities, and magnetic properties provides a generic method to make advanced materials for a broad range of applications.

bioprinting↗
DOE OSTI2016

A 10-node composite tetrahedral finite element for solid mechanics

We present a reformulation of the composite tetrahedral finite element first introduced by Thoutireddy et al. By choosing a different numerical integration scheme, we obtain an element that is more accurate than the one proposed in the original formulation. We also show that in the context of Lagrangian approaches, the gradient and projection operators derived from the element reformulation admit fully analytic expressions, which offer a significant improvement in terms of accuracy and computational expense. For plasticity applications, a mean-dilatation approach on top of the underlying Hu–Washizu variational principle proves effective for the representation of isochoric deformations. The performance of the reformulated element is shown by hyperelastic and inelastic calculations.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Accelerated fast iterative shrinkage thresholding algorithms for sparsity-regularized cone-beam CT image reconstruction

Purpose: The development of iterative image reconstruction algorithms for cone-beam computed tomography (CBCT) remains an active and important research area. Even with hardware acceleration, the overwhelming majority of the available 3D iterative algorithms that implement nonsmooth regularizers remain computationally burdensome and have not been translated for routine use in time-sensitive applications such as image-guided radiation therapy (IGRT). In this work, two variants of the fast iterative shrinkage thresholding algorithm (FISTA) are proposed and investigated for accelerated iterative image reconstruction in CBCT. Methods: Algorithm acceleration was achieved by replacing the original gradient-descent step in the FISTAs by a subproblem that is solved by use of the ordered subset simultaneous algebraic reconstruction technique (OS-SART). Due to the preconditioning matrix adopted in the OS-SART method, two new weighted proximal problems were introduced and corresponding fast gradient projection-type algorithms were developed for solving them. We also provided efficient numerical implementations of the proposed algorithms that exploit the massive data parallelism of multiple graphics processing units. Results: The improved rates of convergence of the proposed algorithms were quantified in computer-simulation studies and by use of clinical projection data corresponding to an IGRT study. The accelerated FISTAs were shown to possess dramatically improved convergence properties as compared to the standard FISTAs. For example, the number of iterations to achieve a specified reconstruction error could be reduced by an order of magnitude. Volumetric images reconstructed from clinical data were produced in under 4 min. Conclusions: The FISTA achieves a quadratic convergence rate and can therefore potentially reduce the number of iterations required to produce an image of a specified image quality as compared to first-order methods. We have proposed and investigated accelerated FISTAs for use with two nonsmooth penalty functions that will lead to further reductions in image reconstruction times while preserving image quality. Moreover, with the help of a mixed sparsity-regularization, better preservation of soft-tissue structures can be potentially obtained. The algorithms were systematically evaluated by use of computer-simulated and clinical data sets.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-G-201-11: Exploring the Upper Limits of Dose Sculpting Capacity of the Novel Direction Modulated Brachytherapy (DMBT) Tandem Applicator

Purpose: To explore and quantify the upper limits in dose sculpting capacity of the novel direction modulated brachytherapy (DMBT) tandem applicator compared with conventional tandem design for {sup 192}Ir-based HDR planning. Methods: The proposed DMBT tandem applicator is designed for image-guided adaptive brachytherapy (IGABT), especially MRI, of cervical cancer. It has 6 peripheral holes of 1.3-mm width, grooved along a 5.4-mm diameter nonmagnetic tungsten alloy rod of density 18.0 g/cc, capable of generating directional dose profiles – leading to enhanced dose sculpting capacity through inverse planning. The external dimensions are identical to that of conventional tandem design to ensure clinical compatibility. To explore the expansive dose sculpting capacity, we constructed a hypothetical circular target with 20-mm radius and positioned the DMBT and conventional tandems at the center. We then incrementally shifted the positions laterally away from the center of up to 15 mm, at 1-mm steps. The in-house coded gradient projection-based inverse planning system was then used to generate inverse optimized plans ensuring identical V100=100% coverage. Conformity index (CI) was calculated for all plans. Results: Overall, the DMBT tandem generates more conformal dose distributions than conventional tandem for all lateral positional shifts of 0-15 mm (CI=0.91–0.52 and 0.99–0.34, respectively), with an exception at the central position due to the ideal circular dose distribution, generated by the {sup 192}Ir, fitting tightly around the circular target (CI = 0.91 and 0.99, respectively). The DMBT tandem is able to generate dose conformity of CI>0.8 at up to 6-mm positional shift while the conventional tandem violates this past 2-mm shift. Also, the CI ratio (=DMBT/conv.) increases rapidly until about 8 mm and then stabilizes beyond. Conclusion: A substantial enhancement in the dose sculpting capacity has been demonstrated for the novel DMBT tandem applicator. While further studies are warranted, the concept is promising for potential clinical translation.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

On the Solution of ℓ 0 -Constrained Sparse Inverse Covariance Estimation Problems

The sparse inverse covariance matrix is used to model conditional dependencies between variables in a graphical model to fit a multivariate Gaussian distribution. Estimating the matrix from data are well known to be computationally expensive for large-scale problems. Sparsity is employed to handle noise in the data and to promote interpretability of a learning model. Although the use of a convex ℓ 1 regularizer to encourage sparsity is common practice, the combinatorial ℓ 0 penalty often has more favorable statistical properties. In this paper, we directly constrain sparsity by specifying a maximally allowable number of nonzeros, in other words, by imposing an ℓ 0 constraint. Here, we introduce an efficient approximate Newton algorithm using warm starts for solving the nonconvex ℓ 0 -constrained inverse covariance learning problem. Numerical experiments on standard data sets show that the performance of the proposed algorithm is competitive with state-of-the-art methods.

$\ell_0$-Constrained↗
NASA NTRS1990

Euler and potential computational results for selected aerodynamic configurations

A selection of CFD successes and failures is evaluated, on the basis of experimental data/CFD result correlations involving full-potential and Euler computations of the aerodynamics of four commercial transport wings and two low aspect ratio delta wings. An effort is made to ascertain optimum values for grid density and distribution, artificial dissipation, Courant-Friedrichs-Lewy number, enthalphy damping, and a multigrid scheme for each flow condition and configuration analyzed. It is demonstrated that CFD solutions can assist the experimentalist prior to a test by indicating the locations of high pressure gradients and projecting test condition limitations due to balance design limits.

Hicks, R. M.↗
NASA NTRS1973

Aircraft symmetric flight optimization

Review of the development of gradient techniques and their application to aircraft optimal performance computations in the vertical plane of flight. Results obtained using the method of gradients are presented for attitude- and throttle-control programs which extremize the fuel, range, and time performance indices subject to various trajectory and control constraints, including boundedness of engine throttle control. A penalty function treatment of state inequality constraints which generally appear in aircraft performance problems is outlined. Numerical results for maximum-range, minimum-fuel, and minimum-time climb paths for a hypothetical supersonic turbojet interceptor are presented and discussed. In addition, minimum-fuel climb paths subject to various levels of ground overpressure intensity constraint are indicated for a representative supersonic transport. A variant of the Gel'fand-Tsetlin 'method of ravines' is reviewed, and two possibilities for further development of continuous gradient processes are cited - namely, a projection version of conjugate gradients and a curvilinear search.

Falco, M.↗
DOE OSTI2021

Seismic Spatial Gradients and Machine Learning-Based Classifiers for Explosion Monitoring (LDRD 218327)

This final report summarizes the work completed under the Laboratory Directed Research and Development (LDRD) project “Seismic Spatial Gradients as a Machine Learning-Based Classifier for Explosion Monitoring.” The overarching goal of the project was to explore the efficacy of using machine learning-based classification algorithms where the input data are the spatial gradient of the seismic wavefield collected at a single point on the Earth’s surface. The methods that I describe here are in direct contrast to conventional methods of seismic discrimination which typically rely on a spatially extended network of instruments and physics-based wavefield attributes such as, for example, the ratio between $\textit{P}$ and $\textit{S}$ waves. Rather, we use the spatial gradient of the seismic wavefield observed at a single point on the Earth’s surface and data processing approaches inspired by the machine learning community. We tested two algorithms, a neural network and a modified version of principal component analysis termed Spectrally Filtered Principal Component Analysis (SFPCA). To test these algorithms, we first conducted a series of numerical tests using synthetic data and then conducted a small-scale controlled field experiment. The tests using synthetic data showed that both algorithms had high success rates on gradiometric data, even when simulated noise was added to the signal. Furthermore, we found that using seismic spatial gradients increased the performance of our discrimination algorithms when compared to using just the traditional translational motion seismic data. The tests with field data also showed a high degree of discriminative success.

58 GEOSCIENCES↗
NASA NTRSDate not supplied

Development of a Self-Aligned Compact Focusing Schlieren System for NASA Test Facilities

A novel compact, inherently self-aligned focusing schlieren system is presented for use in experiments that require flow visualization. Unlike existing focusing schlieren systems that require a separate, and precisely aligned, source grid and cutoff grid to achieve sensitivity to density gradients, this system projects an image of a physical grid element onto a background and then reimages this projection back onto the physical grid element thereby achieving self-alignment. Further, unlike existing focusing schlieren systems where the image of the source grid is physically translated relative to the cutoff grid to achieve sensitivity to density gradients, a polarization prism in the system acts on the polarization state of the light of the projected grid element to provide sensitivity adjustment. By adjusting the position of the prism along the optical axis of the instrument, changes to the sensitivity can be realized. Since the outgoing and returning light used to project an image of the grid element are orthogonally polarized to one another, reflections from windows that may be present between the lens and background of the system can be eliminated. The system can also be implemented in test facilities where optical access is limited and/or exists on only one side. A discussion on different aspects of the development of this technology is provided, as are sample images from the system at the different stages of development. Images that demonstrate the ability to make on-the-fly adjustments to the instrument's field-of-view and working distance are also presented.

Brett F Bathel↗
NASA NTRS1986

Diagnosis of the role of vertical deformation in a two-dimensional primitive equation model of upper-level frontogenesis

In the present equations for the temporal rates of change of the magnitudes of potential temperature and absolute momentum vector gradients that are projected onto vertical planes transverse to straight frontal zones, the terms involving the transverse ageostrophic circulation are of the same mathematical form, and in a kinematic sense are also analogous to, the divergence and deformation terms that involve the horizontal wind field in Pettersen's (1936) classic equation for frontogenesis in the potential temperature field. The proposed frontogenesis equation form for the magnitude of the potential temperature gradient in the transverse plane is illustrated with the results of two simulations from an idealized, two-dimensional primitive equation model in which upper level frontogenesis proceeds by very different mechanisms.

Keyser, D.↗
DOE OSTI2024

What is the gradient of a scalar function defined on a subspace of square matrices?

We illustrate a technique to calculate the gradient of scalar functions that are defined on any arbitrary matrix subspace. It generalizes our earlier work titled “What is the gradient of a scalar function of a symmetric matrix ?”(Indian Journal of Pure and Applied Mathematics (2022), doi:10.1007/s13226-022-00313-x), in which we considered the special case of the subspace of symmetric matrices. Extant methods to calculate the gradient in such cases have an inherent flaw which leads to spurious results that populate several publications, as well as respected textbooks and handbooks on matrix calculus. We examine these sources and results in a rigorous and concrete mathematical setting of a finite-dimensional inner-product space and discover the inherent flaw and also a remedy. We demonstrate two ways to calculate the derivative/gradient and second derivative for scalar functions of matrices defined over an arbitrary matrix subspace; the first method is by considering any (differentiable) extension to the space of square matrices and projection of its gradient onto the given subspace. The second method utilizes an ordered basis and computes each component of the gradient through evaluation of the directional derivative. All the ideas presented are illustrated by non-trivial examples, namely, considering the subspace of 3 × 3 circulant and Toeplitz matrices and presenting the results of gradient-descent with both the spurious and correct gradients. Moreover, our bibliography makes it clear that a rigorous approach to matrix calculus is not common in practice, and our presentation of matrix calculus in the language of inner-product spaces will be significant and meaningful for applied mathematicians, engineers and researchers working in inter-disciplinary fields to avoid the conceptual pitfalls that exist.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

What is the gradient of a scalar function defined on a subspace of square matrices?

We illustrate a technique to calculate the gradient of scalar functions that are defined on any arbitrary matrix subspace. It generalizes our earlier work titled “What is the gradient of a scalar function of a symmetric matrix ?”(Indian Journal of Pure and Applied Mathematics (2022), https://doi.org/10.1007/s13226-022-00313-x), in which we considered the special case of the subspace of symmetric matrices. Extant methods to calculate the gradient in such cases have an inherent flaw which leads to spurious results that populate several publications, as well as respected textbooks and handbooks on matrix calculus. Here, we examine these sources and results in a rigorous and concrete mathematical setting of a finite-dimensional inner-product space and discover the inherent flaw and also a remedy. We demonstrate two ways to calculate the derivative/gradient and second derivative for scalar functions of matrices defined over an arbitrary matrix subspace; the first method is by considering any (differentiable) extension to the space of square matrices and projection of its gradient onto the given subspace. The second method utilizes an ordered basis and computes each component of the gradient through evaluation of the directional derivative. All the ideas presented are illustrated by non-trivial examples, namely, considering the subspace of 3 x 3 circulant and Toeplitz matrices and presenting the results of gradient-descent with both the spurious and correct gradients. Moreover, our bibliography makes it clear that a rigorous approach to matrix calculus is not common in practice, and our presentation of matrix calculus in the language of inner-product spaces will be significant and meaningful for applied mathematicians, engineers and researchers working in inter-disciplinary fields to avoid the conceptual pitfalls that exist.

97 MATHEMATICS AND COMPUTING↗