SEARCH · Engineering Papers

Results for “gradient methods”

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 433 records · Page 24

All sources NASA NTRS DOE OSTI

NASA NTRS1989

Robotic control of the seven-degree-of-freedom NASA laboratory telerobotic manipulator

A computationally efficient robotic control scheme for the NASA Laboratory Telerobotic Manipulator (LTM) is presented. This scheme utilizes the redundancy of the seven-degree-of-freedom LTM to avoid joint limits and singularities. An analysis to determine singular configurations is presented. Performance criteria are determined based on the joint limits and singularity analysis. The control scheme is developed in the framework of resolved rate control using the gradient projection method, and it does not require the generalized inverse of the Jacobian. An efficient formulation for determining the joint velocities of the LTM is obtained. This control scheme is well suited for real-time implementation, which is essential if the end-effector trajectory is continuously modified based on sensory feedback. Implementation of this scheme on a Motorola 68020 VME bus-based controller of the LTM is in progress. Simulation results demonstrating the redundancy utilization in the robotic mode are presented.

Dubey, R. V.↗

NASA NTRS1991

Internally mounted thin-liquid-film skin-friction meter - Comparison with floating element method with and without pressure gradient

A new, robust oil film skin friction meter was designed and constructed. This enables skin friction measurements remotely and from within the model, as well as avoiding the need to know the location of the leading edge of the film. The instrument was tested by comparing measurements with those given by a floating element gage in a zero pressure gradient flat plate turbulent boundary layer. Both instruments agreed satisfactorily with the well-known curve for this case. Significant discrepancies between the two instruments were observed in the case of adverse and favorable pressure gradients. The discrepancies were of opposite sign for opposite-sign pressure gradients as is consistent with the error expected from floating-element gages. Additional confidence in the oil film technique is supplied by the good agreement of the behavior of the film profile with predictions from lubrication theory.

Hornung, Hans↗

NASA NTRS1991

Inverse problems in diffraction

A two-dimensional problem of diffraction of a plane electromagnetic wave on a smooth 2 pi-periodic surface is considered. A numerical algorithm solving this problem is developed. An inverse problem of determination of the shape of 2 pi-periodic surface using the performance data of reverse scattering is considered. The inverse problem was solved by means of minimization of the residual functional with the help of the gradient descent method. The initial data were calculated with the help of the numerical method. On each step of the iterative method of minimization, the residual functional was calculated approximately with the help of the small slope method. The examples of the shape determination are considered.

Mikheev, Andrew G.↗

NASA NTRS1993

Grid-independent upwind scheme for multidimensional flow

Recent advances in the development of a grid-independent finite volume scheme for the Euler equations of gas dynamics are described. In the proposed method, flowfield gradient data are reconstructed locally (on a triangle) using five elementary planar waves, and an upwind numerical flux function for grid-oblique waves is developed to model the effect of the passage of these waves on the data in a cell. Numerical examples for several two-dimensional test problems are included. These results show high wave resolution and nearly monotone strong-wave transitions.

Parpia, Ijaz H.↗

NASA NTRS1993

The role of simulation in the design of a neural network chip

An iterative, simulation-based design procedure for a neural network chip is introduced. For this design procedure, the goal is to produce a chip layout for a neural network in which the weights are determined by transistor gate width-to-length ratios. In a given iteration, the current layout is simulated using the circuit simulator SPICE, and layout adjustments are made based on conventional gradient-decent methods. After the iteration converges, the chip is fabricated. Monte Carlo analysis is used to predict the effect of statistical fabrication process variations on the overall performance of the neural network chip.

Desai, Utpal↗

NASA NTRS1990

Direct handling of equality constraints in multilevel optimization

In recent years there have been several hierarchic multilevel optimization algorithms proposed and implemented in design studies. Equality constraints are often imposed between levels in these multilevel optimizations to maintain system and subsystem variable continuity. Equality constraints of this nature will be referred to as coupling equality constraints. In many implementation studies these coupling equality constraints have been handled indirectly. This indirect handling has been accomplished using the coupling equality constraints' explicit functional relations to eliminate design variables (generally at the subsystem level), with the resulting optimization taking place in a reduced design space. In one multilevel optimization study where the coupling equality constraints were handled directly, the researchers encountered numerical difficulties which prevented their multilevel optimization from reaching the same minimum found in conventional single level solutions. The researchers did not explain the exact nature of the numerical difficulties other than to associate them with the direct handling of the coupling equality constraints. The coupling equality constraints are handled directly, by employing the Generalized Reduced Gradient (GRG) method as the optimizer within a multilevel linear decomposition scheme based on the Sobieski hierarchic algorithm. Two engineering design examples are solved using this approach. The results show that the direct handling of coupling equality constraints in a multilevel optimization does not introduce any problems when the GRG method is employed as the internal optimizer. The optimums achieved are comparable to those achieved in single level solutions and in multilevel studies where the equality constraints have been handled indirectly.

Renaud, John E.↗

NASA NTRS1994

Thermal analysis of a functionally graded material subject to a thermal gradient using the boundary element method

The boundary element method is utilized in this study to conduct thermal analysis of functionally graded composites, materials in which the internal microstructure or properties are explicitly tailored in order to obtain an optimal response, on the micromechanical (constituent) scale. A unique feature of the boundary element formulations used here is the use of circular shape functions to convert the two-dimensional integrations of the composite fibers to one dimensional integrations. Using the computer code BEST-CMS, the through the thickness temperature profiles are computed for a representative material with varying numbers of fibers and fiber spacing in the thickness direction. The computed temperature profiles are compared to those obtained using an alternate analytical theory which explicitly couples the heterogeneous microstructure to the global analysis. The boundary element results compared favorably to the analytical calculations, with discrepancies that are explainable based on the boundary element formulation. The results serve both to demonstrate the ability of the boundary element method to analyze these types of materials, and to verify the accuracy of the analytical theory.

Goldberg, Robert K.↗

NASA NTRS1995

Minimization of the Vibration Energy of Thin-Plate Structures and the Application to the Reduction of Gearbox Vibration

While the vibration analysis of gear systems has been developed, a systematic approach to the reduction of gearbox vibration has been lacking. The technique of reducing vibration by shifting natural frequencies is proposed here for gearboxes and other thin-plate structures using the theories of finite elements, modal analysis, and optimization. A triangular shell element with 18 degrees of freedom is developed for structural and dynamic analysis. To optimize, the overall vibration energy is adopted as the objective function to be minimized at the excitation frequency by varying the design variable (element thickness) under the constraint of overall constant weight. Modal analysis is used to determine the sensitivity of the vibration energy as a function of the eigenvalues and eigenvectors. The optimum design is found by the gradient projection method and a unidimensional search procedure. By applying the computer code to design problems for beams and plates, it was verified that the proposed method is effective in reducing vibration energy. The computer code is also applied to redesign the NASA Lewis gear noise rig test gearbox housing. As one example, only the shape of the top plate is varied, and the vibration energy levels of all the surfaces are reduced, yielding an overall reduction of 1/5 compared to the initial design. As a second example, the shapes of the top and two side plates are varied to yield an overall reduction in vibration energy of 1/30.

Inoue, Katsumi↗

NASA NTRS1996

Analysis of the Hessian for Aerodynamic Optimization: Inviscid Flow

In this paper we analyze inviscid aerodynamic shape optimization problems governed by the full potential and the Euler equations in two and three dimensions. The analysis indicates that minimization of pressure dependent cost functions results in Hessians whose eigenvalue distributions are identical for the full potential and the Euler equations. However the optimization problems in two and three dimensions are inherently different. While the two dimensional optimization problems are well-posed the three dimensional ones are ill-posed. Oscillations in the shape up to the smallest scale allowed by the design space can develop in the direction perpendicular to the flow, implying that a regularization is required. A natural choice of such a regularization is derived. The analysis also gives an estimate of the Hessian's condition number which implies that the problems at hand are ill-conditioned. Infinite dimensional approximations for the Hessians are constructed and preconditioners for gradient based methods are derived from these approximate Hessians.

Arian, Eyal↗

NASA NTRS2000

Experimental Verification of Electric Drive Technologies Based on Artificial Intelligence Tools

In this report, a fully integrated prototype of a flight servo control system is successfully developed and implemented using brushless dc motors. The control system is developed by the fuzzy logic theory, and implemented with a multilayer neural network. First, a neural network-based architecture is introduced for fuzzy logic control. The characteristic rules and their membership functions of fuzzy systems are represented as the processing nodes in the neural network structure. The network structure and the parameter learning are performed simultaneously and online in the fuzzy-neural network system. The structure learning is based on the partition of input space. The parameter learning is based on the supervised gradient decent method, using a delta adaptation law. Using experimental setup, the performance of the proposed control system is evaluated under various operating conditions. Test results are presented and discussed in the report. The proposed learning control system has several advantages, namely, simple structure and learning capability, robustness and high tracking performance and few nodes at hidden layers. In comparison with the PI controller, the proposed fuzzy-neural network system can yield a better dynamic performance with shorter settling time, and without overshoot. Experimental results have shown that the proposed control system is adaptive and robust in responding to a wide range of operating conditions. In summary, the goal of this study is to design and implement-advanced servosystems to actuate control surfaces for flight vehicles, namely, aircraft and helicopters, missiles and interceptors, and mini- and micro-air vehicles.

Rubaai, Ahmed↗

NASA NTRS2004

Crystal Growth of CdTe by Gradient Freeze in Universal Multizone Crystallizator (UMC)

In the case of unsealed melt growth of an array of II-VI compounds, namely, CdTe, CdZnTe and ZnSe, there is a tremendous amount of experimental data describing the correlations between melt conditions and crystal quality. The results imply that the crystallinity quality can be improved if the melt was markedly superheated or long-time held before growth. It is speculated that after high superheating the associated complex dissociate and the spontaneous nucleation is retarded. In this study, crystals of CdTe were grown from melts which have undergone different thermal history by the unseeded gradient freeze method using the Universal Multizone Crystallizator (UMC). The effects of melt conditions on the quality of grown crystal were studied by various characterization techniques, including Synchrotron White Beam X-ray Topography (SWSXT), infrared microscopy, chemical analysis by glow discharge mass spectroscopy (GDMS), electrical conductivity and Hall measurements.

Su, Ching-Hua↗

NASA NTRS1996

Magnetic resonance imaging of the calcaneus: preliminary assessment of trabecular bone-dependent regional variations in marrow relaxation time compared with dual X-ray absorptiometry

RATIONALE AND OBJECTIVES: Marrow transverse relaxation time (T2*) in magnetic resonance (MR) imaging may be related to the density and structure of the surrounding trabecular network. We investigated regional variations of T2* in the human calcaneus and compared the findings with bone mineral density (BMD), as measured by dual X-ray absorpiometry (DXA). Short- and long-term precisions were evaluated first to determine whether MR imaging would be useful for the clinical assessment of disease status and progression in osteoporosis. METHODS: Gradient-recalled echo MR images of the calcaneus were acquired at 1.5 T from six volunteers. Measurements of T2* were compared with BMD and (for one volunteer) conventional radiography. RESULTS: T2* values showed significant regional variation; they typically were shortest in the superior region of the calcaneus. There was a linear correlation between MR and DXA measurements (r = .66 for 1/T2* versus BMD). Differences in T2* attributable to variations in analysis region-of-interest placement were not significant for five of the six volunteers. Sagittal MR images had short- and long-term precision errors of 4.2% and 3.3%, respectively. For DXA, the precision was 1.3% (coefficient of variation). CONCLUSION: MR imaging may be useful for trabecular bone assessment in the calcaneus. However, given the large regional variations in bone density and structure, the choice of an ROI is likely to play a major role in the accuracy, precision, and overall clinical efficacy of T2* measurements.

Non-NASA Center↗

NASA NTRS2002

Compatibility of segmented thermoelectric generators

It is well known that power generation efficiency improves when materials with appropriate properties are combined either in a cascaded or segmented fashion across a temperature gradient. Past methods for determining materials used in segmentation weremainly concerned with materials that have the highest figure of merit in the temperature range. However, the example of SiGe segmented with Bi2Te3 and/or various skutterudites shows a marked decline in device efficiency even though SiGe has the highest figure of merit in the temperature range. The origin of the incompatibility of SiGe with other thermoelectric materials leads to a general definition of compatibility and intrinsic efficiency. The compatibility factor derived as = (Jl+zr - 1) a~ is a function of only intrinsic material properties and temperature, which is represented by a ratio of current to conduction heat. For maximum efficiency the compatibility factor should not change with temperature both within a single material, and in the segmented leg as a whole. This leads to a measure of compatibility not only between segments, but also within a segment. General temperature trends show that materials are more self compatible at higher temperatures, and segmentation is more difficult across a larger -T. The compatibility factor can be used as a quantitative guide for deciding whether a material is better suited for segmentation orcascading. Analysis of compatibility factors and intrinsic efficiency for optimal segmentation are discussed, with intent to predict optimal material properties, temperature interfaces, and/or currentheat ratios.

Thermoelectric power generation segmented↗

NASA NTRS1993

Trajectory Learning Using Hierarchy of Oscillatory Modules

To this date, the most successful approaches to learning have been either the back-propagation or gradient descent method.

hierarchical and modular approach trajectory learn↗

NASA NTRS1993

Ru(sub 2)Ge(sub 3): Crystal Growth and Come Properties

Large samples of Ru(sub 2)Ge(sub 3) were grown from Ge-rich off-stoichiometric melts at a temperature close to 1460 C by a vertical gradient freeze method in graphite and glassy carbon crucibles.

off-stoichiometric melt twinning crack formation h↗

NASA NTRS2007

Crystal Growth and Characterization of CdTe Grown by Vertical Gradient Freeze

In this study, crystals of CdTe were grown from melts by the unseeded vertical gradient freeze method. The quality of grown crystal were studied by various characterization techniques including Synchrotron White Beam X-ray Topography (SWBXT), chemical analysis by glow discharge mass spectroscopy (GDMS), low temperature photoluminescence (PL), and Hall measurements. The SWBXT images from various angles show nearly strain-free grains, grains with inhomogeneous strains, as well as twinning nucleated in the shoulder region of the boule. The GDMS chemical analysis shows the contamination of Ga at a level of 3900 ppb, atomic. The low temperature PL measurement exhibits the characteristic emissions of a Ga-doped sample. The Hall measurements show a resistivity of 1 x l0(exp 7) ohm-cm at room temperature to 3 x 10(exp 9) ohm-cm at 78K with the respective hole and electron concentration of 1.7 x 10(exp 9) cm(exp -3) and 3.9 x 10(exp 7) cm(exp -3) at room temperature.

Su, Ching-Hua↗

NASA NTRS2005

Kurtosis Approach Nonlinear Blind Source Separation

In this paper, we introduce a new algorithm for blind source signal separation for post-nonlinear mixtures. The mixtures are assumed to be linearly mixed from unknown sources first and then distorted by memoryless nonlinear functions. The nonlinear functions are assumed to be smooth and can be approximated by polynomials. Both the coefficients of the unknown mixing matrix and the coefficients of the approximated polynomials are estimated by the gradient descent method conditional on the higher order statistical requirements. The results of simulation experiments presented in this paper demonstrate the validity and usefulness of our approach for nonlinear blind source signal separation Keywords: Independent Component Analysis, Kurtosis, Higher order statistics.

Duong, Vu A.↗

NASA NTRS2005

Kurtosis Approach for Nonlinear Blind Source Separation

kurtosis↗