Search NASASearch

SEARCH · Search NASA

Results for “Leaky”

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 37 records · Page 2

Characterization of metal-polymer boundaries using Stoneley and Leaky waves

Optical and acoustooptical techniques for bonded surface inspection and interface wave detection are discussed. Interface wave propagation models based on viscous, elastic, and viscoelastic mechanical properties are reviewed. Gross boundary anomalies such as large defects and localized disbond regions were measured using conventional high frequency pulse echo and pitch catch ultratechniques. Bond durabilty is estimated by comparing detected wave attenuation versus frequency as a function of prebond surface preparation with the experimental limit obtained for perfect welded contact. Sensitivity of interface acoustic field periodicity to both residual and dc applied stress at the boundary surface is observed as well as hysteresis due to viscoelastic polymeric behavior as the applied stress is cycled.

Claus, R. O.

A Leaky Waveguide Model for MHD Wave Driven Winds from Coronal Holes

Magnetohydrodynamic (MHD) waves, driven by the large scale convective motions of the photosphere are suggested as a possible source of additional acceleration for the stellar wind. Most of the turbulent power in a coronal hole is carried by MHD waves with periods of a few hundred seconds or longer. This is evident from direct observations of turbulence in the solar photosphere, as well as in situ observations of turbulence in the solar wind. But waves with periods this long have wavelengths which are typically as large as the transverse scale of the coronal hole flux tube itself. For these waves boundary effects are important and the coronal hole must be treated as a waveguide. The propagation of MHD waves using this waveguide approach is discussed. The simple model presented demonstrates that coronal holes can act as waveguides for MHD waves. For typical solar parameters the waves are compressible and can generate a wave tensile force which tends to cancel at least part of the wave pressure force. This effect tends to decrease the efficiency of MHD wave acceleration.

Davila, J. M.

A leaky magnetohydrodynamic waveguide model for the acceleration of high-speed solar wind streams in coronal holes

It is now reasonably well established that there is a correlation between high-speed solar wind streams and coronal holes. It has been concluded that a significant addition of momentum and/or energy in the region of supersonic flow is needed to explain the observed particle flux and flow speed observed in the high-speed streams. The most likely source of this additional momentum appears to be magnetohydrodynamic (MHD) waves propagating up from the solar surface. The present investigation is concerned with the propagation of MHD waves in a structure of finite transverse size, taking into account the consequences for the acceleration of high-speed solar wind streams. A waveguide solution for a model coronal hole is described, giving attention to a geometric or ray analysis of the slab waveguide, a wave mode analysis, an analytic solution of the dispersion relation for high-frequency waves, and the calculation of the time-averaged wave force.

Davila, J. M.

Stochastic acceleration of cometary pickup ions - The classic leaky box model

The acceleration of cometary pickup ions by magnetohydrodynamic waves at P/Giacobini-Zinner is examined in a model where acceleration predominantly occurs downstream of the bow shock throughout the cometosheath where intense magnetic turbulence exists. The mean free path for scattering by the magnetic fluctuations in this region is less than the characteristic dimension of the cometosheath so that pickup ions are rendered isotropic and energized by a modest amount in the process. This principal loss mechanism for the ions is spatial diffusion out of the acceleration region moderated by the self-same accelerating waves. This particular feature constrains the model in a way that the predicted ion spectrum is uniquely determined by the power spectrum of the magnetic turbulence. At both P/Giacobini-Zinner and P/Halley, the turbulence is non-Kolmogoroff with a spectral index of 2 resulting in an ion spectral behavior that is approximately an exponential in ion speed, consistent with a recent analysis of the Giacobini-Zinner data (see Richardson et al.).

Barbosa, D. D.

One GHz leaky SAW velocity of metal layers and bilayers evaporated onto fused quartz

An acoustic microscope operating at 1 GHz was used to determine the surface acoustic wave velocities v of thin film metal layers and metal bilayers deposited onto fused quartz. V(0)'s influence was reduced by gating. This produced a calibrated accuracy of 3 percent. A program was constructed from the explicitly solved 6 x 6 theoretical determinant. Single film thicknesses were decided upon by using this theory to produce a v equal to a standard. Since the single film v's were linear with respect to their thickness, half the single film thicknesses were used for the bilayered films. The velocities for these bilayered films agreed with theory. This experimentally confirms the theoretical technique used here to examine bilayered systems, or a prototype composite interphase. No discrepancy was seen for gold films as others have reported. V(z) seemed insensitive to the formation of intermetallics or CuO. Some annealed and unannealed platinum films did not change the v from that of fused quartz. Two platinum films whose v's were in agreement with theory peeled off with the tape test.

Walikainen, Dale

Development of Leaky Wave Antennas for Layered Ridge Dielectric Waveguide

The millimeter wave, especially above 100 GHz, and the submillimeter wave frequency spectrum offers the possibility for narrow-beam, high-resolution antennas which are critical for high definition radars required for space debris tracking, airport ground avoidance radars, and missile tracking. In addition, the frequency which most atmospheric constituents may be detected lie in this part of the frequency spectrum. Therefore, the development of electronic components for millimeter/submillimeter wave passive sensors is required for environmental monitoring of the Earth's atmosphere. Typical microwave transmission lines such as microstrip and coplanar waveguide rely on two or more electrical conductors to concentrate and guide the electromagnetic energy. Unfortunately, the surface resistance of the conductors increases as the square root of frequency. In addition, the circuit dimensions must be decreased with increasing frequency to maintain a single mode transmission line which further increases the conductor loss. An alternative family of transmission lines are formed from two or more insulating materials and rely on the differences in the permittivities between the two materials to guide the wave. No metal conductors are required although some dielectric waveguides do utilize a metallic ground plane to facilitate the interconnections of active electrical elements or to reduce the transmission line size. Examples of such transmission lines are image guides, insulated image guides, trapped image guides, ridge guide, and layered ridge dielectric waveguide (LRDW). Although most dielectric waveguides have dimensions on the order of lambda to provide sufficient field confinement, the LRDW has been shown to provide good field confinement for electrically small lines. This offers an advantage in circuit integration. It has been shown that a periodic array of metallic strips placed either along or on top of a dielectric waveguide forms an effective radiator. This antenna is easy to fabricate and there is good background of microstrip type antenna design information in the literature. This paper reports the development of the first frequency scanning antenna fed by a LRDW.

Ponchak, George E.

ReLU, Sparseness, and the Encoding of Optic Flow in Neural Networks

Accurate self-motion estimation is critical for various navigational tasks in mobile robotics. Optic flow provides a means to estimate self-motion using a camera sensor and is particularly valuable in GPS- and radio-denied environments. The present study investigates the influence of different activation functions—ReLU, leaky ReLU, GELU, and Mish—on the accuracy, robustness, and encoding properties of convolutional neural networks (CNNs) and multi-layer perceptrons (MLPs) trained to estimate self-motion from optic flow. Our results demonstrate that networks with ReLU and leaky ReLU activation functions not only achieved superior accuracy in self-motion estimation from novel optic flow patterns but also exhibited greater robustness under challenging conditions. The advantages offered by ReLU and leaky ReLU may stem from their ability to induce sparser representations than GELU and Mish do. Our work characterizes the encoding of optic flow in neural networks and highlights how the sparseness induced by ReLU may enhance robust and accurate self-motion estimation from optic flow.

97 MATHEMATICS AND COMPUTING

Energy dependence of the P(bar)/P ratio in cosmic rays

The antiproton/proton /P(bar)/P/ratio is presented for seven kinetic energy intervals from 4.4 - 13.4 GeV. The P(bar) were observed in 1979 by the New Mexico State University balloon-borne magnet spectrometer. The observations are compared to various models including secondary production in a leaky box. A decline in the P(bar)/P ratio at low energies is expected if the P(bar) are produced as secondaries. The data are consistent with such a decline but the P(bar)/P ratio is much higher than expected. The confidence level is less than 0.0001 for a model based on a leaky box with a mean-matter-traversed lambda of 7 gm/sq cm. The best leaky-box-model fit obtained by varying lambda gives a confidence level of 0.12 for lambda l= 21 + or - 3 gm/sq c. Results are also presented for fits based on the closed-galaxy and other models.

Golden, R. L.

Dielectric Covered Planar Antennas at Submillimeter Wavelengths for Terahertz Imaging

Most optical systems require antennas with directive patterns. This means that the physical area of the antenna will be large in terms of the wavelength. When non-cooled systems are used, the losses of microstrip or coplanar waveguide lines impede the use of standard patch or slot antennas for a large number of elements in a phased array format. Traditionally, this problem has been solved by using silicon lenses. However, if an array of such highly directive antennas is to be used for imaging applications, the fabrication of many closely spaced lenses becomes a problem. Moreover, planar antennas are usually fed by microstrip or coplanar waveguides while the mixer or the detector elements (usually Schottky diodes) are coupled in a waveguide environment. The coupling between the antenna and the detector/ mixer can be a fabrication challenge in an imaging array at submillimeter wavelengths. Antennas excited by a waveguide (TE10) mode makes use of dielectric superlayers to increase the directivity. These antennas create a kind of Fabry- Perot cavity between the ground plane and the first layer of dielectric. In reality, the antenna operates as a leaky wave mode where a leaky wave pole propagates along the cavity while it radiates. Thanks to this pole, the directivity of a small antenna is considerably enhanced. The antenna consists of a waveguide feed, which can be coupled to a mixer or detector such as a Schottky diode via a standard probe design. The waveguide is loaded with a double-slot iris to perform an impedance match and to suppress undesired modes that can propagate on the cavity. On top of the slot there is an air cavity and on top, a small portion of a hemispherical lens. The fractional bandwidth of such antennas is around 10 percent, which is good enough for heterodyne imaging applications.The new geometry makes use of a silicon lens instead of dielectric quarter wavelength substrates. This design presents several advantages when used in the submillimeter-wave and terahertz bands: a) Antenna fabrication compatible with lithographic techniques. b) Much simpler fabrication of the lens. c) A simple quarter-wavelength matching layer of the lens will be more efficient if a smaller portion of the lens is used. d) The directivity is given by the lens diameter instead of the leaky pole (the bandwidth will not depend anymore on the directivity but just on the initial cavity). The feed is a standard waveguide, which is compatible with proven Schottky diode mixer/detector technologies. The development of such technology will benefit applications where submillimeter- wave heterodyne array designs are required. The main fields are national security, planetary exploration, and biomedicine. For national security, wideband submillimeter radars could be an effective tool for the standoff detection of hidden weapons or bombs concealed by clothing or packaging. In the field of planetary exploration, wideband submillimeter radars can be used as a spectrometer to detect trace concentrations of chemicals in atmospheres that are too cold to rely on thermal imaging techniques. In biomedicine, an imaging heterodyne system could be helpful in detecting skin diseases.

Chattopadhyay, Goutam

Uncertainty propagation in feed-forward neural network models

We develop new uncertainty propagation methods for feed-forward neural network architectures with leaky ReLU activation functions subject to random perturbations in the input vectors. In particular, we derive analytical expressions for the probability density function (PDF) of the neural network output and its statistical moments as a function of the input uncertainty and the parameters of the network, i.e., weights and biases. A key finding is that an appropriate linearization of the leaky ReLU activation function yields accurate statistical results even for large perturbations in the input vectors. This can be attributed to the way information propagates through the network. We also propose new analytically tractable Gaussian copula surrogate models to approximate the full joint PDF of the neural network output. To validate our theoretical results, we conduct Monte Carlo simulations and a thorough error analysis on a multi-layer neural network representing a nonlinear integro-differential operator between two polynomial function spaces. Our findings demonstrate excellent agreement between the theoretical predictions and Monte Carlo simulations.

MLP networks

Cosmic ray antiprotons in closed galaxy model

The flux of secondary antiprotons expected for the leaky-box model was calculated as well as that for the closed galaxy model of Peters and Westergard (1977). The antiproton/proton ratio observed at several GeV is a factor of 4 higher than the prediction for the leaky-box model but is consistent with that predicted for the closed galaxy model. New low energy data is not consistent with either model. The possibility of a primary antiproton component is discussed.

Protheroe, R.

Cosmic ray antiprotons in the closed galaxy model

A calculation is made of the flux of secondary antiprotons expected for the leaky box model and for the closed galaxy model of Peters and Westergaard (1977). The antiproton/proton ratio observed at several GeV is a factor of 4 higher than the prediction for the leaky box model but is consistent with that predicted for the closed galaxy model. It is found that new low-energy data are not consistent with either model. Attention is given to the possibility of a primary antiproton component.

Protheroe, R. J.