Search NASA⌕ Search

SEARCH · Search NASA

Results for “Convolution”

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 1,441 records · Page 80

Exploring End-to-end Deep Learning Applications for Event Classification at CMS

An essential part of new physics searches at the Large Hadron Collider (LHC) at CERN involves event classification, or distinguishing potential signal events from those coming from background processes. Current machine learning techniques accomplish this using traditional hand-engineered features like particle 4-momenta, motivated by our understanding of particle decay phenomenology. While such techniques have proven useful for simple decays, they are highly dependent on our ability to model all aspects of the phenomenology and detector response. Meanwhile, powerful deep learning algorithms are capable of not only training on high-level features, but of performing feature extraction. In computer vision, convolutional neural networks have become the state-of-the-art for many applications. Motivated by their success, we apply deep learning algorithms to low-level detector data from the 2012 CMS Simulated Open Data to directly learn useful features, in what we call, end-to-end event classification. We demonstrate the power of this approach in the context of a physics search and offer solutions to some of the inherent challenges, such as image construction, image sparsity, combining multiple sub-detectors, and de-correlating the classifier from the search observable, among others.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Real-Time Movement-Based Traffic Volume Prediction at Signalized Intersections

The traffic volume of each movement at signalized intersections can provide valuable information on real-time traffic conditions that enable traffic control systems to dynamically respond to the fluctuated traffic demands. Real-time movement-based traffic volume prediction is challenging due to various nonlinear spatial relationships at different locations/approaches and the complicated underlying temporal dependencies. In this study, a novel deep intersection spatial-temporal network (DISTN) is developed for real-time movement-based traffic volume prediction at signalized intersections, which considers both spatial and temporal features by the convolutional neural network (CNN) and long short-term memory (LSTM), respectively. In addition, the within-day, daily, and weekly periodic trends of traffic volume are also considered in the proposed model. This is the first time that a deep-learning method has been applied for movement-based traffic volume prediction at signalized intersections. In the numerical experiment, the proposed model is evaluated using real-world data and simulation data to demonstrate its effectiveness. The impacts of various structures of traffic networks on the proposed model are also discussed. Results show that the proposed model outperforms some of the state-of-the-art volume prediction methods currently in the literature.

99 GENERAL AND MISCELLANEOUS↗

Formation of the diphenyl molecule in the crossed beam reaction of phenyl radicals with benzene

The chemical dynamics to form the D5-diphenyl molecule, C{sub 6}H{sub 5}C{sub 6}D{sub 5}, via the neutral-neutral reaction of phenyl radicals (C{sub 6}H{sub 5}) with D6-benzene (C{sub 6}D{sub 6}), was investigated in a crossed molecular beams experiment at a collision energy of 185 kJ mol{sup -1}. The laboratory angular distribution and time-of-flight spectra of the C{sub 6}H{sub 5}C{sub 6}D{sub 5} product were recorded at mass to charge m/z of 159. Forward-convolution fitting of our data reveals that the reaction dynamics are governed by an initial addition of the phenyl radical to the {pi} electron density of the D6-benzene molecule yielding a short-lived C{sub 6}H{sub 5}C{sub 6}D{sub 6} collision complex. The latter undergoes atomic deuterium elimination via a tight exit transition state located about 30 kJ mol{sup -1} above the separated reactants; the overall reaction to form D5-diphenyl from phenyl and D6-benzene was found to be weakly exoergic. The explicit identification of the D5-biphenyl molecules suggests that in high temperature combustion flames, a diphenyl molecule can be formed via a single collision event between a phenyl radical and a benzene molecule.

74 ATOMIC AND MOLECULAR PHYSICS↗

Characterization of majority and minority carrier deep levels in p-type GaN:Mg grown by molecular beam epitaxy using deep level optical spectroscopy

Deep level defects in p-type GaN:Mg grown by molecular beam epitaxy were characterized using steady-state photocapacitance and deep level optical spectroscopy (DLOS). Low frequency capacitance measurements were used to alleviate dispersion effects stemming from the deep Mg acceptor. Use of DLOS enabled a quantitative survey of both deep acceptor and deep donor levels, the latter being particularly important due to the limited understanding of minority carrier states for p-type GaN. Simultaneous electron and hole photoemissions resulted in a convoluted deep level spectrum that was decoupled by emphasizing either majority or minority carrier optical emission through control of the thermal filling time conditions. In this manner, DLOS was able to resolve and quantify the properties of deep levels residing near both the conduction and valence bandedges in the same sample. Bandgap states through hole photoemission were observed at E{sub v}+3.05 eV, E{sub v}+3.22 eV and E{sub v}+3.26 eV. Additionally, DLOS revealed levels at E{sub c}-3.24 eV and E{sub c}-2.97 eV through electron emission to the conduction band with the former attributed to the Mg acceptor itself. The detected deep donor concentration is less than 2% of activated [Mg] and demonstrates the excellent quality of the film.

36 MATERIALS SCIENCE↗

Effect of the intrinsic width on the piezoelectric force microscopy of a single ferroelectric domain wall

Intrinsic domain wall width is a fundamental parameter that reflects bulk ferroelectric properties and governs the performance of ferroelectric memory devices. We present closed-form analytical expressions for vertical and lateral piezoelectric force microscopy PFM profiles of a single ferroelectric domain wall for the conical and disk models of the tip, beyond point charge and sphere approximations. The analysis takes into account the finite intrinsic width of the domain wall and dielectric anisotropy of the material. These analytical expressions provide insight into the mechanisms of PFM image formation and can be used for a quantitative analysis of the PFM domain wall profiles. The PFM profile of a realistic domain wall is shown to be the convolution of its intrinsic profile and the resolution function of PFM.

Morozovska, Anna N↗

Schwinger-Dyson operators as invariant vector fields on a matrix model analog of the group of loops

For a class of large-N multimatrix models, we identify a group G that plays the same role as the group of loops on space-time does for Yang-Mills theory. G is the spectrum of a commutative shuffle-deconcatenation Hopf algebra that we associate with correlations. G is the exponential of the free Lie algebra. The generating series of correlations is a function on G and satisfies quadratic equations in convolution. These factorized Schwinger-Dyson or loop equations involve a collection of Schwinger-Dyson operators, which are shown to be right-invariant vector fields on G, one for each linearly independent primitive of the Hopf algebra. A large class of formal matrix models satisfying these properties are identified, including as special cases, the zero momentum limits of the Gaussian, Chern-Simons, and Yang-Mills field theories. Moreover, the Schwinger-Dyson operators of the continuum Yang-Mills action are shown to be right-invariant derivations of the shuffle-deconcatenation Hopf algebra generated by sources labeled by position and polarization.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Low-Frequency Seismic Ground Motion At The Pier Positions Of The Planned Messina Straits Bridge For A Realistic Earthquake Scenario

We estimated longer-period (period T>0.5 s) components of the ground motion at the piers of the planned Messina straits bridge. As the shortest fault-to-site distance is only 3-5 km, the kinematic earthquake rupture process has to be described in a realistic way and thus, the causative fault is represented by a dense grid of subfaults. To model the 1908 event, we assume a Mw = 7 earthquake, with a 40x20 km rectangular fault, and pure reverse dip-slip. The horizontal upper side of the rectangle is at 3-km depth, and the N corner of the rectangle is just between the piers. For the fault nucleation point, the least favorable place is assumed and a randomized rupture velocity is used in a particular run. In a typical simulation, the fault motion is initially represented by the time history of slip in each of the subfaults and by the distribution of the final seismic moment among the subsources (forming 'asperities'), both generated as lognormal random functions. The time histories are then filtered in order to fit a chosen source spectral model. The parameters that are conditioning the random functions can be based on the bulk of published fault inversions, or reproduced from an earlier successful attempt to simulate ground motions in the epicentral zone of the 1994, M = 6.7 Northridge, California, earthquake. In the second step of calculations, the Green functions (for each subfault and pier combination) are calculated for a layered halfspace model of the pier foundation stratigraphy, using an advanced Green function calculator, that allows an accurate calculation over the entire relevant frequency band including static terms. Finally, the 3-components of the strong ground motion are obtained at the two piers through convolution and summation over the different subsources. We compare a set of response horizontal velocity spectra (PRV) obtained from our calculations with a reference PRV that is considered as a reasonable upper bound for the possible ground motion near the piers. Our results suggest that the seismic ground motion under Torre Sicilia dominates over this under Torre Calabria and that the median (average log) PRV is generally above the reference one, about 1.1-1.3 times for T>4 s, and up to 2 times for 1<T{<=}4 s. The use of advanced fault and medium models, accounting also for the natural scatter of individual PRV due to events with the same gross source parameters, provides a sound basis for the deterministic engineering estimates of future seismic ground motion.

58 GEOSCIENCES↗

Evaluation of Biomaterials Using Micro-Computerized Tomography

Micro-computed tomography or Micro-CT is a high resolution, non-invasive, x-ray scanning technique that allows precise three-dimensional imaging and quantification of micro-architectural and structural parameters of objects. Tomographic reconstruction is based on a cone-beam convolution-back-projection algorithm. Micro-architectural and structural parameters such as porosity, surface area to volume ratio, interconnectivity, pore size, wall thickness, anisotropy and cross-section area of biomaterials and bio-specimens such as trabecular bone, polymer scaffold, bio-ceramics and dental restorative were evaluated through imaging and computer aided manipulation of the object scan data sets.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Anomalous Broadening of Balmer H{sub {alpha}} Line in Aluminum and Copper Hollow Cathode Glow Discharges

The presented results are concerned with the shape of Balmer alpha line emitted from a low pressure DC glow discharge with aluminum (Al) and copper (Cu) hollow cathode (HC) in pure H{sub 2} and Ar-H{sub 2} gas mixture. The analysis indicates that the line profile represents a convolution of Gaussian profiles resulting from different collision excitation processes.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Modelling and Evaluation of Spectra in Beam Aided Spectroscopy

The evaluation of active beam induced spectra requires advanced modelling of both active and passive features. Three types of line shapes are addressed in this paper: Thermal spectra representing Maxwellian distribution functions described by Gaussian-like line shapes, secondly broad-band fast ion spectra with energies well above local ion temperatures, and, finally, the narrow lines shapes of the equi-spaced Motion Stark multiplet (MSE) of excited neutral beam particles travelling through the magnetic field confining the plasma. In each case additional line shape broadening caused by Gaussian-like instrument functions is taken into account. Further broadening effects are induced by collision velocity dependent effective atomic rates where the observed spectral shape is the result of a convolution of emission rate function and velocity distribution function projected into the direction of observation. In the case of Beam Emission Spectroscopy which encompasses the Motional Stark features, line broadening is also caused by the finite angular spread of injected neutrals and secondly by a ripple in the acceleration voltage associated with high energy neutral beams.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

MAIZE: a 1 MA LTD-Driven Z-Pinch at The University of Michigan

Researchers at The University of Michigan have constructed and tested a 1-MA Linear Transformer Driver (LTD), the first of its type to reach the USA. The Michigan Accelerator for Inductive Z-pinch Experiments, (MAIZE), is based on the LTD developed at the Institute of High Current Electronics in collaboration with Sandia National Labs and UM. This LTD utilizes 80 capacitors and 40 spark gap switches, arranged in 40 'bricks,' to deliver a 1 MA, 100 kV pulse with 100 ns risetime into a matched resistive load. Preliminary resistive-load test results are presented for the LTD facility.Planned experimental research programs at UM include: a) Studies of Magneto-Raleigh-Taylor instability of planar foils, and b) Vacuum convolute studies including cathode and anode plasma.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Soldered Contact and Current Risetime Effects on Negative Polarity Wire Array Z-pinches

The Cornell University COBRA pulser is a nominal 1 MA machine, capable of driving up to 32 wire cylindrical Z-pinch arrays. COBRA can operate with variable current risetimes ranging from 100 ns to 200 ns (short and long pulse, respectively). Wires are typically strung with a 'press' contact to the electrode hardware, where the wire is loosely pulled against the hardware and held there to establish electrical contact. The machine is normally negative, but a bolt-on convolute can be used to modify the current path and effectively produce positive polarity operation at the load.Previous research with single wires on a 1-5 kA pulser has shown that soldering the wire, thereby improving the wire/electrode contact, and operating in positive polarity can improve the energy deposition into the wire and enhance wire core expansion. Negative polarity showed no difference. Previous experiments on the negative polarity, 20 MA, 100 ns Z accelerator have shown that improving the contact improved the x-ray yield.Cornell data were collected on 16-wire Aluminum Z-pinch arrays in negative polarity. Experiments were conducted with both short and long current pulses with soldered and no-soldered wire/electrode contacts. The initiation, ablation, implosion and stagnation phases were compared for these four conditions. Time dependent x-ray signals were measured using diodes and diamond detectors. An inductive voltage monitor was used to infer minimum current radius achieved, as defined by a uniform shell of current moving radially inward, producing a time dependent inductance. Total energy data were collected with a metal-strip bolometer. Self-emission data were collected by an XUV 4-frame camera and an optical streak camera.In negative polarity and with short pulses, soldering appeared to produce a smaller radius pinch and decrease variations in the x-ray pulse shape. The bolometer, laser backlighter, 4-frame and streak cameras showed negligible differences in the initiation ablation and implosion phases. X-ray yield was increased with soldered arrays in negative polarity with long pulses. The bolometer also showed a 50% increase in radial power emitted for soldered arrays. 4-frame images showed soldered arrays have a more pronounced 'Christmas Tree' effect originating from the cathode. The inductive voltage monitor showed, with both long and short pulses, that soldered and no-solder arrays reached the same minimum current radius at the same time. However, with long pulses soldered arrays radiate x-rays at the time of minimum current radius while no-solder array x-ray output is delayed by {approx}20 ns, resulting in decreased x-ray yield.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

UNIPIC code for simulations of high power microwave devices

In this paper, UNIPIC code, a new member in the family of fully electromagnetic particle-in-cell (PIC) codes for simulations of high power microwave (HPM) generation, is introduced. In the UNIPIC code, the electromagnetic fields are updated using the second-order, finite-difference time-domain (FDTD) method, and the particles are moved using the relativistic Newton-Lorentz force equation. The convolutional perfectly matched layer method is used to truncate the open boundaries of HPM devices. To model curved surfaces and avoid the time step reduction in the conformal-path FDTD method, CP weakly conditional-stable FDTD (WCS FDTD) method which combines the WCS FDTD and CP-FDTD methods, is implemented. UNIPIC is two-and-a-half dimensional, is written in the object-oriented C++ language, and can be run on a variety of platforms including WINDOWS, LINUX, and UNIX. Users can use the graphical user's interface to create the geometric structures of the simulated HPM devices, or input the old structures created before. Numerical experiments on some typical HPM devices by using the UNIPIC code are given. The results are compared to those obtained from some well-known PIC codes, which agree well with each other.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A dedicated superbend x-ray microdiffraction beamline for materials, geo-, and environmental sciences at the advanced light source

A new facility for microdiffraction strain measurements and microfluorescence mapping has been built on beamline 12.3.2 at the advanced light source of the Lawrence Berkeley National Laboratory. This beamline benefits from the hard x-radiation generated by a 6 T superconducting bending magnet (superbend). This provides a hard x-ray spectrum from 5 to 22 keV and a flux within a 1 {mu}m spot of {approx}5x10{sup 9} photons/s (0.1% bandwidth at 8 keV). The radiation is relayed from the superbend source to a focus in the experimental hutch by a toroidal mirror. The focus spot is tailored by two pairs of adjustable slits, which serve as secondary source point. Inside the lead hutch, a pair of Kirkpatrick-Baez (KB) mirrors placed in a vacuum tank refocuses the secondary slit source onto the sample position. A new KB-bending mechanism with active temperature stabilization allows for more reproducible and stable mirror bending and thus mirror focusing. Focus spots around 1 {mu}m are routinely achieved and allow a variety of experiments, which have in common the need of spatial resolution. The effective spatial resolution ({approx}0.2 {mu}m) is limited by a convolution of beam size, scan-stage resolution, and stage stability. A four-bounce monochromator consisting of two channel-cut Si(111) crystals placed between the secondary source and KB-mirrors allows for easy changes between white-beam and monochromatic experiments while maintaining a fixed beam position. High resolution stage scans are performed while recording a fluorescence emission signal or an x-ray diffraction signal coming from either a monochromatic or a white focused beam. The former allows for elemental mapping, whereas the latter is used to produce two-dimensional maps of crystal-phases, -orientation, -texture, and -strain/stress. Typically achieved strain resolution is in the order of 5x10{sup -5} strain units. Accurate sample positioning in the x-ray focus spot is achieved with a commercial laser-triangulation unit. A Si-drift detector serves as a high-energy-resolution ({approx}150 eV full width at half maximum) fluorescence detector. Fluorescence scans can be collected in continuous scan mode with up to 300 pixels/s scan speed. A charge coupled device area detector is utilized as diffraction detector. Diffraction can be performed in reflecting or transmitting geometry. Diffraction data are processed using XMAS, an in-house written software package for Laue and monochromatic microdiffraction analysis.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Assessment of a time-of-flight detection technique for measuring small velocities of cold atoms

A low noise time-of-flight detection system for laser cooled atoms has been constructed and incrementally optimized. Here, a thorough description of the construction is presented along with an analysis of the capabilities of the system. The quality of the detection (the resolution, the reproducibility, the sensitivity, etc.) is crucial for, e.g., the ability to see details in the velocity distribution profile, which is of interest for fundamental studies of statistical physics and of the laser cooling processes, and for detection of small initial velocities of an atomic cloud, important, e.g., when studying small drifts induced by Brownian motors and ratchets. We estimate the signal-to-noise ratio of our signal to be better than 1000:1 for a typical single shot, and we discuss the effect of the initial atomic cloud size, the probe size, and the effects of the wave packet spread during the fall time on the measured quantities. We show that the shape of the velocity distribution is well conserved during the mapping done in the detection, i.e., in the convolution with the probe beam, and that velocities as small as a few percent of the single photon recoil velocity can be resolved.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Finite orbit width effect in ion collisional transport in TJ-II

The validity of the traditional local diffusive approach and of the use of monoenergetic calculations has been studied for the stellarator TJ-II [Alejaldre et al., Fusion Technol. 17, 131 (1990)]: it is shown to be doubtful, under some circumstances, even in a purely collisional description of transport. The diffusion in physical space starting from Dirac-delta-like initial conditions has been studied using the code Integrator of Stochastic Differential Equations for Plasmas by Castejon et al. [Plasma Phys. Controlled Fusion 49, 753 (2007)]. Particles may experience large radial excursions from their original magnetic surfaces in a single collisional time. The contribution of these particles to the flux may make it nondiffusive; non-Gaussian density distributions, characterized by long tails, are observed. In the velocity space, there are important variations in the average particle kinetic energy after one collision time. We discuss the effect of this fact over the calculation of monoenergetic transport coefficients and their convolution. A simple analysis based on Hurst exponents has shown nevertheless that the description of transport by means of a pinch term and an effective transport coefficient is more correct than expected.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Three dimensional quantum geometry and deformed symmetry

We study a three dimensional noncommutative space emerging in the context of three dimensional Euclidean quantum gravity. Our starting point is the assumption that the isometry group is deformed to the Drinfeld double D(SU(2)). We generalize to the deformed case the construction of E{sup 3} as the quotient of its isometry group ISU(2) by SU(2). We show that the algebra of functions on E{sup 3} becomes the noncommutative algebra of SU(2) distributions, C(SU(2))*, endowed with the convolution product. This construction gives the action of ISU(2) on the algebra and allows the determination of plane waves and coordinate functions. In particular, we show the following: (i) plane waves have bounded momenta; (ii) to a given momentum are associated several SU(2) elements leading to an effective description of {phi} is an element of C(SU(2))* in terms of several physical scalar fields on E{sup 3}; (iii) their product leads to a deformed addition rule of momenta consistent with the bound on the spectrum. We generalize to the noncommutative setting the 'local' action for a scalar field. Finally, we obtain, using harmonic analysis, another useful description of the algebra as the direct sum of the algebra of matrices. The algebra of matrices inherits the action of ISU(2): rotations leave the order of the matrices invariant, whereas translations change the order in a way we explicitly determine.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Gyrokinetic simulation tests of quasilinear and tracer transport

A nonlinear gyrokinetic simulation code is used to test the quasilinear transport approximation (QLTA) with simulated nonlinear spectral (potential field) intensity. Two common forms of the QLTA are defined. The first uses the linear mode spectrum (mQLTA) and the second uses the complete frequency spectrum (fQLTA) for the nonlinear spectral intensity. The mQLTA is tested via two-step linear then nonlinear simulations convoluting a quasilinear weight with a nonlinear field intensity spectral weight to get the quasilinear transport in comparison with the actual nonlinear transport. The fQLTA is tested via one-step simulations that have ion and electron ''plasma species'' at full densities and ''tracer species'' at negligible densities (and making no contribution to the Poisson field solve equation). If the tracer and plasma gyrokinetic equations are identical, then so are their respective energy and particle diffusivities. Comparing tracer and plasma (actual) diffusivities, when the tracer equation nonlinearity is deleted, provides a quantifiable test of the fQLTA form. The mQLTA preserves ambipolarity but the two-step test includes only the leading linear normal modes at each wave number. The one-step test of the fQLTA subsumes all normal modes but precludes ambipolar particle flow. The mQLTA and fQLTA quasilinear weights (per normal mode) are shown to be identical for a commonly used (but unphysical) mode frequency line width model. In successful cases, quasilinear diffusivities are typically 1.4-1.8 (1.2-1.4) larger than actual diffusivities for mQLTA (fQLTA). The QLTA is expected to make best predictions in the ratios of energy and particle flows. Electron to ion energy flow ratios are well approximated but both forms of the QLTA appear to breakdown most evidently for ratios of particle to energy flows in cases with strongly pinched (and impractically large) particle flows. An example of the so-called passive tracer diffusivity, which includes only linear and nonlinear ExB motion, is given for comparison with actual diffusivities.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗