Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data smoothing”

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 1,207 records · Page 67

Finding MIDDLE Ground: Scalable and Secure Distributed Learning

Edge computing methods allow devices to efficiently train a high-performing, robust, and personalized model for predictive tasks. However, these methods succumb to privacy and scalability concerns such as adversarial data recovery and expensive model communication. Furthermore, edge computing methods unrealistically assume that all devices train an identical model. In practice, edge devices have varying computational and memory constraints which may not allow certain devices to have the space or speed to train a specific model. To overcome these issues, we propose MIDDLE: a model independent distributed learning algorithm which allows heterogeneous edge devices to assist each other’s training while communicating only non-sensitive information. MIDDLE unlocks the ability for edge devices, regardless of computational or memory constraints, to assist each other even with completely different model architectures. Furthermore, MIDDLE does not require model or gradient communication which greatly reduces communication size and time. We prove that MIDDLE attains the optimal convergence rate O(1/sqrt(TM)) of stochastic gradient descent for convex and non-convex smooth optimization (for total iterations T and batch size M). Finally, our experimental results demonstrate that MIDDLE (even in non-IID data settings) attains robust and high-performing models without model or gradient communication.

Bornstein, Marc I.↗

Potential Mars Surveyor 2001 Landing Sites: Low-Elevation Cratered "Highlands" in Central and Eastern Sinus Meridiani and Near Amenthes Fossae

The main scientific goal for the Mars Surveyor Program 2001 (MSP 01) landed mission is to collect and characterize 91 rock and 13 soil core samples using an integrated instrument suite onboard the Athena Rover. If possible, these samples will be retrieved and returned to Earth via a MSP 05 or MSP 07 mission. Preliminary engineering constraints for the MSP 01 landing site call for a location that lies between 15 S and 30 N, and below about 2 km elevation (based on Viking-era topography). Desirable landing sites for MSP 01 are to be located "in the ancient highlands where the environmental conditions may have been favorable to the preservation of evidence of possible prebiotic or biotic processes including the emergence (and, potentially, the persistence) of life". We interpret this to mean that the desirable sites include those that have evidence of aqueous sediments that might have been deposited during the Noachian and/or Hesperian Epochs of Mars' history. In addition to the search for subaqueous sedimentary deposits, we took into consideration the fact that the rover, Athena, will need to be able to access these materials. Thus, a site where aeolian deflation has occurred might be desirable because it might expose, in situ, layered sedimentary deposits. Deflated areas, of course, might include potential landing hazards in the form of meterscale buttes and mesas (e.g., Christmas Lake Valley, OR), thus careful study of such sites with high resolution images will be required before a decision is made to land. We have been examining three regions that have potential to be considered for MSP 01 landing sites. This work is based on Viking (VIS, IRTM) and Phobos 2 (Termoskan) observations and should be regarded as preliminary because we believe that the final site selection should also be based upon analysis of Mars Global Surveyor observations that help constrain mineralogy (TES) and local geomorphology (MOC, MOLA). (1) Eastern Sinus Meridiani Region (proposed by K. S. Edgett) Sinus Meridiani is a persistent low-albedo (< 0. 16) region on the martian equator that has been recognized for about 400 years. All of Sinus Meridiani is below the 2 km elevation constraint for MSP 01. The region includes cratered highlands, valley networks, aeolian dunes, and possible aqueous sedimentary deposits. The landing site study region is located in the eastern portion of Sinus Meridiani. It is bounded by latitudes 10 S to 2 N, longitudes 355 W to 345 W, and the elevations are mostly 1-2 km. The center of this area contains a medium-albedo (0.19-0.21), relatively smooth-surfaced deposit that was suggested by Rice to be a lacustrine deposit. Several potential landing areas can be suggested within this region. Based on Viking and Phobos 2 data, the favored sites so far are centered at 7.6 S, 346.9 W and 0.8 S, 349 W. Thermal inertias are 3.2-7.0 x 10(exp -3) cal /sq cm s(exp -0.5)/K; and rock abundances are around 2-6%. The site at 7.6 S, 346.9 W is at the southern end of the smooth, medium-albedo unit that might have a lacustrine origin. At this location, numerous channels appear to have drained toward the smooth unit. Viking images from orbit 747A show this area at about 15 m/pixel ground resolution. The images reveal that aeolian deflation has occurred along the deposit's margins. Bright (i.e., albedo >= 0.21) aeolian dunes are present on the channel floors and in some of the depressions on the smooth unit. The bright, apparently active dunes might consist of material (perhaps lakedeposited sands) that has been eroded from the smooth unit. The site at approximately 0.8 S, 349 W is selected because it offers an opportunity to solve a long-standing puzzle about Mars remote sensing. There are three main "color" units on Mars: "dark red, dark gray, and bright red". This landing site would allow the Athena rover an opportunity to investigate all three materials within close proximity (the best place on Mars to do so). There are no high resolution (better than 100 m/pixel) Viking or Mariner images of this site. (2) Central Sinus Meridiani Region (proposed by K. S. Edgett and T. J. Parker) Central Sinus Meridiani is characterized by two types of surfaces [4]. One is like typical martian cratered highlands elsewhere- there are old valley networks and old impact craters. The other is relatively smooth and flat. These two units are in contact around 3.1 S between 5 W and 4 E longitudes. Valley networks- including one at 6'S, 358 W that rivals the Grand Canyon of Arizona- once drained toward the smooth unit. Edgett and Parker [4] proposed that the smooth unit might consist of sediments laid down in a large Noachian-aged sea/ocean that would have covered much of the northern hemisphere. Schultz and Lutz [I I I suggested that it is a paleopolar layered deposit. Regardless, the smooth unit where it contacts the cratered terrain would make an excellent site for Athena rover to investigate. The site is best seen in Viking high resolution images from orbits 408B (about 30 m/pixel) and 746A (about 12 m/pixel). These images suggest that aeolian deflation has occurred along the margin of the smooth unit, and this deflation has exposed horizontal larrs of material. The elevation is about 0.5 km; thermal inertias are 6.5-8.0 x 10(exp -3) cal /sq cm s(exp -0.5) / K; rock abundances are 2-4%; and the surface is probably sandy with dark drifts and ripples but almost no actual dunes. We suggest a landing around 3.2 S, 3.0 W would test the aqueous sediment hypothesis and provide a potentially smooth surface on which to land. (3) Amenthes Fossae Region (proposed by S. N. Huntwork and K. S. Edgett) The Amenthes Fossae are a series of graben/fissures that are circumferential to the southeast side of Isidis Planitia. These fissures cross a variety of ancient, heavily cratered Noachian terrain and younger, Hesperian and Amazonian terrain. We focused our search on a region 0-15 N, 250 - 270 W. Elevations are -0.5 to 2 km. Depending upon whether Isidis Planitia was ever a water-rich environment, this region might have been influenced by aqueous sedimentation. Valley networks are common, and they drained toward the north and northwest. We focused our work on a set of Viking orbiter high-resolution images, 719A 1-48. These have resolutions 16-24 rri/pixel. We examined images 20-23, centered at 2 N, 258 W. This site, on the plains just southwest of a 42 km-diameter crater, includes a valley network channel, a relatively young crater ejecta deposit, a few buttes composed of presumably ancient, Noachian bedrock, and a "plains" unit. The plains might make an ideal landing surface, except for the presence of some fine-scale ridges (oriented approximately N-S). The ridges are probably yardangs, thus this site offers a place where aeolian deflation has probably exposed some of the layered rock units that comprise the "plains". Thermal inertias are 7.9-8.3 x 10(exp -3) cal/sq cm s(exp -0.5) /K and rock abundances are 10-15%. A rover traverse might include the opportunity to go down to the floor (and sample along the walls) of the valley network channel at 2 N 258.1 W.

Edgett, K. S.↗

Water Impact Test and Simulation of a Composite Energy Absorbing Fuselage Section

In March 2002, a 25-ft/s vertical drop test of a composite fuselage section was conducted onto water. The purpose of the test was to obtain experimental data characterizing the structural response of the fuselage section during water impact for comparison with two previous drop tests that were performed onto a rigid surface and soft soil. For the drop test, the fuselage section was configured with ten 100-lb. lead masses, five per side, that were attached to seat rails mounted to the floor. The fuselage section was raised to a height of 10-ft. and dropped vertically into a 15-ft. diameter pool filled to a depth of 3.5-ft. with water. Approximately 70 channels of data were collected during the drop test at a 10-kHz sampling rate. The test data were used to validate crash simulations of the water impact that were developed using the nonlinear, explicit transient dynamic codes, MSC.Dytran and LS-DYNA. The fuselage structure was modeled using shell and solid elements with a Lagrangian mesh, and the water was modeled with both Eulerian and Lagrangian techniques. The fluid-structure interactions were executed using the fast general coupling in MSC.Dytran and the Arbitrary Lagrange-Euler (ALE) coupling in LS-DYNA. Additionally, the smooth particle hydrodynamics (SPH) meshless Lagrangian technique was used in LS-DYNA to represent the fluid. The simulation results were correlated with the test data to validate the modeling approach. Additional simulation studies were performed to determine how changes in mesh density, mesh uniformity, fluid viscosity, and failure strain influence the test-analysis correlation.

Fasanella, Edwin L.↗

A novel approach to characterizing the surface topography of niobium superconducting radio frequency (SRF) accelerator cavities

As superconducting niobium radio-frequency (SRF) cavities approach fundamental material limits, there is increased interest in understanding the details of topographical influences on realized performance limitations. Micro- and nano-roughness are implicated in both direct geometrical field enhancements as well as complications of the composition of the 50 nm surface layer in which the super-currents typically flow. Interior surface chemical treatments such as buffered chemical polishing (BCP) and electropolishing (EP) used to remove mechanical damage leave surface topography, including pits and protrusions of varying sharpness. These may promote RF magnetic field entry, locally quenching superconductivity, so as to degrade cavity performance. A more incisive analysis of surface topography than the widely used average roughness is needed. In this study, a power spectral density (PSD) approach based on Fourier analysis of surface topography data acquired by both stylus profilometry and atomic force microscopy (AFM) is introduced to distinguish the scale-dependent smoothing effects, resulting in a novel qualitative and quantitative description of Nb surface topography. The topographical evolution of the Nb surface as a function of different steps of well-controlled EP is discussed. This study will greatly help to identify optimum EP parameter sets for controlled and reproducible surface levelling of Nb for cavity production.

43 PARTICLE ACCELERATORS↗

A fast, high-order solver for the Grad–Shafranov equation

We present a new fast solver to calculate fixed-boundary plasma equilibria in toroidally axisymmetric geometries. By combining conformal mapping with Fourier and integral equation methods on the unit disk, we show that high-order accuracy can be achieved for the solution of the equilibrium equation and its first and second derivatives. Smooth arbitrary plasma cross-sections as well as arbitrary pressure and poloidal current profiles are used as initial data for the solver. Equilibria with large Shafranov shifts can be computed without difficulty. Spectral convergence is demonstrated by comparing the numerical solution with a known exact analytic solution. A fusion-relevant example of an equilibrium with a pressure pedestal is also presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Synthesis and characteristics of spray deposited CuInS{sub 2} nanocrystals thin films for photovoltaic applications

Graphical abstract: - Highlights: • CuInS{sub 2} nanocrystals thin films were synthesized by spray pyrolysis technique. • They are polycrystalline and have chalcopyrite (tetragonal) structure. • They have high absorption coefficient ∼10{sup 4} cm{sup −1} and optimum band gap of 1.55 eV. • They showed excellent opto-electronic properties employable in photovoltaics. - Abstract: Nanocrystalline thin films of CuInS{sub 2}, an attractive absorber material for highly efficient and terrestrial photovoltaic devices, were deposited on ultraclean glass substrates using spray pyrolysis technique. The prepared films were characterized by FESEM, FETEM, HRTEM, AFM, XRD, optical absorption spectroscopy, photoluminescence and current–voltage characteristics. The films exhibit almost smooth, dense and uniform topography; and have nano-sized particles 40–60 nm of CuInS{sub 2}. XRD data show that the films are polycrystalline and have chalcopyrite (tetragonal) structure with crystallite size 45–60 nm. Optical absorption studies show that the band gap of spray deposited CuInS{sub 2} films is 1.55 eV, while absorption coefficient is of ∼10{sup 5} cm{sup −1}. PL spectra have only one peak at 1.43 eV which may be attributed to the excitonic recombination through donor–acceptor impurity levels in the sample. Furthermore, CdS/CuInS{sub 2} heterojunction was also produced onto ITO coated glass substrate to evaluate solar cell parameters such as short circuit current density, open circuit voltage, fill factor and conversion efficiency.

36 MATERIALS SCIENCE↗

Search for dijet resonances in proton–proton collisions at $\sqrt{s}$ = 13 TeV and constraints on dark matter and other models

A search is presented for narrow resonances decaying to dijet final states in proton–proton collisions at s=13TeV using data corresponding to an integrated luminosity of 12.9 fb −1 . The dijet mass spectrum is well described by a smooth parameterization and no significant evidence for the production of new particles is observed. Upper limits at 95% confidence level are reported on the production cross section for narrow resonances with masses above 0.6 TeV. In the context of specific models, the limits exclude string resonances with masses below 7.4 TeV, scalar diquarks below 6.9 TeV, axigluons and colorons below 5.5 TeV, excited quarks below 5.4 TeV, color-octet scalars below 3.0 TeV, W′ bosons below 2.7 TeV, Z′ bosons below 2.1 TeV and between 2.3 and 2.6 TeV, and RS gravitons below 1.9 TeV. These extend previous limits in the dijet channel. Vector and axial-vector mediators in a simplified model of interactions between quarks and dark matter are excluded below 2.0 TeV. The first limits in the dijet channel on dark matter mediators are presented as functions of dark matter mass and are compared to the exclusions of dark matter in direct detection experiments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

HIERARCHICAL STRUCTURE OF MAGNETOHYDRODYNAMIC TURBULENCE IN POSITION-POSITION-VELOCITY SPACE

Magnetohydrodynamic turbulence is able to create hierarchical structures in the interstellar medium (ISM) that are correlated on a wide range of scales via the energy cascade. We use hierarchical tree diagrams known as dendrograms to characterize structures in synthetic position-position-velocity (PPV) emission cubes of isothermal magnetohydrodynamic turbulence. We show that the structures and degree of hierarchy observed in PPV space are related to the presence of self-gravity and the global sonic and Alfvenic Mach numbers. Simulations with higher Alfvenic Mach number, self-gravity and supersonic flows display enhanced hierarchical structure. We observe a strong dependency on the sonic and Alfvenic Mach numbers and self-gravity when we apply the statistical moments (i.e., mean, variance, skewness, kurtosis) to the leaf and node distribution of the dendrogram. Simulations with self-gravity, larger magnetic field and higher sonic Mach number have dendrogram distributions with higher statistical moments. Application of the dendrogram to three-dimensional density cubes, also known as position-position-position (PPP) cubes, reveals that the dominant emission contours in PPP and PPV are related for supersonic gas but not for subsonic. We also explore the effects of smoothing, thermal broadening, and velocity resolution on the dendrograms in order to make our study more applicable to observational data. These results all point to hierarchical tree diagrams as being a promising additional tool for studying ISM turbulence and star forming regions for obtaining information on the degree of self-gravity, the Mach numbers and the complicated relationship between PPV and PPP data.

79 ASTRONOMY AND ASTROPHYSICS↗

OBSERVATIONS OF THE M82 SN 2014J WITH THE KILODEGREE EXTREMELY LITTLE TELESCOPE

We report observations of the bright M82 supernova 2014J serendipitously obtained with the Kilodegree Extremely Little Telescope (KELT). The supernova (SN) was observed at high cadence for over 100 days, from pre-explosion, to early rise and peak times, through the secondary bump. The high cadence KELT data with high signal-to-noise ratio is completely unique for SN 2014J and for any other SNIa, with the exception of the (yet) unpublished Kepler data. Here, we report determinations of the SN explosion time and peak time. We also report measures of the ''smoothness'' of the light curve on timescales of minutes/hours never before probed, and we use this to place limits on energy produced from short-lived isotopes or inhomogeneities in the explosion or the circumstellar medium. From the non-observation of significant perturbations of the light curves, we derive a 3σ upper limit corresponding to 8.7 × 10{sup 36} erg s{sup –1} for any such extra sources of luminosity at optical wavelengths.

79 ASTRONOMY AND ASTROPHYSICS↗

MarmotViz

MarmotViz is an illustrative visualization plug-in for ParaView. It enables users to identify and track time-varying features in simulation datasets. It permits the application of illustrative visualization effects to these features including: selective visualization, feature coloring, boundary smoothing, haloing, silhouette outlining, speedlines, and strobe silhouettes. These techniques serve to assist in exploration and interpretation of simulation data and can be used to generate enhanced renderings for presentation. The MarmotViz plug-in is developed as a flexible framework, and can be easily extended to incorporate new region-of-interest (ROI) identification algorithms, feature matching and tracking algorithms, and illustrative visualization effects.

Rattner, Alexander↗

Automated RF Phase Adjustment for Beam Stabilization in the Fermilab Linac

The Fermilab Linac experiences longitudinal beam phase drift, leading to increased particle loss, conventionally corrected through labor-intensive manual RF adjustments. This project explores machine learning-based automation for drift correction, employing a prototype-based classification approach. Our model utilizes a 34-dimensional feature set (RF settings and BPM readings) and leverages a 7x27 response matrix for system modeling. To overcome limited real-world data, we generate synthetic data, enhancing model training and generalizability. Custom loss functions, including a surrogate energy-consistent loss and a temporal smoothness constraint, ensure physically plausible drift predictions. The goal is a robust system for autonomous phase adjustments, ensuring stable beam acceleration and reduced manual intervention.

Chichili, R. R. [Illinois U., Chicago]↗

Upwelling studies with satellites

Two different ways of obtaining surface temperature structures with data from an orbiting platform are: (1) Analysis of radiometric recordings during one overpass over apparently cloud-free regions has provided the surface structures of an area. This method can be used to analyze a synoptic recording and has the further advantage of noting rapid changes in the sea surface temperatures over typical scales of a few days. (2) Mapping the sea surface temperature through use of a multispectral method which detects cloud-free conditions and uses the radiation for the determination of the sea surface temperature. This method includes data recorded from orbiting platforms over a period of 2 to 4 weeks; therefore, short-time fluctuations are smoothed out. The method is applicable to map ocean surfaces on a global scale. Both methods were used to study areas in oceans where upwelling exists.

Szekielda, K.↗

Lunar topography from Apollo 15 and 16 laser altimetry

In the orbital plane of Apollo 15 the mean lunar radius is 1737.3 km, the mean altitude of terrae above maria is about 3 km, and the center-of-figure is displaced from the center-of-mass by about 2 km away from longitude 25 E. The Apollo 16 laser altimeter obtained a total of about 7.5 revolutions of partially overlapping data. The principal difference in results from Apollo 16 is the absence of any great far-side basin similar to the 1400-km wide feature found by Apollo 15, 1200 km to the south. This absence of a far-side depression in the Apollo 16 orbital plane largely accounts for a greater mean radius: 1738.1 km; a greater mean altitude of terrae above maria: about 4 km; and a greater offset of centers: about 3 km, also away from 25 E. In the Apollo 16, as well as Apollo 15, data the far-side terrae are much 'rougher' than the near-side terrae. Mare surfaces are generally smooth to within plus or minus 150 m, and have slopes of 1:500 to 1:2000 persisting over distances as great as 500 km.

Kaula, W. M.↗

Drag and heat transfer on surfaces with small longitudinal fins

An investigation is conducted regarding the drag characteristics of longitudinally ribbed surfaces using various configurations which attempt to confine the turbulent wall bursts to regions of small transverse extent in the initial birth region. The ratio of heat transfer increase to drag increase for the high area ratio rib configurations is determined and the heat transfer efficiency of the present surfaces is compared with data found in the literature. The drag data obtained indicate local 'wetted area skin friction' reductions but due to the increased surface area the planform drag is increased in most instances. There is no indication of drag reduction with the rectangular ribbed models that had sharp leading and fin tip edges. Heat transfer data indicate that several of the triangular grooved models have efficiencies approximately 10 percent higher than that of a smooth flat plate.

Walsh, M. J.↗

Viking bistatic radar experiment - Summary of first-order results emphasizing north polar data

Initial results of bistatic radar observations of Mars made by the Viking Orbiter spacecraft are presented with particular emphasis on the previously unstudied polar regions. Bistatic radar scattering experiments were performed in near-equatorial regions by the Viking Orbiter 1, while other regions of the planet were observed by the polar-orbiting Orbiter 2, with scattered signals received by stations of the NASA Deep Space Network. In the equatorial region, the bistatic radar estimates of rms surface slope are found to be in qualitative agreement with results obtained using earth-based transmitter-receivers, showing a nearly 2:1 decrease in rms surface roughness between 20 and 22 deg N with no appreciable change in dielectric constant. Data in the north polar region reveal variations in surface roughness from 1 to 6 deg, with areas of smooth material generally located within rougher environs, surface roughness decreasing along Vastitas Borealis in the direction of the north pole, and north polar cap rms roughnesses on the order of 2.5-3.0 deg. Trends in radar reflectivity suggest a decreasing surface density with increasing latitude, consistent with a layer of seasonal CO2 or H2O snow increasing in depth as the pole is approached.

Simpson, R. A.↗

Rotor-to-stator rub vibration in centrifugal compressor

One example of excessive vibration encountered during loading of a centrifugal compressor train (H type compressor with HP casing) is discussed. An investigation was made of the effects of the dynamic load on the bearing stiffness and the rotor-bearing system critical speed. The high vibration occurred at a "threshold load," but the machine didn't run smoothly due to rubs even when it had passed through the threshold load. The acquisition and discussion of the data taken in the field as well as a description of the case history which utilizes background information to identify the malfunction conditions is presented. The analysis shows that the failures, including full reverse precession rub and exact one half subharmonic vibration, were caused by the oversize bearings and displacement of the rotor center due to foundation deformation and misalignment between gear shafts, etc. The corrective actions taken to alleviate excessive vibration and the problems which remain to be solved are also presented.

Gao, J. J.↗

Neuromorphic learning of continuous-valued mappings from noise-corrupted data. Application to real-time adaptive control

The ability of feed-forward neural network architectures to learn continuous valued mappings in the presence of noise was demonstrated in relation to parameter identification and real-time adaptive control applications. An error function was introduced to help optimize parameter values such as number of training iterations, observation time, sampling rate, and scaling of the control signal. The learning performance depended essentially on the degree of embodiment of the control law in the training data set and on the degree of uniformity of the probability distribution function of the data that are presented to the net during sequence. When a control law was corrupted by noise, the fluctuations of the training data biased the probability distribution function of the training data sequence. Only if the noise contamination is minimized and the degree of embodiment of the control law is maximized, can a neural net develop a good representation of the mapping and be used as a neurocontroller. A multilayer net was trained with back-error-propagation to control a cart-pole system for linear and nonlinear control laws in the presence of data processing noise and measurement noise. The neurocontroller exhibited noise-filtering properties and was found to operate more smoothly than the teacher in the presence of measurement noise.

Troudet, Terry↗