Making risk minimization tolerant to label noise

In many applications, the training data, from which one needs to learn a classifier, is corrupted with label noise. Many standard algorithms such as SVM perform poorly in the presence of label noise. In this paper we investigate the robustness of risk minimization to label noise. We prove a sufficient condition on a loss function for the risk minimization under that loss to be tolerant to uniform label noise. We show that the 0-1 loss, sigmoid loss, ramp loss and probit loss satisfy this condition though none of the standard convex loss functions satisfy it. We also prove that, by choosing a sufficiently large value of a parameter in the loss function, the sigmoid loss, ramp loss and probit loss can be made tolerant to nonuniform label noise also if we can assume the classes to be separable under noise-free data distribution. Through extensive empirical studies, we show that risk minimization under the 0-1 loss, the sigmoid loss and the ramp loss has much better robustness to label noise when compared to the SVM algorithm. (C) 2015 Elsevier B.V. All rights reserved.

Analysis on the atomization characteristics of aviation biofuel discharging from simplex swirl atomizer

The atomization characteristics of aviation biofuel discharging from a simplex swirl atomizer into quiescent atmospheric air are studied. The aviation biofuel is a mixture of 90% commercially available camelina-derived biofuel and 10% VonSol-53 (aromatics). The experiments are conducted in a spray test facility at varying fuel flow rate conditions. The measured characteristics include atomizer flow number, spray cone angle, breakup length of liquid sheet, wavelength of undulations on liquid sheet, and spray droplet size. The characteristics of biofuel sheet breakup are deduced from the captured images of biofuel spray. The measurements of spray droplet size distribution are obtained using Spraytec. The experimentally measured characteristics of the biofuel sheet breakup are compared with the predictions obtained from the liquid film breakup model proposed by Senecal et al. (1999). The measurements of wavelength and breakup length of the biofuel sheet discharging from the simplex swirl atomizer agree well with the model predictions. The model-predicted droplet size for the biofuel spray is significantly higher than the experimentally measured Sauter mean diameter (SMD). The spray droplets formed from the liquid sheet breakup undergo secondary atomization until 35-45 mm from the atomizer exit and thereafter the SMD increases downstream due to the combined effect of fuel evaporation and droplet coalescence. A good comparison is observed between the experimentally measured SMD of the biofuel spray and the predictions obtained using the empirical correlation reported in literature for sprays discharging from simplex swirl atomizers. (C) 2015 Elsevier Ltd. All rights reserved.

Atomization Characteristics of Camelina-Based Alternative Aviation Fuels Discharging From Dual-Orifice Injector

The atomization characteristics of blends of bioderived camelina hydrogenated renewable jet (HRJ) alternative fuel with conventional aviation kerosene (Jet A-1) discharging into ambient atmospheric air from a dual-orifice atomizer used in aircraft engines are described. The spray tests are conducted in a spray test facility at six different test flow conditions to compare the atomization of alternative fuels with that of Jet A-1. The fuel sprays are characterized in terms of fuel discharge, spray cone angle, drop size distribution, and spray patternation. The measurements of spray drop size distribution are obtained using laser diffraction based Spraytec equipment. The characteristics of fuel discharge and cone angle of alternative fuel sprays do not show any changes from that of Jet A-1 sprays. The characteristics of spray drop size, evaluated in terms of the variation of mean drop size along the spray axis, for the alternative fuel sprays remain unaffected by the variation in fuel properties between the alternative fuels and Jet A-1. The measurements on spray patternation, obtained using a mechanical patternator at a distance 5.1 cm from the atomizer exit, show an enhanced fuel concentration in the vicinity of spray axis region for the alternative fuel sprays discharging from the dual-orifice atomizer.

Numerical study of solidification of A356 aluminum alloy flowing on an oblique plate with experimental validation

Molten A356 aluminum alloy flowing on an oblique plate is water cooled from underneath. The melt partially solidifies on plate wall with continuous formation of columnar dendrites. These dendrites are continuously sheared off into equiaxed/fragmented grains and carried away with the melt by producing semisolid slurry collected at plate exit. Melt pouring temperature provides required solidification whereas plate inclination enables necessary shear for producing slurry of desired solid fraction. A numerical model concerning transport equations of mass, momentum, energy and species is developed for predicting velocity, temperature, macrosegregation and solid fraction. The model uses FVM with phase change algorithm, VOF and variable viscosity. The model introduces solid phase movement with gravity effect as well. Effects of melt pouring temperature and plate inclination on hydrodynamic and thermo-solutal behaviors are studied subsequently. Slurry solid fractions at plate exit are 27%, 22%, 16%, and 10% for pouring temperatures of 620 degrees C, 625 degrees C, 630 degrees C, and 635 degrees C, respectively. And, are 27%, 25%, 22%, and 18% for plate inclinations of 30, 45, 60, and 75, respectively. Melt pouring temperature of 625 degrees C with plate inclination of 60 generates appropriate quality of slurry and is the optimum. Both numerical and experimental results are in good agreement with each other. (C) 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Geomorphology reveals active decollement geometry in the central Himalayan seismic gap

The similar to 700-km-long ``central seismic gap'' is the most prominent segment of the Himalayan front not to have ruptured in a major earthquake during the last 200-500 yr. This prolonged seismic quiescence has led to the proposition that this region, with a population >10 million, is overdue for a great earthquake. Despite the region's recognized seismic risk, the geometry of faults likely to host large earthquakes remains poorly understood. Here, we place new constraints on the spatial distribution of rock uplift within the western similar to 400 km of the central seismic gap using topographic and river profile analyses together with basinwide erosion rate estimates from cosmogenic Be-10. The data sets show a distinctive physiographic transition at the base of the high Himalaya in the state of Uttarakhand, India, characterized by abrupt strike-normal increases in channel steepness and a tenfold increase in erosion rates. When combined with previously published geophysical imaging and seismicity data sets, we interpret the observed spatial distribution of erosion rates and channel steepness to reflect the landscape response to spatially variable rock uplift due to a structurally coherent ramp-flat system of the Main Himalayan Thrust. Although it remains unresolved whether the kinematics of the Main Himalayan Thrust ramp involve an emergent fault or duplex, the landscape and erosion rate patterns suggest that the decollement beneath the state of Uttarakhand provides a sufficiently large and coherent fault segment capable of hosting a great earthquake.

EXPERIMENTAL INVESTIGATION OF UNSTEADY FLOW IN A TRANSONIC UNI-STAGE AXIAL COMPRESSOR

Detailed steady and unsteady experimental measurements and analysis were performed on a Single stage Transonic Axial Compressor with asymmetric rotor tip clearance for studying the compressor stall phenomena. The installed compressor had asymmetric tip clearance around the rotor casing varying from about 0.65mm to 1.25mm. A calibrated 5-hole aerodynamic probe was traversed radially at exit of rotor and showed the characteristics of increased flow angle at lower mass flow rates for all the speeds. Mach number distribution and boundary layer effects were also clearly captured. Unsteady measurements for velocity were carried out to study the stall cell behavior using a single component calibrated hotwire probe oriented in axial and tangential directions for choke/free flow and near stall conditions. The hotwire probe was traversed radially across the annulus at inlet to the compressor and showed that the velocity fluctuations were dissimilar when probe was aligned axial and tangential to the flow. Averaged velocities across the annulus showed the reduction in velocity as stall was approached. Axial mean flow velocity decreased across the annulus for all the speeds investigated. Tangential velocity at free flow condition was higher at the tip region due to larger radius. At stall condition, the tangential velocity showed decreased velocities at the tip and slightly increased velocities at the hub section indicating that the flow has breakdown at the tip region of the blade and fluid is accelerated below the blockage zone. The averaged turbulent intensity in axial and tangential flow directions increased from free flow to stall condition for all compressor rated speeds. Fast Fourier Transform (FFT) of the raw signals at stall flow condition showed stall cell and its corresponding frequency of occurrence. The stalling frequency of about half of rotational speed of the rotor along with large tip clearance suggests that modal type stall inception was occurring.

Phase Inversion of Agitated Liquid-Liquid Dispersions in the Presence of Micrometer-Sized Particles

We have investigated the impact of partially wetting particles of tens of micrometers on inversion instability of agitated liquid liquid dispersions. Particles of this size can be easily separated from the exit streams to avoid downstream processing-related issues. The results show that the presence of hydrophilic particles in small quantities (volume fraction range of 2 X 10(-4) to 1.25 x 10(-2)) significantly decreases the dispersed phase fraction at which water-in-oil (w/o) dispersions invert but leaves the inversion of oil-in-water (o/w) dispersions nearly unaffected. The addition of the same particles after they are hydrophobized decreases the dispersed phase fraction at which o/w dispersions invert but leaves the inversion of w/o dispersions unaffected. These findings suggest an increased rate of coalescence of drops when particles wet drops preferentially and a marginal decrease when they wet the continuous phase preferentially. High-speed conductivity measurements on w/o dispersion show transient conduction of a few hundred milliseconds duration through voltage pulses. Close to the inversion point, voltage pulses appear at high frequency for even 7 cm separation between the electrodes. The presence of hydrophilic particles produces a nearly identical signal at a significantly lower dispersed phase fraction itself, close to the new lowered inversion point in the presence of particles. We propose formation of elongated domains of the conducting dispersed phase through a rapid coalescence-deformation-breakup process to explain the new observations. The voltage signal appears as a forerunner of inversion instability.

Ferromagnetism and the effect of free charge carriers on electric polarization in the double perovskite Y2NiMnO6

The double perovskite Y2NiMnO6 displays ferromagnetic transition at T-c approximate to 81 K. The ferromagnetic order at low temperature is confirmed by the saturation value of magnetization (Ms) and also validated by the refined ordered magnetic moment values extracted from neutron powder diffraction data at 10 K. This way, the dominant Mn4+ and Ni2+ cationic ordering is confirmed. The cation-ordered P2(1)/n nuclear structure is revealed by neutron powder diffraction studies at 300 and 10 K. Analysis of the frequency-dependent dielectric constant and equivalent circuit analysis of impedance data take into account the bulk contribution to the total dielectric constant. This reveals an anomaly which coincides with the ferromagnetic transition temperature (T-c). Pyrocurrent measurements register a current flow with onset near T-c and a peak at 57 K that shifts with temperature ramp rate. The extrinsic nature of the observed pyrocurrent is established by employing a special protocol measurement. It is realized that the origin is due to reorientation of electric dipoles created by the free charge carriers and not by spontaneous electric polarization at variance with recently reported magnetism-driven ferroelectricity in this material.

Transitions and blowoff of unconfined non-premixed swirling flame

The present experimental work reports the first observations of primary and secondary transitions in the time-averaged flame topology in a non-premixed swirling flame as the geometric swirl number S-G (a non dimensional number used to quantify the intensity of imparted swirl) is varied from a magnitude of zero till flame blowout. First observations of two transition types viz. primary and secondary transitions are reported. The primary transition represents a transformation from yellow straight jet flame (at S-G = 0) to lifted flame with blue base and finally to swirling seated (burner attached) yellow flame. Time-averaged streamline plot obtained from 2D PIV in mid-longitudinal plane shows a recirculation zone (RZ) at the immediate vicinity of burner exit. The lifted flame is stabilized along the vortex core of this RZ. Further, when S-G similar to 1.4-3, the first occurrence of vortex breakdown (VB) induced internal recirculation zone (IRZ) is witnessed. The flame now stabilizes at the upstream stagnation point of the VB-IRZ, which is attached to the burner lip. The secondary transition represents a transformation from a swirling seated flame to swirling flame with a conical tailpiece and finally to a highly-swirled near blowout oxidizer-rich flame. This transition is understood to be the result of transition in vortex breakdown modes of the swirling flow field from dual-ring VB bubble to central toroidal recirculation zone (CTRZ). The physics of transition is described on the basis of modified Rossby number (Ro(m)). Finally, when the swirl intensity is very high i.e. SG similar to 10, the flame blows out due to excessive straining and due to entrainment of large amount of oxidizer due to partial premixing. The present investigation involving changes in flame topology is immensely important because any change in global flame structure causes oscillatory heat release that can couple with dynamic pressure and velocity fluctuations leading to unsteady combustion. In this light, understanding mechanisms of flame stabilization is essential to tackle the problem of thermo-acoustic instability. (C) 2015 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Micro-shock waves generated inside a fluid jet impinging on plane wall

An Nd:glass laser pulse (18 ns, 1.38 J) is focused in a tiny area of about 100-mum diam under ambient conditions to produce micro-shock waves. The laser is focused above a planar surface with a typical standoff distance of about 4 mm, The laser energy is focused inside a supersonic circular jet of carbon dioxide gas produced by a nozzle with internal diameter of 2.9 mm and external diameter of 8 mm, Nominal value of the Mach number of the jet is around 2 with the corresponding pressure ratio of 7.5 (stagnation pressure/static pressure at the exit of the nozzle), The interaction process of the micro-shock wave generated inside the supersonic jet with the plane wall is investigated using double-pulse holographic interferometry. A strong surface vortex field with subsequent generation of a side jet propagating outward along the plane wail is observed. The interaction of the micro-shock wave with the cellular structure of the supersonic jet does not seem to influence the near surface features of the flowfield. The development of the coherent structures near the nozzle exit due to the upstream propagation of pressure waves seems to be affected by the outward propagating micro-shock wave. Mach reflection is observed when the micro-shock wave interacts with the plane wall at a standoff distance of 4 mm, The Mach stem is slightly deflected, indicating strong boundary-layer and viscous effects near the wall. The interaction process is also simulated numerically using an axisymmetric transient laminar Navier-Stokes solver. Qualitative agreement between experimental and numerical results is good.

Forebody Compressibility Research of Hypersonic Vehicle

Three kinds of forebody model of hypersonic vehicles were studied with numerical simulation method. It shows that the two- order compressive ramp model is the best selection among the three for its good evaluative parameters value at the cowl of the inlet . This model can provide higher value of flux coefficient and total pressure recovery coefficient and lower average Mach number compared with those of the other two models . Simultaneously different compressive angles may have different effects . The configuration which the firstorder of compressive angle is 4°and the second 5°is the optimum combination. Furthermore factors such as attack angle were concerned. Better result may be obtained with a range of attack angles . Based on the work above the integrated design for forebodyPinlet of a hypersonic vehicle was performed. The numerical result shows that this integrated model provides good flow field quality for inlet and engine work.

Application of Similarity Theory to the Characterization of Non-Transferred Laminar Plasma Jet Generation

Laminar-flow non-transferred DC plasma jets were generated by a torch with an inter-electrode insert by which the arc column was limited to a length of about 20 mm. Current–voltage characteristics, thermal efficiency and jet length, a parameter which changes greatly with the generating parameters in contrast with the almost unchangeable jet length of the turbulent plasma, were investigated systematically, by using the similarity theory combined with the corresponding experimental examination. Formulae in non-dimensional forms were derived for predicting the characteristics of the laminar plasma jet generation, within the parameter ranges where no transfer to turbulent flow occurs. Mean arc temperature in the torch channel and mean jet-flow temperature at the torch exit were obtained, and the results indicate that the thermal conductivity feature of the working gas seems to be an important factor affecting thermal efficiency of laminar plasma generation.

Evaluation of the Cumulative Formation of High-Speed Liquid Jets

This paper describes the generation of pulsed, high-speed liquid jets using the cumulation method. This work mainly includes (1) the design of the nozzle assembly, (2) the measurement of the jet velocity and (3) flow visualization of the injection sequences. The cumulation method can be briefly described as the liquid being accelerated first by the impact of a moving projectile and then further after it enters a converging section. The experimental results show that the cumulation method is useful in obtaining a liquid jet with high velocity. The flow visualization shows the roles of the Rayleigh-Taylor and Kelvin-Helmholtz instabilities in the breakup of the liquid depend on the jet diameter and the downstream distance. When the liquid jet front is far downstream from the nozzle exit, the jet is decelerated by air drag. Meanwhile, large coherent vortex structures are formed surrounding the jet. The liquid will break up totally by the action of these vortices. Experimental results showing the effect of the liquid volume on the jet velocity are also included in this paper. Finally, a method for measuring the jet velocity by cutting two carbon rods is examined.

Thermocapillary Effect in a Thermal Rheological Jet

The process of die swell in polymer jets is an important feature within polymer processing and can be explained through a study of its rheological effects. The existence of a thermocapillary effect, driven by the gradient of its surface tension, should be considered when examining a thermal jet that has a non-uniform temperature distribution on its free surface, as in various polymer processings. Both the rheological effect and thermocapillary effect on die swell can be studied numerically through a finite element method as used on a two-dimensional and unsteady model, in which a Coleman-Noll second-order fluid model is employed. The results show that the expanding angle depends on both the rheological property of the fluid and the pressure at the vessel exit. Although both the thermocapillary and the rheological effects contribute to the cross-section expansion of the fluid jet, the latter is more important in determining the expansion.

Computational study of viscous effects on lobed mixer flow features and performance

This article describes a computational study of viscous effects on lobed mixer flowfields. The computations, which were carried out using a compressible, three-dimensional, unstructured-mesh Navier-Stokes solver, were aimed at assessing the impacts on mixer performance of inlet boundary-layer thickness and boundary-layer separation within the lobe. The geometries analyzed represent a class of lobed mixer configurations used in turbofan engines. Parameters investigated included lobe penetration angles from 22 to 45 deg, stream-to-stream velocity ratios from 0.5 to 1.0, and two inlet boundary-layer displacement thicknesses. The results show quantitatively the increasing influence of viscous effects as lobe penetration angle is increased. It is shown that the simple estimate of shed circulation given by Skebe et al. (Experimental Investigation of Three-Dimensional Forced Mixer Lobe Flow Field, AIAA Paper 88-3785, July, 1988) can be extended even to situations in which the flow is separated, provided an effective mixer exit angle and height are defined. An examination of different loss sources is also carried out to illustrate the relative contributions of mixing loss and of boundary-layer viscous effects in cases of practical interest.