Influence functions of two families of robust estimators under proportional scatter matrices

In this paper, under a proportional model, two families of robust estimates for the proportionality constants, the common principal axes and their size are discussed. The first approach is obtained by plugging robust scatter matrices on the maximum likelihood equations for normal data. A projection- pursuit and a modified projection-pursuit approach, adapted to the proportional setting, are also considered. For all families of estimates, partial influence functions are obtained and asymptotic variances are derived from them. The performance of the estimates is compared through a Monte Carlo study. © 2006 Springer-Verlag.

Robust plug-in estimators in proportional scatter models

We address the problem of estimating the principal axes and their size in the case of several populations under the assumption of a proportional model. We propose robust estimators for the common principal axes and their size. The robust estimators are based on asymptotically normal and equivariant robust scatter estimators. The asymptotic distribution of the robust estimators including the proportionality constants are derived. © 2003 Elsevier B.V. All rights reserved.

The influence of solute carbon in cold-rolled steels on shear band formation and recrystallization texture

Two experiments were conducted to clarify the roles of grain size, solute carbon and strain in determining the recrystallization textures of cold-rolled and annealed steels. In the first experiment, samples of coarse-grained low-carbon (LC) and interstitial-free (IF) steels were cold-rolled to a 75% reduction in thickness. One sample from each steel was polished and cold-rolled an additional 5%, while the remaining samples were annealed for various times at 650°C. In the second experiment, three samples from a commercial LC steel sheet were rolled 70% at 300°C. Two of the samples were given a further rolling reduction of 5% of the original thickness, with one of the samples being given this additional reduction at 300°C and the other at room temperature. Goss recrystallization textures are strengthened by coarse initial grain sizes, the presence of solute carbon and rolling at a temperature where dynamic strain ageing occurs, but are weakened by additional rolling beyond a reduction of 70%, especially when this extra rolling is conducted at a temperature where dynamic strain ageing does not occur. Characterization of key features of the deformed and recrystallized steels using optical microscopy, scanning electron microscopy (SEM) and electron back-scatter diffraction (EBSD) supports a rationale for these effects based on the repeated activation and deactivation of shear bands and the influence of solute carbon and dynamic strain ageing on the operating life of the bands and the accumulation of strain within them.

In-situ investigation of twinning behaviour in Mg-3Al-1Zn

Magnesium alloys are attractive for automotive and aerospace industries, due to their low density. One problem with these alloys is their limited formability at room temperature. Twinning plays a dominant role in deformation behaviour and it can be expected that an increased understanding of twinning will help improve formability. In the present work, the behaviour of different twinning systems in as-cast Mg-3AI-IZn is investigated using in-situ tensile tests in a scanning electron microscope. Electron backscatter diffraction and back scatter electron imaging were carried out during the tests. The results show both "tension" and "compression" twinning are active at room temperature and that twinning and untwinning occur both during loading and unloading.

Dynamic softening of ferrite during large strain warm deformation of a plain-carbon steel

The evolution of dynamic ferrite softening in a plain-carbon steel was investigated by torsion tests during warm deformation at 810 °C, in the two-phase (ferrite + austenite) region, and strain rate of 0.1 s−1 with different strains up to 50. The warm flow behaviour and ferrite microstructural parameters, such as grain size, misorientation angle across ferrite/ferrite boundaries, and the fraction of high-angle and low-angle grain/subgrain boundaries were quantified using electron back scatter diffraction. The results show that with increasing strain up to not, vert, similar2, the ferrite grain size and fraction of high-angle boundaries rapidly decrease and the fraction of low-angle boundaries increases. However, these parameters remain approximately unchanged with increasing strain from not, vert, similar2 to 50. The dynamic softening mechanism observed during large strain ferritic deformation is explained by dynamic recovery and continuous dynamic recrystallization.

A signal processing approach for elastic wave-based structural health monitoring using active piezoelectrics

A series of digital frequency filters (DFFs) were designed to screen diverse noises and the spectrographic analysis was conducted to isolate complex boundary reflection, which obscures the damage-induced signals. The scale-averaged wavelet power (SAP) technique was applied to enhance
the measurement accuracy of Time of Flight (TOF). As an example, the propagation characteristics of elastic wave in a structural beam of square cross-section were analyzed using such an approach and verified experimentally and numerically, with the consideration of the complicated wave scatter caused by the non-ignorable section dimensions.

Individual discriminative subspace for face recognition under uncontrolled conditions

Most face recognition (FR) algorithms require the face images to satisfy certain restrictions in various aspects like view angle, illumination, occlusion, etc. But what is needed in general is the techniques that can recognize any face images recognizable by human beings. This paper provides one potential solution to this problem. A method named Individual Discriminative Subspace (IDS) is proposed for robust face recognition under uncontrolled conditions. IDS is the subspace where only the images from one particular person converge around the origin while those from others scatter. Each IDS can be used to distinguish one individual from others. There is no restriction on the face images fed into the algorithm, which makes it practical for real-life applications. In the experiments, IDS is tested on two large face databases with extensive variations and performs significantly better than 12 existing FR techniques.

Modelling the diffuse fraction of global solar radiation on a horizontal surface

For various applications it is necessary to know not only global solar radiation values, but also the diffuse and beam components. Because often only global values are available, there have been several models developed to establish correlations between the diffuse fraction and various predictors. These typically include the clearness index, but also may include the solar angle, temperature and humidity. The clearness index is the proportion of extraterrestrial radiation reaching a location, where the extraterrestrial value used in the calculation varies with latitude and time of year. These correlations have been developed using data principally from latitudes greater than 40°, often using only data from a few locations and with few exceptions have not used solar altitude as a predictor. Generally the data consist of hourly integrated values. A model has been developed using hourly data from a weather station set up at Deakin University, Geelong. Another model has also been developed for 15 minute data values in order to ascertain if the smoothing generated by using hourly data makes a significant difference to overall results. The construction of such models has been investigated, enabling an extension to the research, inclusive of other stations, to be performed systematically. A final investigation was carried out, using data from other Australian locations, to explain some of the considerable scatter by adding apparent solar time as a predictor, which proved to be significantly better than solar altitude.

High temporal resolution observations of the pulsar microstructure at 102 MHz

We report observations of the microstructure of 7 pulsars at 102 MHz with a time resolution of up to 1 μs. We have detected fine pulse structure with a μs time-scale for all seven pulsars. We observe that interstellar scattering does not broaden the μs time-scale pulsar microstructure, and find structure of a time-scale much shorter than the interstellar scatter broadening. In addition to the previously known structure with time-scale τ_μ of tens of μs, we have detected structures with characteristic time-scale of a few μs, nearly equal to the reciprocal of the analysis bandwidth. This suggests that even shorter micropulses may exist.
Micropulses usually cluster in groups, each of duration tens to hundreds of μs, which may be identified as microstructure observed previously with smaller time resolution.
In addition to the previously known quasi-periodicity of hundreds of milliseconds, we reveal a structures with characteristic period of tens of μs. We observe structures with short quasi-periodicity to have a limited bandwidth. The characteristic period P_μ of these structures in PSR B0950+08 does not have a common value for different frequency regions in the same single pulse.
Scattering was revealed as a train of quasi-periodic micropulses with an exponentially damped envelope. The scatter-broadening and decorrelation bandwidth were measured. We report two time-scales of decorrelation bandwidth of pulsar PSR B0950+08 and propose two branches of Δν_d(ν) dependence.
The dispersion measure of PSR B0950+08 as DM= 2.9686 ± 0.0001 pc cm⁻³ has been determined.

Ploughzone archaeology on an Australian historic site

Archaeologists are often confronted with sites featuring post-occupation disturbance. At rural sites, this disturbance often comes in the form of agricultural activity, such as ploughing and grazing. These disturbances can call into question the value of site spatial relationships and broader data integrity. Between 2006 and 2007, archaeologists from La Trobe University and New Zealand-based consultancy firm Geometria carried out a programme of fieldwork at an 18411861 cottage in Gippsland, Victoria. The site is now an open grazing paddock that has been ploughed on several occasions in the past. The survey techniques used by the archaeological team, which included geomagnetic survey and artefact surface scatter mapping, allowed for testing the integrity of the ploughed archaeological deposits prior to excavation, and provide a case study for the applicability of ploughzone archaeology techniques to Australian historic sites.

Nutrient dynamics in Victorian rural soils

Experimentation indicates that water and nutrients move through heterogeneous matrix flow pathways and that nutrients show little variation in concentration with fertilizer application rate after 2 months. However a scatter in concentrations independent of application rates after 4 and 7 months, indicates that external factors are affecting nutrient concentrations within the soil after this time period. Grassmere and Rutherglen were the sample sites.

Dielectrophoretically tuneable optical waveguides using nanoparticles in microfluidics

We present a tuneable optical waveguide using dielectrophoretically controlled nanoparticles in microfluidics. Silicon dioxide nanoparticles of different sizes in de-ionized water are channelled through a microfluidic system. An array of microelectrodes generates the dielectrophoretic force to funnel nanoparticles, forming narrowbands at the center of the microfluidics at different applied voltages and frequencies. It is observed that these narrowbands either scatter or guide the coupled light under selected conditions. The realization of such a system offers exciting possibilities for the development of a new class of optofluidics, which are tuned by the positioning of nanoparticles on demand.

Characterising material and process variation effects on springback robustness for a semi-cylindrical sheet metal forming process

Variation in the incoming sheet material and fluctuations in the press setup is unavoidable in many stamping plants. The effect of these variations can have a large influence on the quality of the final stamping, in particular, unpredictable springback of the sheet when the tooling is removed. While stochastic simulation techniques have been developed to simulate this problem, there has been little research that connects the influence of the noise sources to springback. This paper characterises the effect of material and process variation on the robustness of springback for a semi-cylindrical channel forming operation, which shares a similar cross-section profile as many automotive structural components. The study was conducted using the specialised sheet metal forming package AutoForm^TM Sigma, for which a series of stochastic simulations were performed with each of the noise sources incrementally introduced. The effective stress and effective strain scatter in a critical location of the part was examined and a response window, which indicates the respective process robustness, was defined. The incremental introduction of the noise sources allows the change in size of the stressstrain response window to be tracked. The results showed that changes to process variation parameters, such as BHP and friction coefficient, directly affect the strain component of the stressstrain response window by altering the magnitude of external work applied to forming system. Material variation, on the other hand, directly affected the stress component of the response window. A relationship between the effective stressstrain response window and the variation in springback was also established.

On the stress function-based OWA determination method with optimization criteria

The ordered weighted averaging (OWA) determination method with stress function was proposed by Yager, and it makes the OWA operator elements scatter in the shape of the stress function. In this paper, we extend the OWA determination with the stress function method using an optimization model. The proposed method transforms the OWA optimal solution elements into the interpolation points of the stress function. The proposed method extends the basic form of the stress function method with both scale and vertical shift transformations.We also explore a number of properties of this optimization-based stress function method. The OWA operator optimal solution elements can distribute as the shape of the given stress function in a parameterized way, in which case, the solution always possesses the arithmetic average operator as a special case.