Search results

Grain size, resistivity and thermal conductivity of the boron doped diamond used are 10μm, 5~37×10Ω·m and 440~580W/m·K.
Boron doped diamond particles with a grain size of φ10μm were sintered together with cobalt as a solvent (catalyst).
Evaluation items were number of times of wire breakage, machining speed, surface roughness and surface integrity.

The Ti-adsorbed ZnO (1120) surfaces have metallic characteristics because of the number of outer electron transferred from Ti to O is extremely limited and the area of negative charge of the surface is large.
In general, the segregation properties of the transition elements applied on the ZnO surface also determine the metal-ZnO contacts [4-6] and conduction properties of grain boundaries [7].
However, the (1120) surfaces is constituted by the chain of Zn and O consisting of alternating and the ditches in the [0001] direction, the lattice plane which perpendicular to the (110) direction includes the same number of Zn and O atoms, and all are parallel dimer exists in ZnO.
Each atoms on the surface lost one position in four-coordinated, and thus become a three-coordinated, and it bonding with one atom of the surface and two atoms of the sub-surface. the results of this paper indicate that the geometry of (1120) surfaces is similar to bulk phase, their surface relaxation is small, and(1120) surfaces relaxation, the number of bonding and the number of dangling bonds is same in this two surface, but the surface structure is different, M.Calatayud [8] think that is due to (1010) plane shear needs to be rotated 18°, but (1120) surfaces needs to be rotated 34.6° caused.
Ti-O shows obviously better performance than Zn-O in covalent bond. the Ti-adsorbed ZnO (1120) surfaces have metallic characteristics because of the number of outer electron transferred from Ti to O is extremely limited and the area of negative charge of the surface is large.

Each layer can be associated with a vector of numbers, called neurons.
One layer can be composed of any number of filters of the same size.
The idea is to add outputs from i-th layer to (i+k)-th layer, where k is usually small in comparison to number of layers in the whole network.
We changed pretrained architecture replacing the last dense layers with convolutional ones with filter of the size of 1 in first two dimensions, creating so called Fully Convolutional Network (FCNN) [5] with number of filters in the last layer corresponding to the number of classes in the data set.
This way, each output has a size equal to number of classes x 36 x 48, and a tensor of size number of classes x 1 x 1 can be understood as evaluation of classification for small part of input image.

Dislocations, grain-boundaries, phase boundaries, and free surfaces are other types of defects of crystalline solids.
Both the numbers of participants and papers showed a large increase, from 174 to 272 and from 153 to 317, respectively.
Extensive investigations on various problems of diffusion in solids have been carried out at the institute by a number of scientists, in particular by Chr.
A number of contributions to these Proceedings from the 'Munster school' confirm that the institute is one of the leading teams in diffusion research.

It is based on minimization of the following objective function: ， where m is any real number greater than 1, uij is the degree of membership of xi in the cluster j, xi is the ith of d-dimensional measured data, cj is the d-dimension center of the cluster, and ||*|| is any norm expressing the similarity between any measured data and the center.
It is suggested that the grid takes a reasonably coarse-grain scale so that the mapping would not result in too many fine data points.
The number of clusters could be decided by estimating the probability density contours for the two-component mixture distribution.
Considering the scale of charging requirement spots, clustering number is chosen six.

The oil field unites a number of local structures of the II and IV orders: Asomkinskaya, Vostochno-Asomkinskaya, Sredne-Asomkinskaya, Zapadno-Fainskaya and a number of smaller nameless V-order highs.
The YUS11 formation sedimentation is mainly presented by polymictic, small-grain, moderate- and poorly-sorted, clayish, with mudstone interbeds and dense aleurolite sandstones.
The median size of the sandstone grains and aleurolites varies depending on the structural characteristics and lies within 0.12-0.21 mm (sandstones) and 0.09-0.16 mm (aleurolites).
The presence of sediments with high structural maturity (medium-grained sandstones with high degree of assortment) predetermines the significant, though local, growth of the reservoir rock permeability.
When all these requirements are observed on the Jurassic formations of the Ugutskoye field, in a number of cases with acid BFZP the qoil increment was 30-50 tons/day.

Sampling temperature, which is one of the important parameters, has been found to influence the releasing rate of the volatiles from rice grain and could be optimized to obtain the best results.
Although a large number of odorous molecules in the rice aroma have been determined, only a few have been found to contribute to the characteristics of rice aroma [1].
Sample temperature, which is one of the important measurement parameters, has been found to influence the releasing rate of the volatiles from the rice grain and can effect to the e-nose results.
For future prospect, we need to increase a number of components in the sensor array to improve efficiency of the e-nose as well as to develop a commercial prototype.

A number of different aging/heat treating procedures have been utilised that have resulted in significantly different mechanical properties.
Introduction Rolling at cryogenic temperatures (cryo-rolling) has been demonstrated to be a successful technique to produce materials with very fine grain sizes [1-3] and such materials offer a unique set of properties [4].
When examining the solution treated material it could be seen to contain a population of particles that are known as the T phase [6] and these could be observed using the AsB detector due to the atomic number contrast, Fig. 1b.

From the existing literature, a significant number of closely related works are found.
The strength of the line source was calculated as a function of Peclet number and conductivity with the assumption of a conical profile of the keyhole.
They also show that the weld shape is a strong function of Peclet number and conductivity of material.
The model is validated by predicting the weld-bead dimensions and comparing them with experimental data [43].In an investigation, it has been revealed that that the HAZ grain boundary migration rate is approaching the upper theoretical limit (represented by a time exponent n = 0.5 in the grain growth equation) owing to the high driving forces involved.
In the presence of extensive phosphorus and nitrogen segregations, a transition in the grain boundary velocity has, however, been observed during high heat input weld simulation, which largely will prevent further coarsening of the austenite grains.

To maximize their error correction strength, LDPC codes demand NAND flash memory carry out fine-grained (i.e., soft-decision) sensing, which nevertheless results in a significant read latency penalty.
The speculation error could of course degrade the decoding error correction strength and/or increase the number of decoding iterations, where the latter could offset the potential decoding speed improvement.
Although soft-decision message-passing LDPC code decoding can achieve very strong error correction strength, it demands fine-grained soft-decision flash memory on-chip sensing.
As we count the the number of failed parity checks in which each bit participates (i.e., uik), we immediately compare the current uik with the speculated uspck, and l If uik equals to uspck, flip the hard decision of current bit; l If uik > uspck, update uspck by setting uspck = uik and flip the hard decision of current bit.
LDPC decoder ASIC design result Decoder Logic Memory Total BF (mm2) 0.281 0.594 0.875 SBF (mm2) 0.273 0.149 0.422 Reduction (%) 2.7 75.0 51.8 With the same clock frequency, LDPC decoding throughput depends on two factors, including (i) the number of clock cycles per decoding iteration, and (ii) the average number of decoding iterations.