Wafer Defect Map Pattern Recognition Based on Improved ResNet

The defect detection of wafers is an important part of semiconductor manufacturing.The wafer defect map formed from the defects can be used to trace back the problems in the production process and make improvements in the yield of wafer manufacturing.Therefore,for the pattern recognition of wafer defects,this paper uses an improved ResNet convolutional neural network for automatic pattern recognition of seven common wafer defects.On the basis of the original ResNet,the squeeze-and-excitation(SE)attention mechanism is embedded into the network,through which the feature extraction ability of the network can be improved,key features can be found,and useless features can be suppressed.In addition,the residual structure is improved,and the depth separable convolution is added to replace the traditional convolution to reduce the computational and parametric quantities of the network.In addition,the network structure is improved and the activation function is changed.Comprehensive experiments show that the precision of the improved ResNet in this paper reaches 98.5%,while the number of parameters is greatly reduced compared with the original model,and has well results compared with the common convolutional neural network.Comprehensively,the method in this paper can be very good for pattern recognition of common wafer defect types,and has certain application value.

Multifractal Simulation of GeochemicalMap Patterns

Using a simple multifractal model based on the model De Wijs, various geochemical map patterns for element concentration values are being simulated. Each pattern is self-similar on the average in that a similar pattern can be derived by application of the multiplicative cascade model used to any small subarea on the pattern. In other experiments, the original, self-similar pattern is distorted by superimposing a 2-dimensional trend pattern and by mixing it with a constant concentration value model. It is investigated how such distortions change the multifractal spectrum estimated by means of the 3-step method of moments. Discrete and continuous frequency distribution models are derived for patterns that satisfy the model of De Wijs. These simulated patterns satisfy a discrete frequency distribution model that as upper bound has a continuous frequency distribution to which it approaches in form when the subdivisions of the multiplicative cascade model are repeated indefinitely. This limiting distribution is lognormal in the center and has Pareto tails. Potentially, this approach has important implications in mineral and oil resource evaluation.

Mapping-Based 3D Hexahedral Finite Element Mesh Generation Method

Mapping mesh generation is widely applied in pre-processes of Finite Element Method （FEM）. In this study, the basic 3D mapping equations by Lagrange interpolating function are founded. Based these equations, a mapping pattern library, which maps essential configurations e.g. line, circle, rotary body, sphere etc. to hexahedral FEM mesh, has been built. Then available FEM mesh will be generated by clipping and assembling the mapped essential objects. Study case illustrates that the proposed method is simple and efficient to generate valid FEM mesh for complex 3D engineering structure.

PATTERN MAPPING ON BETA-SPLINE SURFACES——BILINEAR MAPPING METHOD

Texture pattern mapping is one of the most important techniques for high quality image syn- thesis. It can largely enhance the visual richness of raster-scan images. In this paper is presented a new method of mapping planar texture pattern onto beta-spline curved surfaces——bilinear mapping method which can map planar texture pattern onto curved surfaces with less distortion, and also can fulfill the geometric transformation of the texture pattern on the curved surfaces by operating the pattern win- dow. It is valuable to both CAD/CAM in artistic field and computer graphics.

Application of time–frequency entropy from wake oscillation to gas–liquid flow pattern identification

Gas–liquid two-phase flow abounds in industrial processes and facilities. Identification of its flow pattern plays an essential role in the field of multiphase flow measurement. A bluff body was introduced in this study to recognize gas–liquid flow patterns by inducing fluid oscillation that enlarged differences between each flow pattern. Experiments with air–water mixtures were carried out in horizontal pipelines at ambient temperature and atmospheric pressure. Differential pressure signals from the bluff-body wake were obtained in bubble, bubble/plug transitional, plug, slug, and annular flows. Utilizing the adaptive ensemble empirical mode decomposition method and the Hilbert transform, the time–frequency entropy S of the differential pressure signals was obtained. By combining S and other flow parameters, such as the volumetric void fraction β, the dryness x, the ratio of density φ and the modified fluid coefficient ψ, a new flow pattern map was constructed which adopted S（1–x）φ and （1–β）ψ as the vertical and horizontal coordinates, respectively. The overall rate of classification of the map was verified to be 92.9% by the experimental data. It provides an effective and simple solution to the gas–liquid flow pattern identification problems.

Experimental Research on Gas-Liquid Two-Phase Spiral Flow in Horizontal Pipe

In view of the importance of gas-liquid two-phase spiral flow and the few research reports at home and abroad,the gas-liquid two-phase spiral flow patterns have been researched in a horizontal pipe with different parameters investigated by means of observation and a high-speed camera.Since the appearance of spiral flow makes the distribution of twophase flow more complicated,the flow patterns appearing in the experiments were divided into the Spiral Wavy Stratified Flow(SWS),the Spiral Bubble Flow(SB),the Spiral Slug Flow(SS),the Spiral Linear Flow(SL),the Spiral Axial Flow(SA),and the Spiral Dispersed Flow(SD) by the observations and with reference to the predecessors' research achievements.A flow pattern map has been drawn up.The influence of velocity,vane angle and vane area on flow pattern conversion boundary and pressure drop has been studied,with a solid foundation laid for the future research work.

Spatiotemporal characteristics of the sea level anomaly in the Kuroshio Extension using a self-organizing map

Satellite altimeter SSH data in the Kuroshio Extension （KE） region gathered during the period January 1993 to December 2014 are analyzed using self-organizing map （SOM） analysis. Four spatial patterns （SOM1, SOM2, SOM3, and SOM4） are extracted, and the corresponding time series are used to characterize the variation of the sea level anomaly. Except in some individual months, SOM1 and SOM2 with single-branch jet structures appear alternately during the periods 1993-1998 and 2002-2011. However, during 1999-2001 and 2012-2014, SOM3 and SOM4 with double-branch jet structures are dominant.The sea level anomalies exhibit interannual variations, while the KE stream demonstrates decadal variation. For SOM1, the change in the KE path is less evident, although the KE jet is strong and narrow. For SOM2, the KE jet is weakened and widened and its jet axis moves towards the southwest. Compared with the SOM3, for SOM4 the trough and ridge in the upstream KE region are deeper in the northeast-southwest direction, and accompanied by a jet weakening and splitting.This study shows that SOM analysis is a useful approach for characterizing KE variability.

EFFECT OF SURFACTANT ON TWO-PHASE FLOW PATTERNS OF WATER-GAS IN CAPILLARY TUBES

Flow patterns of liquid-gas two-phase flow were experimentally investigated. The experiments were carried out in both vertical and horizontal capillary tubes having inner diameters of 1,60 mm, The working liquid was the mixture of water and Sodium Dodecyl Benzoyl Sulfate （SDBS）. The working gas was Nitrogen. For the water/SDBS mixture-gas flow in the vertical capillary tube, flow-pattern transitions occurred at lower flow velocities than those for the water-gas flow in the same tube. For the water/SDBS mixture-gas flow in the horizontal capillary tube, surface tension had little effect on the bubbly-intermittent transition and had only slight effect on the plug-slug and slug-annular transitions. However, surface tension had significant effect on the wavy stratified flow regime. The wavy stratified flow regime of water/SDBS mixture-gas flow expanded compared with that of water-gas.