• International Journal of Korean Welding Society
• /
• 제1권2호
• /
• pp.51-60
• /
• 2001

A prediction method for determining the welding residual stress by artificial neural network is proposed. A three-dimensional transient thermo-mechanical analysis has been performed for the $CO_2$ arc welding using the finite element method. The first part of numerical analysis performs a three-dimensional transient heat transfer analysis, and the second part then uses the results of the first part and performs a three-dimensional transient thermo-elastic-plastic analysis to compute transient and residual stresses in the weld. Data from the finite element method are used to train a back propagation neural network to predict the residual stress. Architecturally, the fully interconnected network consists of an input layer for the voltage and current, a hidden layer to accommodate the failure mechanism mapping, and an output layer for the residual stress. The trained network is then applied to the prediction of residual stress in the four specimens. It is concluded that the accuracy of the neural network predicting method is fully comparable with the accuracy achieved by the traditional predicting method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 학술대회 논문집 정보 및 제어부문
• /
• pp.189-191
• /
• 2005

In the scalable extension of H.264/AVC, the codec is based on a layered approach to enable spatial scalability. In each layer, the basic concepts of motion compensated prediction and intra prediction are employed as in standard H.264/AVC. Additionally inter-layer prediction algorithm between successive spatial layers is applied to remove redundancy. In the inter-layer prediction, as the prediction we can use the signal that is the upsampled signal of the lower resolution layer. In this case, coding efficiency can be variable as the kinds of interpolation filter. In this paper, we investigate the approach to select the interpolation filter for residual signal in order to optimal prediction.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2006년도 춘계학술대회 논문집
• /
• pp.132-135
• /
• 2006

Injection molding is a flexible production technique for the manufacturing of polymer products, but introduces residual stresses. Residual stresses in a structural material or component are those stresses which exist in the object without other external loads. The layer removal and hole drilling method are used for the measurement of residual stress in injection molded polystyrene part. The hole drilling method is potentially more flexible for determining residual stress in complex geometries and can be used as an adoptable technique for the measurement of residual stress in polymeric materials. Results obtained by experiments agree with each other.

• 한국해양공학회지
• /
• 제15권1호
• /
• pp.45-51
• /
• 2001

In order to clarify the seasonal variation of the residual current and the material transportation process in Hiroshima Bay, JAPAN, the real-time simulation of residual current and particle tracking by using the Euler-Lagrange model were carried out. The calculated tidal current, water temperature, and salinity showed good agreement with the observed ones. The residual currents showed a southward flow pattern at the upper layer, and a northward flow pattern at the lower layer. The flow structure of the residual current in Hiroshima Bay is an estuarine circulation affected by density flow and wind driven current. The residual current plays an important role of material transportation in the bay.

• 한국멀티미디어학회논문지
• /
• 제23권7호
• /
• pp.803-811
• /
• 2020

This paper proposes an image classification algorithm that transforms the number of convolution layers in the residual block of ResNet, CNN's representative method. The proposed method modified the structure of 34/50 layer of ResNet structure. First, we analyzed the performance of small and many convolution layers for the structure consisting of only shortcut and 3 × 3 convolution layers for 34 and 50 layers. And then the performance was analyzed in the case of small and many cases of convolutional layers for the bottleneck structure of 50 layers. By applying the results, the best classification method in the residual block was applied to construct a 34-layer simple structure and a 50-layer bottleneck image classification model. To evaluate the performance of the proposed image classification model, the results were analyzed by applying to the cifar10 dataset. The proposed 34-layer simple structure and 50-layer bottleneck showed improved performance over the ResNet-110 and Densnet-40 models.

• 한국세라믹학회지
• /
• 제27권3호
• /
• pp.329-336
• /
• 1990

In this investigation, bar-shaped specimens which consisted of three layers are prepared to study the effects of residual compressive stress on the mechanical properties in ZTA. The outer layers contained Al2O3 and unstabilized ZrO2 and the central layer contained Al2O3 and stabilized ZrO2(with 5.10wt% Y2O3). When cooled from the sintering temperature, some of zirconia in the outer layers transformed to the monoclinic form while zirconia in the central layer was retained in the tetragonal form. The transformation which induces to dilatational expansion led to the estabilishmenet of compressive stress in the outer layers and balances tensile stress in the central layer. Decrease of outer layer thickness(for a fixed total thickness)increases residual compressive stress. Because of residual compressive stress in the outer layers, the fracture toughness of outer layers of 3-layers composite is 10.21 Mpam1/2, which is increased to 25% above in comparison with 1-layer specimens in ZTA. Also, the 3-layers composite is believed to exhibite greater fracture resistance in contact damage environment from thermal shock test.

• 반도체디스플레이기술학회지
• /
• 제14권3호
• /
• pp.61-66
• /
• 2015

Recently, the major trends of NIL are high throughput and large area patterning. For UV NIL, if it can be proceeded in the non-vacuum environment, which greatly simplifies tool construction and greatly shorten process times. However, one key issue in non-vacuum environment is air bubble formation problem. In this paper, numerical analysis of bubble defect of UV NIL is performed. Fluent, flow analysis focused program was utilized and VOF (Volume of Fluid) skill was applied. For various resist-substrate and resist-mold angles, effects of velocity inlet and residual layer thickness of resist on bubble defect formation were investigated. The numerical analyses show that the increases of velocity inlet and residual layer thickness can cause the bubble defect formation, however the decreases of velocity inlet and residual layer thickness take no difference in the bubble defect formation.

• 한국분말재료학회지
• /
• 제26권1호
• /
• pp.11-15
• /
• 2019

In this study, a method to remove residual powder on a multi-layered graphene and a new approach to transfer multi-layered graphene at once are studied. A graphene one-step transfer (GOST) method is conducted to minimize the residual powder comparison with a layer-by-layer transfer. Furthermore, a residual powder removing process is investigated to remove residual powder at the top of a multi-layered graphene. After residual powder is removed, the sheet resistance of graphene is decreased from 393 to 340 Ohm/sq in a four-layered graphene. In addition, transmittance slightly increases after residual powder is removed from the top of the multi-layered graphene. Optical and atomic-force microscopy images are used to analyze the graphene surface, and the Ra value is reduced from 5.2 to 3.7 nm following residual powder removal. Therefore, GOST and residual powder removal resolve the limited application of graphene electrodes due to residual powder.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1993년도 추계학술대회 논문집
• /
• pp.482-487
• /
• 1993

A prediction method for determining the welding residual stress by artificial neural network is proposed. A three-dimensional transient thermomechanical analysis has been performed for the CO $_{2}$ Arc Welding using the finite element method. The validity of the above results is demonstrated by experimental elastic stress relief method which is called Holl Drilling Method. The first part of numarical analysis performs a three-dimensional transient heat transfer anslysis, and the second part then uses results of the first part and performs a three-dimensional transient thermo-clasto-plastic analysis to compute transient and residual stresses in the weld. Data from the finite element method were used to train a backpropagation neural network to predict residual stress. Architecturally, the finite element method were used to train a backpropagation voltage and the current, a hidden layer to accommodate failure mechanism mapping, and an output layer for residual stress. The trained network was then applied to the prediction of residual stress in the four specimens. The results of predicted residual stress have been very encouraging.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 춘계학술대회논문집
• /
• pp.102-106
• /
• 1999

One of the main causes of unwanted dimensional changes in precision metal mold casting parts is excessive and irregular residual stresses induced by temperature gradients and plastic deformation in the solidifying shell. Residual stresses can also cause stress cracking and lower the fatigue life and fracture strength of the casting parts,. In the present study aluminum alloy casting system with metal mold equipped with electrical heating elements and water cooling units was designed and the casting specimens were produced to quantify the effects of different cooling conditions on the development of residual stresses. the layer removal method was used to measure the biaxial residual stresses in casting specimens produced from the experiments. The experimental results agreed with Tien-Richmond's theoretical model for thermal stress development for the solidifying metal plate