• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.21 no.6
• /
• pp.871-877
• /
• 2012

New materials widely used for automobile related industry, aircraft, space development area are mostly high hardness materials. The hardness value of some hardened materials is over HRC45 and machining of this hardened materials is called as hard turning. Hard turning has its advantage on processing flexibility, cycle time and tool cost reduction. Also this process obtains high efficiency in processing and precise surface roughness through application of the CBN tools. In hard turning process with CBN tool, surface integrity is the important factor for considering the design of machine part and component under high stress and load conditions. A purpose of this study is to analyze optimal condition in hard turning process of AISI 52100 steel (HRC62) with high CBN and low CBN on turning characteristics, tool wear mechanism comparison and surface integrity.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.69-72
• /
• 2002

본 논문에서 화자공간모델 evolution에 기반한 continuous density hidden Markov model (CDHMM)의 online 적응에 대한 새로운 기법을 제안한다. 학습화자의 a priori knowledge을 나타내는 화자공간모델은 factor analysis (FA) 또는 probabilistic principal component analysis (PPCA)와 같은 은닉변수모델(latent variable model)에 의해 효과적으로 나타내어진다. 은닉 변수모델은 화자공간모델뿐아니라 CDHMM 파라메터의 ajoint prior분포를 표시함으로, maximum a posteriori(MAP)적응기법에 직접 적용되어진다. 화자공간모델의 hyperparameters와 CDHMM파라메터를 동시에 순차적으로 적응하기 위해 quasi-Bayes (QB)추정 기술에 기반한 online 적응기법을 제안한다. 연속숫자음 인식과 관련된 화자적응 실험을 통해 제안된 기법은 적은 적응데이터에서 좋은 성능을 나타내며, 데이터가 증가함에 따라 성능이 지속적으로 증가함을 보여준다.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.1 no.2
• /
• pp.173-181
• /
• 1989

Until recently there was a lack of reliable performance data for the design and operation of Ondol heating systems. This paper presents a calculation method on heat flux from Ondol floor surface. Total heat flux from floor consists of radiation and convection component. In order to analyse the characteristics of both radiation and convection heat flux, each surface temperature is measured and several temperatures near each wall are measured vertically and horizontally in a practical Ondol heating space. Radiation heat flux is calculated and analysed by Gebhart's Absorption Factor Method with the consideration of instantaneous radiant exchanges. Convection heat output is derived from the vertical temperature profiles near floor. The vertical temperature profiles could be expressed by nonlinear regression equation models and convection coefficients could be estimated by the equations. As a result, radiation, convection and total heat flux are suggested by the expression of difference between floor surface and room air temperature.

• Journal of the Korean Statistical Society
• /
• v.11 no.1
• /
• pp.36-44
• /
• 1982

A new design concept, called axis-slope-rotatability, is presented for the design of experiments with mixtures. This is an analogue of the Box-Hunter (1957) rotatability for second order response surface designs. By choice of design, it is possible to make the variance of the estimated slopes along the component axes constant for all axial points equidistant from the center point of the factor space. This property is called axis-slope-rotatability for mixture experiments. When the Scheffe's second degree polynomial is used, it is shown that some symmetry conditions are sufficient for axis-slope-rotatability. Several designs having this property are illustrated.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.5
• /
• pp.531-536
• /
• 2021

As artificial intelligence technology advances, it is being applied to various application fields. Artificial intelligence is performing well in the field of image recognition and classification. Chip design specialized in this field is also actively being studied. Artificial intelligence-specific chips are designed to provide optimal performance for the applications. At the design task, memory component optimization is becoming an important issue. In this study, the optimal algorithm for the memory size exploration is presented, and the optimal memory size is becoming as a important factor in providing a proper design that meets the requirements of performance, cost, and power consumption.

• KIEAE Journal
• /
• v.17 no.1
• /
• pp.69-75
• /
• 2017

Purpose: Heat transfer analysis of recently developed 'slim type double-skin system window' were presented. This window system is designed for curtain wall type façade that main energy loss factor of recent elegant buildings. And the double skin system is the dual window system integrated with inner shading component, enclosed gap space made by two windows when both windows were closed and shading component effectively reflect and terminate solar radiation from outdoor. Usually double-skin system requires much more space than normal window systems but this development has limited by 270mm, facilitated for curtain wall façade buildings. In this study, we estimated thermophysical phenomena of our double-skin curtain wall system window with solar load conditions at the summer season. Method: A fully 3-Dimentional analysis adopted for flow and convective and radiative heat transfer. The commercial CFD package were used to model the surface to surface radiation for opaque solid region of windows' frame, transparent glass, fluid region at inside of double-skin and indoor/outdoor environments. Result: Steep angle of solar incident occur at solar summer conditions. And this steep solar ray cause direct heat absorption from outside of frame surface rather than transmitted through the glass. Moreover, reflection effect of shading unit inside at the double-skin window system was nearly disappeared because of solar incident angle. With this circumstances, double-skin window system effectively cuts the heat transfer from outdoor to indoor due to separation of air space between outdoor and indoor with inner space of double-skin window system.

• Tuberculosis and Respiratory Diseases
• /
• v.67 no.6
• /
• pp.569-573
• /
• 2009

Idiopathic pulmonary alveolar proteinosis (PAP) is a rare disorder characterized by surfactant component accumulation in the alveolar space. Idiopathic PAP has recently been recognized as a autoimmune disease of impaired alveolar macrophage function caused by autoantibodies against granulocyte-macrophage colony-stimulating factor (GM-CSF). While whole lung lavage has been the standard treatment, not every patient shows a complete response. Subcutaneous injection or inhalation of GM-CSF is another promising treatment option for PAP. A 45-year-old patient visited our hospital for dyspnea, he was diagnosed as PAP and underwent whole lung lavage. Eighteen months later, the patient had not achieved complete remission in despite of initial response. After then he was administered with GM-CSF (5 ${mu}g/kg/day$, subcutaneous injection) for fivetimes a week during 2 months. Nine months later, the abnormal shadows in high-resolution computed tomography (HRCT) decreased and the patient fully recovered in forced vital capacity. After 60 months, the HRCT scan showed complete remission of PAP.

• Journal of Environmental Impact Assessment
• /
• v.2 no.1
• /
• pp.39-55
• /
• 1993

The importance of urban and regional natural park increases because of the needs for preserving the natural resources and providing with natural recreation space in nature. This planning of natural park management should be established based on the research of the various natural resources in the park. But for the lack of effective data synthesizing methods and concepts, only some restricted factors for zoning plan are considered even though GIS computer system for large complex simulation is used. Therefore, in this study three ecological zoning models such as Basic Factor Model (BFM), Visual Landscape Model (VLM) and Comprehensive Ecological Model (CEM) are proposed and applied to Byounsan Peninsula Nature Park(BPNP) for comparison with the current natural park zoning. The BFM has three components -elevation, slope and vegetation. The VLM has applied with six components -elevation, slope, vegetation, road type, and the visual distance. Finally the CEM's modelling factors have included all of BFM, VLM components are added with the land use type, nature and historic resource factors. The zoning concept of BPNP was based on "Minimization" focused on the specific factors. But introduced modelling concept is "Optimization" based on the total ecological environment. So the result of the modelling has larger area for preservation and development zoning compared with the current zoning whose characteristics are ambiguous which allows the environmental destruction. The future study issues will be the determination of the weighting factor, component reconsideration based on the ground truth data and the agriculture residential area zoning.

• Archives of design research
• /
• v.19 no.2 s.64
• /
• pp.99-108
• /
• 2006

Based on the results from the already published 'A study on factors that make busy of street space' and 'A study on street the image evaluation of streetscape', this study is an attempt to explore factors, other than the structural factors, that compose the street space and affect the image of street to be more lively. First of all, this study was mainly concentrated on the structure of street space that fits the theory of the previous two theses, stating that the structural ratio ($D/H=0.5{\sim}2$) gives the street an interesting image. The next study subject was the street space that exhibits the amenity and busy of image according to the space structure ratio. I defined that exhibiting amenity and busy means the activation of the street space, and I attempted to extract the activation factors from the component elements. The street space that shows amenity and busy image after the activation was named as 'lively street space' in this study. Furthermore, I selected 20 street spaces, after classifying the whole, according to nations and local characteristics as the previous theses had done and looked for the 'lively street space', whose structural ratio was not in the range of $D/H=0.5{\sim}2$ and the factors that contributed to the Image. As the result, I founded that in case of the business areas with the ratio of $D/H=0.5{\sim}2$, street activation factors were hydroponic facilities, sidewalks, and wayside buildings and In case of the commercial areas, the factors were sidewalk, wayside buildings, hydroponic facilities, and illumination facilities. Especially, 5 commercial areas in Korea and 1 business areas in Japan did not have the structural ratio of $D/H=0.5{\sim}2$, but still exhibited lively image as streets. This was because aside from the structural element, other street activation factors such as facilities also had major contribution in these streets. In other words, in commercial areas in Korea have wayside buildings, sidewalks, and hydroponic facilities as activating factors, whereas in street spaces in business areas in Japan, hydroponic facilities, wayside buildings, and sidewalk factors are influential to the activation of street spaces.

• Industrial Engineering and Management Systems
• /
• v.13 no.4
• /
• pp.454-462
• /
• 2014

Retention of possible churning customer is one of the most important issues in customer relationship management, so companies try to predict churn customers using their large-scale high-dimensional data. This study focuses on dealing with large data sets by reducing the dimensionality. By using six different dimension reduction methods-Principal Component Analysis (PCA), factor analysis (FA), locally linear embedding (LLE), local tangent space alignment (LTSA), locally preserving projections (LPP), and deep auto-encoder-our experiments apply each dimension reduction method to the training data, build a classification model using the mapped data and then measure the performance using hit rate to compare the dimension reduction methods. In the result, PCA shows good performance despite its simplicity, and the deep auto-encoder gives the best overall performance. These results can be explained by the characteristics of the churn prediction data that is highly correlated and overlapped over the classes. We also proposed a simple out-of-sample extension method for the nonlinear dimension reduction methods, LLE and LTSA, utilizing the characteristic of the data.