• 문화기술의 융합
• /
• 제8권6호
• /
• pp.867-871
• /
• 2022

최근 빠른 유행의 변화 속에서 디자인의 변화는 패션기업의 매출에 큰 영향을 미치기 때문에 기업들은 신제품디자인 선택에 신중할 수밖에 없다. 최근 인공지능 분야의 발달에 따라 패션시장에서도 소비자들의 선호도를 높이기 위해 다양한 기계학습을 많이 활용하고 있다. 우리는 선호도와 같은 추상적인 개념을 수치화함으로써 신제품 개발에 신뢰성을 높이는 부분에 기여하고자 한다. 이를 위해 3가지 적대적 생성 신경망(Generative adversial netwrok, GAN)을 통하여 기존에 없는 새로운 이미지를 생성하고, 미리 훈련된 합성곱 신경망(Convolution neural networkm, CNN)을 이용하여 선호도라는 추상적인 개념을 수치화시켜 비교하였다. 심층 컨볼루션 적대적 생성 신경망(Deep convolutional generative adversial netwrok, DCGAN), 점진적 성장 적대적 생성 신경망(Progressive growing generative adversial netwrok, PGGAN), 이중 판별기 적대적 생성 신경망(Dual Discriminator generative adversial netwrok, D2GAN)의 3가지 방법을 통해 새로운 이미지를 생성하였고, 판매량이 높았던 제품으로 훈련된 합성곱 신경망으로 유사도를 비교, 측정하였다. 측정된 유사도의 정도를 선호도로 간주하였으며 실험 결과 D2GAN이 DCGAN, PGGAN에 비해 상대적으로 높은 유사도를 보여주었다.

• 한국정밀공학회지
• /
• 제22권9호
• /
• pp.137-146
• /
• 2005

This paper describes a research work to develop a computer-aided design for the product made by progressive working of bending and piercing. An approach to the system for progressive working is based on the knowledge-based rules. Knowledge for the system is formulated from plasticity theories, experimental results and the empirical knowledge of field experts. The system has been written in AutoLISP on the AutoCAD with a personal computer and is composed of three main modules, which are input and shape treatment, flat pattern layout, strip layout modules. The system is designed by considering several factors, such as piercing and bending sequences by fuzzy set theory, complexities of blank geometry, punch profiles, and the availability of a press equipment. Strip layout drawing generated in the strip layout module is presented in 3-D graphic forms, including piercing and bending sequences with punch profiles divided into for external area. Results obtained using the modules enable the manufacturer for progressive working of electric products to be more efficient in this field.

• 대한토목학회논문집
• /
• 제12권1호
• /
• pp.51-58
• /
• 1992

본 연구의 목적은 피로파괴에 대한 해양 구조물의 체계신뢰도해석에 관한 연구이다. 제 2차 모멘트 신뢰도 방법에 기초하여 제안한 새로운 두 방법을 사용하였는데, 하나는 개선된 제 1계 신뢰도방법(IFORM)이고 다른 하나는 수정된 Probabilistic Network Evaluation Technique(MPNET)이다. 이상화된 병렬부재모형을 사용하여 부재당 다수의 연결절점으로 구성된 여러 부재들을 가진 경우에 대하여 진전하는 피로파괴에 대한 신뢰도해석들을 수행하였다. 수치해석결과로부터 본 논문에서 사용한 방법들이 재래적인 방법인, 제 1계 신뢰도방법(FORM)과 PNET 방법보다 더 정확한 결과를 얻을 수 있었다. 이러한 경향은 체계의 파괴모우드 수가 증가됨에 따라 더욱 분명하였다.

• Journal of Welding and Joining
• /
• 제28권6호
• /
• pp.28-34
• /
• 2010

This paper was described on investigation to prevent porosity in high speed MIG Welding of Al 6082-T6. Porosity measurement was carried out by using image analysis of micrographs with the help of an analysis software. The main parameter was arc length and torch progressive angle. The porosity ratio was increased as arc length was increased. The arc length was increased depending upon the output voltage. By proper selection of pulse waveform parameter, the stable arc of one pulse one drop was generated. The porosity ratio of optimum condition in one pulse one drop was lower than high voltage condition. When torch progressive angle was an angle of advance $10^{\circ}$, porosity ratio was minium.

• 복합신소재구조학회 논문집
• /
• 제5권4호
• /
• pp.1-10
• /
• 2014

A progressive failure analysis procedure for composite laminates is developed in here and in the companion paper. An anisotropic plastic constitutive model for fiber-reinforced composite material, is developed, which is simple and efficient to be implemented into computer program for a predictive analysis procedure of composites. In current development of the constitutive model, an incremental elastic-plastic constitutive model is adopted to represent progressively the nonlinear material behavior of composite materials until a material failure is predicted. An anisotropic initial yield criterion is established that includes the effects of different yield strengths in each material direction, and between tension and compression. Anisotropic work-hardening model and subsequent yield surface are developed to describe material behavior beyond the initial yield under the general loading condition. The current model is implemented into a computer code, which is Predictive Analysis for Composite Structures (PACS), and is presented in the companion paper. The accuracy and efficiency of the anisotropic plastic constitutive model are verified by solving a number of various fiber-reinforced composite laminates with and without geometric discontinuity. The comparisons of the numerical results to the experimental and other numerical results available in the literature indicate the validity and efficiency of the developed model.

• 한국정밀공학회지
• /
• 제27권2호
• /
• pp.50-59
• /
• 2010

In this paper, two kind of free-form progressive addition lenses (PALs) were designed with Zernike polynomial surface and anatomically accurate finite presbyopic schematic eyes which have aspheric cornea, aspheric GRIN crystalline lens, aspheric retina, and Gaussian apodization factor. Geometrical and diffraction MTFs were used for the optimization process in sequence. 5th orders of Zernike polynomials were used for the evaluation of progression zones of the two examples. The target MTF was set as 0.22 at 100 lp/mm which satisfies the standard visual resolution. These examples were fabricated with a CNC diamond turning machine controlled by slow tool servo (STS). After polishing process, the wavefront aberrations were measured with a laser interferometer on the ten test points across the progression zones and then compared with three current commercially available PALs on the optical performance. Astigmatic aberrations of the examples are very lower than the three selected PALs and have more increased stabilized progressive intermediate zones and near zones. It is expected to give better clear and comfortable distance, intermediate and near visions than other conventional PALs and to improve the adaptability of presbyopic patients to PALs.

• 근관절건강학회지
• /
• 제15권2호
• /
• pp.155-165
• /
• 2008

Purpose: The purpose of this study was to develop a progressive exercise program based on clinical guidelines for exercise prescription for older adults with osteoarthritis, which was planned to improve the existing arthritis self-help program. Method: We analyzed the contents of the arthritis self-help program and closely reviewed the guideline of exercise prescription for elderly with osteoarthritis. Results: The major contents of the revised arthritis self-help program are as follows: 1) The weekly education is composed of one-hour exercise and one-hour health education about various subjects. 2) The weekly main exercise is composed of muscle strengthening and endurance exercise. 3) Endurance exercise consists of three steps. 4) Muscle strengthening exercise is performed by gravity at first, and then the Thera-Band muscle strengthening exercise is followed on the 4th week. Conclusion: It is necessary to verify the effect of the progressive exercise program for older adults with osteoarthritis. It is thought that this revised program could be adopted as a rehabilitation program for older adults with osteoarthritis.

• 한국전산구조공학회논문집
• /
• 제24권5호
• /
• pp.569-576
• /
• 2011

본 논문에서는 비선형정적해석 방법을 이용하여 기존의 연쇄붕괴 선형정적해석 절차에 수반되는 반복과정을 자동으로 수행하면서도 선형해석 결과와 동일한 결과를 나타내는 새로운 연쇄붕괴 해석절차를 제안하였다. 제안된 해석절차는 GSA 기준에 규정된 기존 선형정적해석법의 단점인 반복적 해석작업 및 그 과정에서 발생할 수 있는 시간과 오류의 가능성을 최소화하여 해석결과의 신뢰성을 확보가 있는 선형정적해석 절차이다. 제안된 해석절차를 검증하기 위하여 철근콘크리트 모멘트골조 및 철골 가새골조의 최하층 기둥부재를 제거한 후 기존 해석법과 제안 해석법을 적용하고 그 결과를 비교 분석하였다. 해석결과에 따르면 제안된 해석절차는 구조물의 붕괴여부의 판정 및 힌지 분포에 있어서 기존의 선형정적해석과 동일한 결과를 나타내었으며, 반복해석 과정이 불필요하므로 기존 해석법에 비하여 매우 짧은 시간에 해석을 수행할 수 있는 것으로 나타났다.

• 한국임상수의학회지
• /
• 제36권5호
• /
• pp.259-265
• /
• 2019

This study was designed to investigate the effects of alpha-glucosyl rutin (G-rutin) and its comparative effects with other antioxidants (glutathione: GSH, catalase: CATA and beta-mercaptoethanol : ${\beta}ME$) on dog sperm freezing. In the first experiment (E1), the spermatozoa were diluted in freezing extender supplemented with 0 (control), 0.001, 0.01, or 0.1% G-rutin and frozen using liquid nitrogen ($LN_2$). The progressive motility, reactive oxygen species (ROS) level and apoptosis of spermatozoa were assessed after sperm thawing at $37^{\circ}C$ for 25 sec. In the second experiment (E2), 0.1% G-rutin group was compared with 10 mM ${\beta}-ME$, $5{\mu}M$ GSH and $50{\mu}M$ CATA groups by assaying progressive motility, viability and gene expression of Bcl-2 and SMCP after sperm freezing and thawing. In E1, 0.1% G-rutin group showed higher (P < 0.05) post-thaw progressive motility and lower (P < 0.05) ROS levels. In E2, the expressions of SMCP in G-rutin group were higher (P < 0.05) than in CATA group while Bcl-2 expression of G-rutin group was higher (P < 0.05) than ${\beta}-ME$ and CATA groups. However, there were no significant differences in progressive motility and viability. Therefore, we suggest that G-rutin can be used as a potentially antioxidative supplement in dog sperm freezing extender on the basis of gene expression related to motility and apoptosis as well as ROS level.

• 대한건축학회논문집:구조계
• /
• 제34권6호
• /
• pp.63-69
• /
• 2018

The machine learning model can capture the dynamics of building systems with less inputs than the first principle based simulation model. The training data for developing a machine learning model are usually selected in a heuristic manner. In this study, the authors developed a machine learning model which can describe supply air temperature from an AHU in a real office building. For rational reduction of the training data, the progressive sampling method was used. It is found that even though the progressive sampling requires far less training data (n=60) than the offline regular sampling (n=1,799), the MBEs of both models are similar (2.6% vs. 5.4%). In addition, for the update of the machine learning model, the normalized mutual information (NMI) was applied. If the NMI between the simulation output and the measured data is less than 0.2, the model has to be updated. By the use of the NMI, the model can perform better prediction ($5.4%{\rightarrow}1.3%$).