• Structural Engineering and Mechanics
• /
• 제30권1호
• /
• pp.77-97
• /
• 2008

In this paper, a strut-and-tie model approach has been proposed to directly calculate the amount of reinforcements in deep beams, and the force transfer mechanisms for this approach were investigated using linear finite element analysis. The proposed strut-and-tie model provides quite similar force transfer mechanisms to the results of linear finite element analysis for the 28 deep beams. The load-carrying capacities calculated from the proposed method are both accurate and conservative with little scatter or trends for the 214 deep beams. The deep beams have different concrete strengths including high-strength, various combinations of web reinforcements, and wide range of and a/d ratios. Good accuracy was also obtained using VecTor2, nonlinear finite element analysis tool based on the Modified Compression Field Theory. Since the proposed method provides a safe and reliable means for design of deep beams, this can serve to improve design provisions in future adjustments and development of design guidelines.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1996년도 봄 학술발표회 논문집
• /
• pp.40-46
• /
• 1996

Three procedures are currently used for the design of deep beams, which are Empirical design method, Nonlinear analysis, and Truss models. The engineering logic and decisions inherent in these design procedures are dependent on the acquired knowledge and experience of the structural engineer. Knowledge-based system is useful to solve problems which require human experties. Therefore, this study presents an application of Knowledge-Based System for design of reinforced concrete deep beams with web openings.

• 실천공학교육논문지
• /
• 제11권2호
• /
• pp.175-180
• /
• 2019

기존에 존재하는 패션 웹 사이트 에서는 상의, 하의 등의 품목에서는 한 가지 종류의 옷에 대한 검색결과만 보여주기 때문에 사용자가 원하는 옷에 대한 조합을 찾을 수 없다. 또 패션 시장이 성장함에 따라 소비자들은 다양한 패션 정보를 찾을 수 플랫폼을 요구하고 있다. 이러한 문제를 해결하고자 하여 딥러닝을 통한 이미지분류를 웹 사이트와 연동하고 SNS 기능을 접목하는 아이디어를 고안해냈다. 웹 사이트에 사용자가 본인의 이미지을 업로드하여 딥러닝 서버를 통해서 이미지의 특징을 파악하고 분류하여 저장한다. 사용자들은 저장된 정보를 가지고 여러 조합을 통해 원하는 이미지들을 검색할 수 있다. 또 SNS 기능을 통해 사용자간의 커뮤니케이션이 활발하게 이루어질 수 있다. 이를 통해서 기존에 존재하는 패션 관련 사이트의 문제를 해결하는 방안을 마련하였다.

• 한국정보통신학회논문지
• /
• 제15권10호
• /
• pp.2272-2278
• /
• 2011

원양어선을 보유한 회사에서 조업 중인 어선의 항해정보를 파악하기 위해 기존의 수작업에 의한 항해 현황판을 개선할 필요가 있다. 본 논문에서는 인터넷에 연결된 PC 또는 DID(Digital-Information Display)를 통해 선박 항해 위경도 표시 및 어획량 등 각종 항해 정보를 지도기반에서 visual하게 display하도록 설계한다. DB에 저장된 항해중인 37척의 원양어선 항해정보를 Flash 기술 기반의 RIA(rich Internet application)를 통해 역동적으로 화면구현이 가능한 Web Application 구축에 중점을 두었다. 다수의 어선상태를 한 화면에서 실시간으로 파악할 수 있기 때문에 항해정보를 모니터링하고, 의사결정을 신속히 가능하게 함으로 운영자 및 사용자에게 정보사용의 만족도를 높일 수 있었다.

• 한국지능시스템학회논문지
• /
• 제19권1호
• /
• pp.115-121
• /
• 2009

Web 2.0 has become the face next generation Web among the business world and research community. Web 2.0 is instant superficial gratification of people. On the other hand, Semantic Web is deep, meaningful and lasting relationship with data. So, it is difficult to apply the Semantic Web to the real world. In this paper, a platform for educational services using the Semantic Web and Web 2.0 is proposed. The proposed platform is based mix of the Semantic Web and Web 2.0, so it is useful to apply in the real world applications. Two services are presented, one is a semantic email system and the other is a cyber study space. The cyber study space adjusted each student is presented. The study environment is called iStudySpace that has personal scheduler, study status plan table, personalized search engine and several gadgets. Finally characteristics and limitations of the Semantic Web and Web 2.0, the organization and components of the platform, evaluation of iStudySpace are shown.

• Computers and Concrete
• /
• 제14권3호
• /
• pp.211-231
• /
• 2014

Prestressed hollow-core slabs (HCS) are widely used for modern lightweight precast floor structures because they are cost-efficient by reducing materials, and have excellent flexural strength and stiffness by using prestressing tendons, compared to reinforced concrete (RC) floor system. According to the recently revised ACI318-08, the web-shear capacity of HCS members exceeding 315 mm in depth without the minimum shear reinforcement should be reduced by half. It is, however, difficult to provide shear reinforcement in HCS members produced by the extrusion method due to their unique concrete casting methods, and thus, their shear design is significantly affected by the minimum shear reinforcement provision in ACI318-08. In this study, a large number of shear test data on HCS members has been collected and analyzed to examine their web-shear capacity with consideration on the minimum shear reinforcement requirement in ACI318-08. The analysis results indicates that the minimum shear reinforcement requirement for deep HCS members are too severe, and that the web-shear strength equation in ACI318-08 does not provide good estimation of shear strengths for HCS members. Thus, in this paper, a rational web-shear strength equation for HCS members was derived in a simple manner, which provides a consistent margin of safety on shear strength for the HCS members up to 500 mm deep. More shear test data would be required to apply the proposed shear strength equation for the HCS members over 500 mm in depth though.

• 문화기술의 융합
• /
• 제7권4호
• /
• pp.751-757
• /
• 2021

본 논문은 웹 서버 기반 환경에서 홀로그램 영상 제작을 위한 파이프라인 시스템을 제안하였다. 기존 홀로그램 영상 제작을 위해 시간 및 공간적인 제약이 존재한다. 제안하는 시스템을 통해 사용자에게 접근성을 높여 고품질의 홀로그램 영상을 획득하는 것을 목적으로 하였다. 웹 환경에서 사용자가 촬영한 동영상을 서버로 전송하여 후반 작업을 거쳐 홀로그램 영상 제작을 위한 프레임으로 변환하는 구조이다. 고품질 홀로그램 영상 획득을 위해 후반 작업은 딥러닝 기반의 알고리즘을 사용하였다. 제안하는 시스템은 사용자 편의를 위해 웹 환경에서 다양한 서비스 도구를 제공하였다. 이 방법을 통하여 제약된 공간이 아닌 웹 환경에서 영상을 촬영하기 때문에 홀로그램 영상 제작 시 사용자 접근성을 높였다.

• Structural Engineering and Mechanics
• /
• 제69권4호
• /
• pp.383-397
• /
• 2019

More than one longitudinal web stiffener may be economical in the design of plate girders that have considerably high width-to-thickness ratio of webs. In this study, the bend-buckling strength of relatively deep webs with two horizontal lines of flat plate-shaped single-sided stiffeners was numerically investigated. Linear eigenvalue buckling analyses were conducted for specially selected hypothetical models of stiffened web panels, in which top and bottom junctions of a web with flanges were assumed to have simply supported boundary conditions. Major parameters in the analyses were the locations of two longitudinal stiffeners, stress ratios in the web, slenderness ratios and aspect ratios of web panels. Based on the application of assumptions on the combined locations of the two longitudinal web stiffeners, simplified equations were proposed for the bend-buckling coefficients and compared to the case of one longitudinal stiffener. It was found that bend-buckling coefficients can be doubled by adopting two longitudinal stiffeners instead of one longitudinal stiffener. For practical design purposes, additional equations were proposed for the required bending rigidity of the longitudinal stiffeners arranged in two horizontal lines on a web.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 2000년도 가을 학술발표회논문집(I)
• /
• pp.577-582
• /
• 2000

This study is aimed to evaluate the shear strength of reinforced concrete deep beams subject to concentrated loads, using a simplified strut-tie model. For the shear strength prediction of deep beams, it is prerequisite to evaluate the effective width of strut and to verify the efficiency factors proposed by MacGregor et al.. The results analyzed by truss models have been compared with those calculated by domestic code for the existing data of 90 deep beam specimens. The shear strength of deep beams were reviewed with respect to concrete strength, the shear span-depth ratio, and the ratio of web reinforcements. The results showed that the shear strength of the proposed model gave a better agreement than the domestic code approach.

• Journal of Information Technology Applications and Management
• /
• 제26권6호
• /
• pp.89-101
• /
• 2019

Many researchers have been focused on designing beauty product recommendation system for a long time because of increased need of customers for personalized and customized recommendation in beauty product domain. In addition, as the application of the deep neural network technique becomes active recently, various collaborative filtering techniques based on the deep neural network have been introduced. In this context, this study proposes a deep neural network model suitable for beauty product recommendation by applying Neural Collaborative Filtering and Generalized Matrix Factorization (NCF + GMF) to beauty product recommendation. This study also provides an implementation of web API system to commercialize the proposed recommendation model. The overall performance of the NCF + GMF model was the best when the beauty product recommendation problem was defined as the estimation rating score problem and the binary classification problem. The NCF + GMF model showed also high performance in the top N recommendation.