• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.6
• /
• pp.451-464
• /
• 2022

The high-level nuclear waste repository is a deep geological disposal system exposed to complex environmental conditions such as high temperature, radiation, and ground-water due to handling spent nuclear fuel. Continuous exposure can lead to cracking and deterioration of the structure over time. On the other hand, the high-level nuclear waste repository requires an ultra-long life expectancy. Thus long-term structural health monitoring is essential. Various sensors such as an accelerometer, earth pressure gauge, and displacement meter can be used to monitor the health of a structure, and a piezoelectric sensor is generally used. Therefore, it is necessary to develop a highly durable sensor based on the durability assessment of the piezoelectric sensor. This study designed an accelerated life test for durability assessment and life prediction of the piezoelectric sensor. Based on the literature review, the number of accelerated stress levels for a single stress factor, and the number of samples for each level were selected. The failure mode and mechanism of the piezoelectric sensor that can occur in the environmental conditions of the high-level waste repository were analyzed. In addition, two methods were proposed to investigate the maximum harsh condition for the temperature stress factor. The reliable operating limit of the piezoelectric sensor was derived, and a reasonable accelerated stress level was set for the accelerated life test. The suggested methods contain economical and practical ideas and can be widely used in designing accelerated life tests of piezoelectric sensors.

• Journal of the Korean Society for Library and Information Science
• /
• v.58 no.2
• /
• pp.151-175
• /
• 2024

This study analyzes the current status of research data management at NST-affiliated government-funded research institutes for the purpose of promoting the sharing and use of research data, and based on this, suggests methods for establishing a research data sharing and management system. The survey on the status of research data management was conducted twice in 2022 and 2023 for a total of 20 research institutes. In addition, difficulties and areas that need to be improved in the management and sharing of research data were identified, and based on this, methods for establishing a research data sharing and management system were proposed by dividing them into policy aspects, system aspects, and linkage system construction aspects. In order to establish a research data sharing system, it would be desirable to prepare a policy basis and present contents such as the definition of research data, scope of application, contents of management, utilization method, and leading institutes. In addition, for systematic and unified research data management, it would be recommended that each institute will establish and manage a repository and management system. By linking this with DataON, the national research data platform, and providing one-stop services, the accessibility and usability of data will be improved.

• The Journal of the Korea Contents Association
• /
• v.4 no.4
• /
• pp.1-10
• /
• 2004

Dialogue speech corpora that contain sufficient dialogue speech features are needed for performance assessment of a spoken language dialogue system. And labeling information of dialogue speech corpora plays an important role for improvement of recognition rate in acoustic and language models. In this paper, we examine the methods by which labeling information of dialogue speech corpora can be structured. More specifically, we examined how to represent features of dialogue speech in a structured document based XML and how to design the repository system of the information.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.33 no.3
• /
• pp.87-92
• /
• 2010

In this paper, we suggest a PLCS-based architecture and operation method for providing interoperability in SBA integrated collaborative environment. Specifically, the suggested architecture is based on the distributed collaborative environment which employes the PLCS application protocol and integrated repository for representing and sharing product data information between collaborators remotely located. As an example of data representation, a military vehicle system is considered and two kinds of information, including state and activity/process, are explained. We expect that the suggested architecture can be used as a reference model to develop an efficient SBA integrated collaborative environment.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2002.11a
• /
• pp.370-376
• /
• 2002

인터넷 환경에서 발생되는 수많은 데이터를 지능적으로 처리할 수 있는 자동화된 분석 시스템의 필요성이 제기된다. 이러한 시스템의 데이터 분석은 크게 지도 학습과 자율 학습으로 나된다. 본 논문에서는 특히 자율학습 군집화에 대한 자동화된 시스템으로서 지능화된 데이터 마이닝 에이전트를 제안한다. 군집화 과정에서는 데이터를 분석하는 분석가가 군집화의 방법과 결과 해석에 실시간으로 관여하기 어렵기 때문에 이러한 작업을 담당하는 지능화된 에이전트가 자동화된 군집화를 담당하면 효과적인 군집화 전략이 될 수 있다. 본 논문의 자동 군집화를 위한 지능화된 데이터 마이닝 에이전트 시스템은 군집화 수행 에이전트와 군집화 성능 평가 에이전트로 구성된 다중 에이전트로서 두 개의 에이전트가 서로 정보를 교환하면서 최적의 군집화를 수행한다. UCI Machine Repository 데이터를 이용한 실험을 통해 제안 시스템의 성능 평가를 수행하였다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2002.11a
• /
• pp.311-320
• /
• 2002

시맨틱 웹은 인간이 이해하는 것처럼 웹 문서의 의미를 컴퓨터가 처리할 수 있도록 하는데 있다. 그러나 인터넷 등 정보통신 기술의 발전으로 인해 정보량이 급증함으로써 이들 정보 자원을 효과적으로 검색하기에는 많은 어려움이 있다. 이러한 문제점을 해결하기 위해 본 논문에서는 주석 에디터를 사용하여 논문에 대한 RDF 메타데이타의 자동 생성 방법을 제안한다. 사용자가 논문을 주석 처리할 때, 문서에 대한 특징을 추출하고 온토로지 인터페이스를 사용하여 문서를 분류한다. 구현된 시스템을 통해 사용자는 추출된 메타데이타를 메타데이타 뷰를 통해 볼 수 있으며, HTML 뷰를 통해 메타데이타를 수동으로 수정이 가능하다. 이 메타데이타는 RDF Repository로 저장할 수 있으며, 주석 뷰를 통하여 RDF 메타데이타 생성을 확인할 수 있다. 이렇게 생성된 RDF 메타데이타는 웹 로봇이 내용의 의미 파악 및 카테고리 정보를 쉽게 알 수 있도록 해준다. 본 논문은 검색 엔진을 통하여 논문 검색시 전체 내용보다 RDF 메타데이타 정보만으로 효율적인 검색을 할 수 있는 방법에 초점을 둔다.

• Korean Journal of Computational Design and Engineering
• /
• v.5 no.3
• /
• pp.232-241
• /
• 2000

Even though we have greater chances to accomplish successful collaborative design by utilizing recent proliferation of networks, current practices do not fully take advantage of the information infrastructure. There are so much data over the networks, but not enough knowledge about the data is available to users. The main objectives of the product data interface system proposed in this paper are to capture more knowledge on managing product data and to provide users effective search capability. We define the metadata model for product data defined in STEP AP 203 and manage the metadata from product data in a repository system. Because we utilize the standard formats such as STEP for product data and RDF for metadata, the proposed approach can be applied to various fields of industries independently on commercial products.

• Korean Journal of Computational Design and Engineering
• /
• v.20 no.4
• /
• pp.401-411
• /
• 2015

Cybrid virtual plant concept is defined as a cyber plant mimicking a physical plant by using plant engineering data and sensor data coming from sensors attached to facilities of the physical plant. Cybrid virtual plant is a new concept for plant industry so that plant managers and operators' requirements need to be captured for systematic application of the concept to the plant industry. The paper proposed an architecture of the cybrid virtual plant, and provided requirement analysis results for a specific plant company. A database, named smart-cube repository, for the proposed cybrid virtual plant is also proposed and its conceptual data structure is described.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.5
• /
• pp.1377-1387
• /
• 2000

Many companies are investing in the development of the PDM(Prodct Data Management) system to improve the productivities of manufacturing since people believe that PDM technology can give a new solution from planning to development. As the requirements of PDM grows, so many industries sped their own budget on developing common requirements and functionalities. In this paper, we describe the framework for a PDM application to promote reuse in a PDM system development area. However, developing the framework is not easy. In this paper, the current state and results of our development are described: 1) the phases of developing our framework, 2) abstruction strategies, 3) programing model based on repository. Although frameworks improve software reuse, frameworks are so large and complex that application developers require to understand framework in order to customize the framework. To support the customization of framework, development environment is developed, also.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.05a
• /
• pp.87-93
• /
• 2005

다양한 정보의 교환 및 공유가 이루어지는 물류분야에서는 정보의 통합관리가 필요하다. 본 논문에서는 서로 다른 물류주체 간의 정보 교환 및 공유가 원활히 이루어지는 RFID에 기반을 둔 물류환경에 적합한 물류정보 통합관리 모형을 제시한다. 본 논문에서 제시하는 물류정보 통합관리 모형은 웹 서비스를 기반으로 물류정보를 저장/관리하는 시스템 모형으로 애플리케이션과의 데이터교환을 위한 인터페이스, 물류데이터 저장, 검색을 위한 모듈, 물류정보의 데이터 무결정과 의미추론을 위한 온톨로지 모듈, XML Repository 및 Respository access 모듈로 구성된다. 데이터 교환을 위한 인터페이스는 SOAP을 이용하며 애플리케이션 및 RFID 미들웨어와 물류정보를 주고 받는다. 데이터 처리모듈은 물류 데이터의 유형에 따라 데이터를 저장/관리하며, 애플리케이션의 정보요청에 대한 결과를 제공한다. 또한 물류정보의 데이터 무결성 검증, 의미추론을 위한 온톨로지 모듈을 포함한다. 본 논문은 XML형식의 물류정보 저장관리 및 시맨틱웹 개념 중심의 모형을 설계한다.