• 산업공학
• /
• 제11권2호
• /
• pp.159-176
• /
• 1998

In the motor manufacturing system with the properties of short delivery and order based production, the process plan is performed individually for each order by the expert of process plan after the completion of the detail design process to satisfy the specification to be required by customer. Also it is hard to establish the standard process plan in reality because part routings and operation times are varied for each order. Hence, the production planner has the problem that is hard to establish the production schedule releasing the job to the factory because there occurs the big difference between the real time to be completed the process plan and the time to be required by the production planner. In this paper, we study the decision supporting system for the process plan based on knowledge base concept. First, we represent the knowledge of process planner as a database model through the modified POI-Feature graph. Then we design and implement the decision supporting system imbedded in the heuristic algorithm in the client/server environment using the ORACLE relational database management system.

• 한국복합재료학회:학술대회논문집
• /
• 한국복합재료학회 2003년도 춘계학술발표대회 논문집
• /
• pp.139-142
• /
• 2003

As composite materials are gaining wide acceptance in aircraft structure, repair of damaged composite is becoming an important issue. The issues in composite repair include high cost, material interchangeability, water ingression, and structural integrity. To address these problems, researchers have studied on the composite repair in various aspects. In this paper, an Internet-based advisory service (called Repair Advisory Service, RAS) for composite repair is proposed to increase efficiency for repair process. In the RAS system the web browser is used as its user interface, which provides easy access to the service. The RAS server provides web-based tools for failure prediction, Structural Repair Manual (SRM), automated prepreg cutting process, material properties, inventory and knowledge base. The computer codes implemented for repair design estimate the tensile failure and shear failure of repaired structures. The prediction of failure is based on the maximum strain criterion for tensile failure while elastic-perfect plastic shear failure model is applied for interfacial failure. The OEM's SRM is provided in the PDF format for viewing and searching by web browsers instead of looking up paper version SRM. The knowledge base in this site offers a room to share and distribute ideas, memos, publications, or suggestions from the repair engineers. The fabrication tool of RAS reads repair geometry from engineers then generates a CNC toolpath to cut prepreg patches. The RAS service is open to public and available at http://nano.gsnu.ac.kr/. Broad feedback from field technicians and engineers is welcome to improve the usefulness of RAS.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제8권3호
• /
• pp.129-136
• /
• 2019

자연어 표현에는 인물, 조직, 장소, 제품 등의 다양한 개체들이 존재한다. 이러한 개체는 다양한 의미를 가질 수 있다. 이러한 개체가 갖는 중의성 문제는 자연어 처리 분야에 있어 매우 도전적인 과제이다. 개체 링킹(Entity Linking)이란 텍스트에 등장한 개체명을 지식베이스 내의 적절한 개체로 연결해주는 작업이다. 개체 링킹을 위한 대표적인 방법론인 짝 연결 접근법(Pairwise based method)은 한 문장에서 등장한 개체가 두 개 이상일 경우 서로의 연관성을 이용해 개체 링킹을 하는 방법이다. 이 방법은 동일 문장에서 등장하는 개체들 간의 상호의존성(interdependence)만을 고려하고 있어 포괄적인 상호의존성(Global interdependence)이 부족하다는 한계를 갖고 있다. 본 논문에서는 개체 링킹을 위해 RDF 형태의 지식베이스 정보를 바탕으로 Word2vec을 활용한 Entity2vec 모델을 생성하였다. 그리고 생성된 모델을 사용하여 각 개체에 대한 랭킹을 하였다. 본 논문에서는 짝 연결 접근법의 한계점을 보완하기 위해 포괄적인 상호의존성을 바탕으로 짝 연결 접근법을 고안하고 구현 및 실험을 통해 기존의 짝 연결 접근법과 비교하였다.

• 한국융합학회논문지
• /
• 제9권10호
• /
• pp.75-80
• /
• 2018

본 논문에서는 헬스 빅데이터 플랫폼에서 이기종 라이프로그 마이닝 모델을 제안한다. 이는 사용자의 라이프 로그를 실시간으로 수집하고 헬스케어 서비스를 제공하기 위한 온톨로지 기반의 마이닝 모델이다. 제안하는 방법은 이기종 라이프 로그 데이터를 분산처리하고, 클라우드 컴퓨팅 환경에서 실시간으로 처리한다. 이를 이기종 온톨로지를 기반으로 구성한 환경에 적합하도록 상위 온톨로지 방식으로 지식베이스를 재구성한다. 재구성한 지식베이스는 Jena 4.0 추론엔진을 이용해 추론 규칙들을 생성하고, 규칙 기반 추론 방법으로 실시간 헬스 서비스를 제공한다. 라이프로그 마이닝을 숨겨진 관계에 대한 분석과 시계열적 생체신호에 대한 예측모델을 구성한다. 이는 관계나 추론규칙에서 포함되지 않은 음의 상관관계나 양의 상관관계를 탐색하여 사용자의 생체신호에 대한 변화를 감지하고 예방 의료 서비스를 현실화하는 실시간 헬스케어 서비스가 가능하다. 성능 평가는 제안한 이기종 라이프로그 마이닝 모델 방법이 정확도에서 0.734, 재현율에서 0.752로 다른 모델에 비해 우수하게 나타난다.

• 한국시뮬레이션학회논문지
• /
• 제20권3호
• /
• pp.111-117
• /
• 2011

본 논문은 규칙기반 에이전트를 이용하여 여객선 화재시 피난행동특성을 고려한 승객행동모델을 제안한다. 기존 시스템들은 승객의 속도만을 반영한 속도기반모델을 사용하였다. 속도기반모델은 승객의 신체적 특징만 고려한 모델로 피난행동특성을 반영하기 어렵다. 이 문제점을 해결하기 위해 규칙기반 에이전트로 피난행동특성을 반영한 승객모델을 모델링하였다. 규칙기반 에이전트는 지식베이스와 추론엔진으로 구성된다. 지식베이스에는 피난행동특성들 중에서 다양한 행동 패턴을 보여줄 수 있는 예제를 선택하여 규칙형태의 지식으로 표현하고, 추론엔진은 지식을 전방추론하여 승객의 행위를 결정하도록 구현하였다. 이승객모델을 IMO MSC/Circ.1238의 문제 8에 적용하여 시뮬레이션을 진행한 결과 규칙기반 에이전트로 모델링한 승객이 다양한 인간의 행동 패턴을 표현할 수 있음을 검증하였다.

• 대한기계학회논문집A
• /
• 제30권3호
• /
• pp.219-230
• /
• 2006

Digital CAD models are one of the most important assets the manufacturer holds. The trend toward concurrent engineering and outsourcing in the distributed development and manufacturing environment has elevated the importance of high quality CAD model and its efficient exchange. But designers have spent a great deal of their time repairing CAD model errors. Most of those poor quality models may be due to designer errors caused by poor or incorrect CAD data generation practices. In this paper, we propose a rule-based approach for healing CAD model errors. The proposed approach focuses on the design history data representation from a commercial CAD model, and the procedural method for building knowledge base to heal CAD model. Through the use of rule-based approach, a CAD model healing system can be implemented, and experiments are carried out on automobile part models.

• International Journal of Safety
• /
• 제2권1호
• /
• pp.50-56
• /
• 2003

There are a lot of uncertainties in the severe accident phenomena and scenarios in nuclear power plants (NPPs) and one of the major issues for severe accident management is the reduction of these uncertainties. The severe accident management aid system using Probabilistic Safety Assessments (PSA) technology is developed for the management staff in order to reduce the uncertainties. The developed system includes the graphical display for plant and equipment status, previous research results by a knowledge-base technique, and the expected plant behavior using PSA. The plant model used in this paper is oriented to identify plant response and vulnerabilities via analyzing the quantified results, and to set up a framework for an accident management program based on these analysis results. Therefore the developed system may playa central role of information source for decision-making for severe accident management, and will be used as a training tool for severe accident management.

• 전기학회논문지P
• /
• 제54권2호
• /
• pp.67-72
• /
• 2005

This paper presents a method to select contingency ranking considering voltage security problems in power systems. Direct method which needs not the detailed knowledge of the post contingency voltage at each bus is used. Based on system operator's experience and knowledge, the membership functions for the MVAR mismatch and allowable voltage violation are justified describing linguistic representation with heuristic rules. Rule base is used for the computation of severity index for each contingency by fuzzy inference. Contingency ranking harmful to the system is formed by the index for security evaluation. Compared with 1P-1Q iteration, this algorithm using direct method and fuzzy inference shows higher computation speed and almost the same accuracy. The proposed method is applied to model system and KEPCO pratical system which consists of 311 buses and 609 lines to show its effectiveness.

• 한국자동차공학회논문집
• /
• 제9권4호
• /
• pp.166-175
• /
• 2001

The high speed tracking control of a 6-6 Stewart platform manipulator (SPM) normally requires knowledge of its complex full dynamics and measurement of its base motion when the SPM operates on a motion nit. In this study, an enhanced sliding mode control scheme has been developed, which is based on the reduced dynamics, not necessitating measurement of the base motion. The enhanced sliding mode control implemented with the perturbation compensation and modified reaching phase alleviation functions has been successfully employed for high speed tacking control of the laboratory SPM, when it is subjected to a virtual base motion.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2007년도 정기 학술대회 논문집
• /
• pp.731-736
• /
• 2007

Ubiquitous computing is a model of computing in which computer functions are integrated into everyday life, providing services and information in anywhere and anytime fashion. Augmented Reality is one of ubiquitous computing technologies to provide new paradigm utilized to interact between human and computer. By adding computer-generated non-visual information to real information and their interaction. user can get the improved and more knowledgeable information about real world. The purpose of this paper is the integration of AR and knowledge-base reasoning technology in ubiquitous computing. Through the introduced concept, it is enable to provide adequate knowledge in the process of ship design and manufacturing easily (Knowledge Everywhere). That is, this is a basic research to construct knowledge-based ubiquitous environment (KAD/KAM) in shipbuilding industry.