http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
확신도 기법을 적용한 선박엔진 고장진단 전문가시스템 구현에 관한 연구
정경열(K.Y.Chung),최욱현(W.H.Choi),임헌호(H.H.Lim),류길수(K.S.Rhyu) 한국해양대학교 해사산업연구소 1998 海事産業硏究所論文集 Vol.8 No.-
This paper presents the design and implementation of a diagnosis system incorporating with a monitoring system. We classified the ship engine into the sub parts as similarly a experts classification method, acquired the causal relation between measuring data and fault states in consultation with experts, constructed a rule-base using them, and adopted a certainty-factor method in order to handle uncertanties of evidences and rules. When alarms occur, the system infers diagnostic conclusion by using certainty factor obtained from an alarm state (one of 4 levels) and an alarm datum's trend (one of 3 levels), and the certainty factor of a related rule (one of 3 levels).
선박 기관실 실시간 모니터링 시스템의 HMI 구현에 관한 연구
정경열(K.Y.Chung),최욱현(W.H.Choi),류길수(K.S.Rhyu) 한국해양대학교 해사산업연구소 1998 海事産業硏究所論文集 Vol.8 No.-
In this paper, the design and implementation of HMI for a real-time monitoring system of ship engine room is presented. The monitoring system gets sensor data of real-time from distributed local units after processing data, displays at CRT and saves in the database. The proposed system consists of four modules, the data management module, the alarm management module, the user interaction module, and the display module. The data management module transfers raw data from a device to engineering data. The alarm management module checks the alarm status of data received from local units. If the alarm occurs, data are displayed at CRT and saved at database.
하이브리드 지식표현을 이용한 선박엔진 고장진단시스템의 개발에 관한 연구
최욱현,소명옥,하창승,류길수 한국전문가시스템학회 1998 학술대회 Vol.2 No.1
This paper presents the design and implementation of ship engine diagnosis system using hybrid knowledge representation method and certainty-factor rule-base. The Monitoring system of ship engine room gets measuring data in real-time from distributed local units, processes, displays an CRT and saves them in the database. The alarm management function checks the alarm state of the monitoring data. If the alarm occurs, it displays on CRT and saves at database. This ship engine diagnosis system uses this monitoring data and alarm data to diagnosis data. The knowledge representation is used object-oriented knowledge representation method. The diagnosis data and abnormal states are represented each objects and abstracted class of diagnosis data and class of abnormal state. We use object-oriented concepts that are abject hierarchical structure, inheritance, method, pattern matching etc. then we can construct knowledge-base effectively. We use certainty-factor method for handling uncertainties of evidence and rule. The certainty-factor rule-base is developed using pattern matching and method that is attached at each abnormal state objects.