• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.129-132
• /
• 1996

In this paper, we suggest a framework to represent the fuzziness in knowledge base as a perspective of the object-oriented paradigm. We divide the knowledge base in two parts. One is the object-base that stores the fuzzy propositions and the explanatory databases. The other is the rule-base that manages the rules between the fuzzy propositions. As the first step, we have to develop a new fuzzy object model that gives an easy way to represent the fuzzy propositions, that is, the fuzzy knowledge in the real world.

• 전자공학회논문지CI
• /
• 제42권6호
• /
• pp.55-62
• /
• 2005

퍼지논리제어기가 이상적인 제어효과를 나타내게 할려면 적합한 규칙집합을 사용하는 것이 아주 중요하다. 퍼지논리제어기의 언어구조는 가상언어정책을 초기 규칙기반으로 사용하는 것을 허용한다. 만약 설계단계에서 적당한 규칙들을 일정하게 잘 조합시킨다면 제어기의 성능을 훨씬 더 향상시킬 수 있을 것이다. 본 논문에서 퍼지제어기 성능을 개선하기 위한 규칙기반 표에서의 원소추이방법을 제안하였다. 제안된 방법은 에러가 증가되면 시스템을 조절하는 출력의 제어효과가 증대될 것이고 반대로 에러가 감소되면 그에 따른 출력의 제어효과가 감소할 것이라는 원리를 기반으로 하였다. 모의실험결과에 의해 제안된 방법은 퍼지제어 규칙기반과 퍼지논리제어기의 성능을 향상시키기 위한 아주 효과적인 방법임을 알 수 있다.

• 한국멀티미디어학회논문지
• /
• 제3권2호
• /
• pp.201-208
• /
• 2000

이 논문에서는 시계열 예측을 위하여 퍼지 엔트로피에 의한 입력공간의 분할과 퍼지 제어규칙을 자동으로 생성하는 방법을 제안하고, Mackey-Glass 데이터 Set을 이용한 시계열 예측 문제에 적용하여 그 성능을 검증한다. 이 방법은 샤논 함수와 퍼지 엔트로피 함수를 이용하여 입력공간을 분할하고, 분할된 부 공간에 대해 이력 데이터와 부합할 수 있는 각각의 규칙에 등급을 정하여 불필요한 제어규칙을 제거하여 최적의 규칙베이스를 구성하도록 한다. 적용되는 퍼지 신경망의 기본적인 구조는 퍼지 제어기의 규칙베이스와 추론의 과정을 신경회로망을 이용하여 구현하며 퍼지 제어규칙의 매개변수들은 최대 급경사 강하법에 의해 적응되어진다. 제안되는 알고리즘을 매개변수의 수를 줄이기 위하여 제어 규칙의 결론부의 출력값은 신경망의 가중치로 구성하여 퍼지 신경망의 복잡도를 줄임으로서 추론형과 기술형 접근법을 혼합한 형태의 학습 알고리즘이다.

• 산업경영시스템학회지
• /
• 제21권48호
• /
• pp.213-222
• /
• 1998

The objective of this study is to develop a knowledge-based scheduler applying simulation and knowledge base. This study utilizes a machine induction to build knowledge base which enables knowledge acquisition without domain expert. In this study, the best job dispatching rule for each order is selected according to the specifications of the order information. And these results are built to the fact base and knowledge base using the attribute-oriented induction method and simulation. When a new order enters in the developed system, the scheduler retrieves the knowledge base in order to find a matching record. If there is a matching record, the scheduling will be carried out by using the job dispatching rule saved in the knowledge base. Otherwise the best rule will be added to the knowledge base as a new record after scheduling to all the rules. When all these above steps finished the system will furnish a learning function.

• 한국해양공학회지
• /
• 제21권2호
• /
• pp.81-86
• /
• 2007

This paper presents a computer aided process planning system integrating case-based reasoning and expert system for block assembly in shipbuilding. Expert rules are extracted from the case-base where cases are represented as a set of constraint-satisfaction problems. Rules for the expert system are extracted by generalizing the constraints. In generalizing the constraints, parts are generalized as variables or as part-types. The system was developed with CLIPS, an expert system shell. As more cases are collected, more rules will be extracted and the existing rules will be updated.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2000년도 춘계 학술대회 및 국제 감성공학 심포지움 논문집 Proceeding of the 2000 Spring Conference of KOSES and International Sensibility Ergonomics Symposium
• /
• pp.253-257
• /
• 2000

the objective of this study is to design and evaluate a methodology for diagnosing the appendicitis in a fuzzy neural network that integrates the partition of input space by fuzzy entropy and the generation of fuzzy control rules and learning algorithm. In particular the diagnosis of appendicitis depends on the rule of thumb of the experts such that it associates with the region, the characteristics, the degree of the ache and the potential symptoms. In this scheme the basic idea is to realize the fuzzy rle base and the process of reasoning by neural network and to make the corresponding parameters of the fuzzy control rules be adapted by back propagation learning rule. To eliminate the number of the parameters of the rules, the output of the consequences of the control rules is expressed by the network's connection weights. As a result we obtain a method for reducing the system's complexities. Through computer simulations the effectiveness of the proposed strategy is verified for the diagnosis of appendicitis.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.983-987
• /
• 2001

This paper deals with an automated computer-aided process planning system by which designer can determine operation sequences even if they have little experience in process planning of wheel bolt products by a multi-stage former. The approach to the system is based on knowledge-based rules and a process knowledge base consisting of design rules is built. Knowledge for the system is formulated from plasticity theories, empirical results and the empirical knowledge of field experts. Programs for the system have been written in AutoLISP for the AutoCAD using a personal computer and are composed of two main modules. An attempt is made to link programs incorporationg a number of expert design rules to form a useful package. Results obtained using the modules enable the designer and manufacturer of wheel bolt product to be more efficient in this field.

• The Journal of the Acoustical Society of Korea
• /
• 제16권1E호
• /
• pp.64-74
• /
• 1997

In many decision making systems, rule-based approaches are used to solve complex problems in the areas of pattern analysis and computer vision. In this paper, we present methods for generating fuzzy IF-THEN rules automatically from training data for pattern classification and high-level computer vision. The rules are generated by construction minimal approximate fuzzy aggregation networks and then training the networks using gradient descent methods. The training data that represent features are treated as linguistic variables that appear in the antecedent clauses of the rules. Methods to generate the corresponding linguistic labels(values) and their membership functions are presented. In addition, an inference procedure is employed to deduce conclusions from information presented to our rule-base. Two experimental results involving synthetic and real are given.

• 한국철도학회논문집
• /
• 제8권5호
• /
• pp.411-418
• /
• 2005

This paper expounds the development of a user-friendly expert system for the optimal stacking sequence design of composite laminates subjected to the various rules constraints. The expert system was realized in the graphic-based design environment. Therefore, users can access and use the system easily. The optimal stacking sequence is obtained by means of integration of a genetic algorithm, finite element analysis. These systems were integrated with the rules of design heuristics under an expert system shell. The optimal stacking sequence combination for the application of interest is drawn from the discrete ply angles and design rules stored in the knowledge base of the expert system. For the integration and management of softwares, a graphic-based design environment that provides multi-tasking and graphic user interface capability is built.

• 지능정보연구
• /
• 제1권2호
• /
• pp.73-83
• /
• 1995

본 논문 에서는 진화 알고리즘을 이용하여 퍼지 규칙 베이스의 최적 규칙들을 자동으로 생성하는 방법을 제안한다. 진화 알고리즘에 의한 퍼지 논리 시스템의 최적 규칙은 전문가의 사전 경험이나 지식이 없이도 자동 설계가 가능하고 이들 규칙을 이용하여 지능 제어를 할 수 있다. 본 논문에서 사용한 접근 방법은 퍼지 규칙 소속함수의 자동 조정으로 규칙을 생성하고, 최적의 제어 규칙 탐색은 퍼지 논리 시스템의 성능 기준으로 정의한 적합도 값을 기반으로 탐색한다. 제안한 방법의 유용성을 보이기 위해 비선형 시스템에서 컴퓨터 모의실험을 행하였다.