• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1849-1850
• /
• 2008

Type-2 퍼지 집합은 Type-1 퍼지 집합의 확장으로 Type-1 퍼지 집합으로는 다루기 힘든 언어적인 불확실성을 다루기 위해 고안되었다. 대표적인 퍼지 논리 시스템(Fuzzy Logic System; FLS)으론 Mamdani FLS 모델과 TSK FLS모델이 있다. 본 논문에서는 Interval Type-2 TSK FLS를 구성한다. FLS 구성을 위한 전반부는 가우시안 형태의 Type-2 멤버쉽 함수를 사용하며, 전.후반부 파라미터들은 오류역전파 알고리즘을 통한 학습으로 결정한다. 본 논문에서는 Type-1 TSK FLS와 Interval Type-2 TSK FLS를 설계하고 가스로 공정 데이터에 적용하여 성능을 비교 분석한다. 또한 노이즈를 추가한 데이터들을 통하여 노이즈에 대한 성능도 비교 분석한다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.1
• /
• pp.173-181
• /
• 2017

As the industries have developed, a myriad of big data have been produced and the inherent uncertainty in the data has also increased accordingly. In this paper, we propose an interval type-2 fuzzy clustering method to deal with the inherent uncertainty in the data and, using this method, design and optimize the fuzzy neural network. Fuzzy rules using the proposed clustering method are designed and carried out the learning process. Genetic algorithms are used as an optimization method and the model parameters are optimally explored. Experiments were performed with two pattern classification, both of the experiments show the superior pattern recognition results. The proposed network will be able to provide a way to deal with the uncertainty increasing.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.3
• /
• pp.73-82
• /
• 1995

The paper demonstrates that an adaptive fuzzy controller can be used effectively for the control of the temperature and density of the Tokarnak fusion recator which is nonlinear and has dynamic uncertainties. The dynamic uncertainties are non-parametric but state dependent. Thus the conventional adaptive nonlinear control methods have difficulties to cope with the problem. The proposed adaptive fuzzy controller can be used as a solution and performs well in a predetermined local space. Simulation result verifies the effectiveness of the scheme.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1851-1852
• /
• 2008

Type-2 퍼지 논리 집합은 언어적인 불확실성을 다루기 위하여 고안된 Type-1 퍼지 논리 집합의 확장한 것이다. Type-2 퍼지 논리 시스템은 외부 노이즈를 효율적으로 다룰 수 있다. 본 논문에서는 불확실성을 표현하기 위해서 전.후반부 멤버쉽 함수로 삼각형 형태의 Type-2 퍼지 집합을 사용한다. 전반부 멤버쉽 함수의 정점을 결정하는데 유전자 알고리즘(Genetic Algorithms)으로 멤버쉽 함수의 정점을 결정한다. 제안된 모델은 모델 평가에 주로 사용되는 가스로 시계열 데이터를 적용하고, 테스트 데이터로 노이즈에 영향 받은 데이터를 사용하여 수치적인 예를 보인다.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.22 no.2
• /
• pp.131-137
• /
• 2018

This paper analyzes the probability of failure for the equivalence ratio error. The control error of the equivalence ratio is affected by the aleatory and epistemic uncertainties. In general, reliability analysis techniques are easily incorporated to handle the aleatory uncertainty. However, the epistemic uncertainty requires a new approach, as it does not provide an uncertainty distribution. The Bayesian inference incorporates the reliability analysis results to handle both uncertainties. The result gives a distribution of failure probability, whose equivalence ratio does not meet the requirement. This technique can be useful in the analysis of most engineering systems, where the aleatory and epistemic uncertainties exist simultaneously.

• Journal of The Korean Association For Science Education
• /
• v.44 no.5
• /
• pp.405-420
• /
• 2024

This study explored the scientific uncertainties raised by students during small-group scientific argumentation and how the uncertainties contributed to the argumentation. A total of 37 seventh-grade middle school students and a teacher participated in the study. They engaged in small-group argumentation on the topic of photosynthesis. We selected three small focal groups, each consisting of 4-5 students, that actively participated in argumentation and raised uncertainties. We conducted small-group interviews with these three focal groups and the teacher. All lesson and interview videos, audio transcripts, student worksheets, and the researcher's field notes were collected and analyzed qualitatively. The findings revealed that there were three major types of uncertainties that contributed to the small-group argumentation. The first type of uncertainties-those about scientific content knowledge-prompted conceptual support from high-achieving peers or the teacher, facilitating the justification of arguments. The second type-uncertainties about data-encouraged students to consider alternative perspectives and arguments. This led students to raise rebuttals and try to reach a consensus, considering the alternatives. Finally, the third type-uncertainties about how to construct scientific arguments-was raised in one small group and prompted epistemic support from the leader, who was more proficient in argumentation. The leader encouraged other students to present their own evidence, rather than just following her opinions. This study provides useful insights for research on scientific uncertainties that students raise in epistemic practices and for developing instructional strategies to support the management of these uncertainties.

• Journal of the Korean Society for Railway
• /
• v.12 no.1
• /
• pp.135-143
• /
• 2009

The majority of development projects fail to meet their target schedule and cost, with the overrun typically between 40 and 200 percent. These overruns happen because the planners underestimate the work or do not consider the need to rework at project planning. Representative schedule planning/management techniques such as Gantt Chart, PERT/CPM etc. that are used in domestic project planning are unable to reflect rework. This paper proposes a method to consider rework to provide more realistic estimates at schedule planning. Additionally, to prevent the underestimation of the work this paper proposes a simulation method that calculates a probabilistic estimated schedule and the associated variance based on the random variable modeling of individual task completion dates.

• Journal of the Korean Geotechnical Society
• /
• v.23 no.11
• /
• pp.109-117
• /
• 2007

Uncertainty is pervasive in rock slope stability analysis due to various reasons and subsequently it may cause serious rock slope failures. Therefore, the importance of uncertainty has been recognized and subsequently the probability theory has been used to quantify the uncertainty since 1980's. However, some uncertainties, due to incomplete information, cannot be handled satisfactorily in the probability theory and the fuzzy set theory is more appropriate for those uncertainties. In this study the random variable is considered as fuzzy number and the fuzzy set theory is employed in rock slope stability analysis. However, the previous fuzzy analysis employed the approximate method, which is first order second moment method and point estimate method. Since previous studies used only the representative values from membership function to evaluate the stability of rock slope, the approximated analysis results have been obtained in previous studies. Therefore, the Monte Carlo simulation technique is utilized to evaluate the probability of failure for rock slope in the current study. This overcomes the shortcomings of previous studies, which are employed vertex method. With Monte Carlo simulation technique, more complete analysis results can be secured in the proposed method. The proposed method has been applied to the practical example. According to the analysis results, the probabilities of failure obtained from the fuzzy Monte Carlo simulation coincide with the probabilities of failure from the probabilistic analysis.

• ICROS
• /
• v.10 no.3
• /
• pp.21-26
• /
• 2004

Structured singular value (SSV)는 robust stability와 robust performance를 매우 엄밀하게 다루기 위해 고안되었다 (Doyle, 1982; Safonov, 1982). 이 엄밀성으로 제어시스템의 설계 및 분석에 광범위하게 사용되고 있다. 강건제어의 단초를 이루었으며 loop failure tolerance, decentralized integral controllability (Campo and Morari. 1994), D-stability (Lee and Edgar, 2001) 등에 SSY가 사용되고 있다. SSV의 중요성이 알려짐에 따라 이것에 관한 많은 연구가 있었다(Fan et at., 1991 ; Pacltard and Pander, 1993), 그러나 이 값의 계산은 매우 어려운 NP-hard인 것으로 판명되었으며 (Braatz et al.. 1994). 실수 불확실 변수에 대한 SSV의 경우 원하는 오차범위 내로 근사 값을 구하는 것도 마찬가지 인 것으로 밝혀졌다(Fu, 1997).(중략)

• Proceedings of the KIEE Conference
• /
• 2008.04a
• /
• pp.57-58
• /
• 2008

Type-2 퍼지 집합은 언어의 불확실성을 다루기 위하여 고안된 Type-1 퍼지집합의 확장이다. TSK 퍼지 로직 시스템(TSK Fuzzy Logic Systems; TSK FLS)은 Mamdani FLS과 함께 가장 널리 사용되는 퍼지 로직 시스템 모델이다. 본 논문에서는 Type-2 퍼지 집합을 이용하여 전반부 멤버쉽 함수를 구성하고 후반부 다항식 함수를 상수와 1차식, 2차식으로 확장한 다항식 Type-2 TSK FLS 설계한다. 다항식 Type-2 TSK FLS의 파라미터를 동정하기 위해 Back-propagation 방법을 사용한다. 제안된 다항식 Type-2 TSK FLS을 노이즈 섞인 비선형 시스템의 모델링에 적용하여 그 성능을 비교 분석한다.