• 대한전기학회논문지:전력기술부문A
• /
• 제48권11호
• /
• pp.1382-1387
• /
• 1999

This paper presents an optimal fuzzy power system stabilizer to damp out low frequency oscillation. So far fuzzy controllers have been applied to power system stabilizing controllers due to its excellent properties on the nonlinear systems. But the design process of fuzzy logic power system stabilizer requires empirical and heuristic knowledge of human experts as well as many trial-and-errors in general. This paper presents and optimal design method of the fuzzy logic stabilizer using the genetic algorithm. Non-symmetric membership functions are optimally tuned over an evaluation function. The present inputs of fuzzy stabilizer are torque angle error and the change of torque angle error without loss of generality. The coding method used in this paper is concatenated binary mapping. Each linguistic fuzzy variable, defined as the peak of a membership function, is assigned by the mapping from a minimum value to a maximum value using eight bits. The tournament selection and the elitism are used to keep the worthy individuals in the next generation. The proposed system is applied to the one-machine infinite-bus model of a power system, and the results showed a promising possibility.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1626-1629
• /
• 2002

This paper describes novel methods to construct fuzzy inference rules with gradient descent. The present methods have a constructive mechanism of the rule unit that is applicable in two parameters: the central value and the width of the membership function in the antecedent part. The first approach is to create the rule unit at the nearest position from the input space, for the central value of the membership function in the antecedent part. The second is to create the rule unit which has the minimum width, for the width of the membership function in the antecedent part. Experimental results are presented in order to show that the proposed methods are effective in difference on the inference error and the number of learning iterations.

• 한국컴퓨터정보학회논문지
• /
• 제16권6호
• /
• pp.61-70
• /
• 2011

본 논문은 퍼지를 적용한 순환 신경망을 이용하여 잡음 제거용 필터를 구현하였다. 제안된 퍼지 순환 신경망 구조는 기본적으로 순환 신경망 구조를 이용하여 가중치 및 반복횟수가 일정한 값에 수렴하도록 하였으며, 하이브리드 퍼지 소속 함수 연산자를 적용하여 수학적인 계산량 및 복잡성를 단순화하였다. 본 논문은 제안된 퍼지 순환 신경망 구조 필터가 일반적인 순환 신경망 구조 필터보다 평균 0.38dB 정도 영상복원이 개선됨을 PSNR을 이용하여 증명하였다. 또한 결과 영상 비교에서 제안된 방법을 적용하여 얻은 영상이 기존 방법을 적용하여 얻은 영상보다 원영상과 더 유사함을 확인하였다.

• 한국안전학회지
• /
• 제36권2호
• /
• pp.49-57
• /
• 2021

In the case of military supplies, any potential failure and causes of failures must be considered. This study is aimed at examining the failure modes of a rotorcraft landing system to identify the priority items. Failure mode and effects analysis (FMEA) is applied to the rotorcraft landing system. In general, the FMEA is used to evaluate the reliability in engineering fields. Three elements, specifically, the severity, occurrence, and detectability are used to evaluate the failure modes. The risk priority number (RPN) can be obtained by multiplying the scores or the risk levels pertaining to severity, occurrence, and detectability. In this study, different weights of the three elements are considered for the RPN assessment to implement the FMEA. Furthermore, the FMEA is implemented using a fuzzy rule base, similarity aggregation model (SAM), and grey theory model (GTM) to perform a comparative analysis. The same input data are used for all models to enable a fair comparison. The FMEA is applied to military supplies by considering methodological issues. In general, the fuzzy theory is based on a hypothesis regarding the likelihood of the conversion of the crisp value to the fuzzy input. Fuzzy FMEA is the basic method to obtain the fuzzy RPN. The three elements of the FMEA are used as five linguistic terms. The membership functions as triangular fuzzy sets are the simplest models defined by the three elements. In addition, a fuzzy set is described using a membership function mapping the elements to the intervals 0 and 1. The fuzzy rule base is designed to identify the failure modes according to the expert knowledge. The IF-THEN criterion of the fuzzy rule base is formulated to convert a fuzzy input into a fuzzy output. The total number of rules is 125 in the fuzzy rule base. The SAM expresses the judgment corresponding to the individual experiences of the experts performing FMEA as weights. Implementing the SAM is of significance when operating fuzzy sets regarding the expert opinion and can confirm the concurrence of expert opinion. The GTM can perform defuzzification to obtain a crisp value from a fuzzy membership function and determine the priorities by considering the degree of relation and the form of a matrix and weights for the severity, occurrence, and detectability. The proposed models prioritize the failure modes of the rotorcraft landing system. The conventional FMEA and fuzzy rule base can set the same priorities. SAM and GTM can set different priorities with objectivity through weight setting.

• 한국지능시스템학회논문지
• /
• 제8권6호
• /
• pp.78-84
• /
• 1998

본 논문에서는 유전 알고리즘을 사용하는 퍼지 교통 제어기를 제안한다. 일반적인 퍼지 교통 제어기들은 사람에 의해 생성된 소속함수들을 사용한다. 그러나 이 방식은 퍼지 제어기를 설계하는데 최적의 해를 보장하지 못한다. 유전 알고리즘은 휴리스틱적인 특정 영역의 지식을 필요로 하는 최적화 문제의 좋은 해결 방법이다. 좋은 성능을 보이는 퍼지 소속함수를 찾기 위해서 적합도 함수가 정의되어야 한다. 그러나 교통 제어에서 적합도 함수를 수치 표현으로 정의하는 것은 쉽지 않다. 따라서 본 논문에서는 교통 시뮬레이터에 의해 얻어지는 성능척도로써 해의 적합도를 결정하는 시뮬레이션 접근법을 사용한다. 제안된 방법은 기존의 퍼지 제어기들에 비하여 우수한 성능을 보여준다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.999-1004
• /
• 2005

An algorithm for a hybrid controller consists of a sliding mode control part and a fuzzy logic part which ar purposely for nonlinear systems. The sliding mode part of the solution is based on "eigenvalue/vector"-type controller is used as the backstepping approach for tracking errors. The fuzzy logic part is a Mamdani fuzzy model. This is designed by applying sliding mode control (SMC) method to the dynamic model. The main objective is to keep the update dynamics in a stable region by used SMC. After that the plant behavior is presented to train procedure of adaptive neuro-fuzzy inference systems (ANFIS). ANFIS architecture is determined and the relevant formulation for the approach is given. Using the error (e) and rate of error (de), occur due to the difference between the desired output value (yd) and the actual output value (y) of the system. A dynamic adaptation law is proposed and proved the particularly chosen form of the adaptation strategy. Subsequently VSC creates a sliding mode in the plant behavior while the parameters of the controller are also in a sliding mode (stable trainer). This study considers the ANFIS structure with first order Sugeno model containing nine rules. Bell shaped membership functions with product inference rule are used at the fuzzification level. Finally the Mamdani fuzzy logic which is depends on adaptive neuro-fuzzy inference systems structure designed. At the transferable stage from ANFIS to Mamdani fuzzy model is adjusted for the membership function of the input value (e, de) and the actual output value (y) of the system could be changed to trapezoidal and triangular functions through tuning the parameters of the membership functions and rules base. These help adjust the contributions of both fuzzy control and variable structure control to the entire control value. The application example, control of a mass-damper system is considered. The simulation has been done using MATLAB. Three cases of the controller will be considered: for backstepping sliding-mode controller, for hybrid controller, and for adaptive backstepping sliding-mode controller. A numerical example is simulated to verify the performances of the proposed control strategy, and the simulation results show that the controller designed is more effective than the adaptive backstepping sliding mode controller.

• Genomics & Informatics
• /
• 제16권4호
• /
• pp.37.1-37.3
• /
• 2018

Gene-gene interaction is a key factor for explaining missing heritability. Many methods have been proposed to identify gene-gene interactions. Multifactor dimensionality reduction (MDR) is a well-known method for the detection of gene-gene interactions by reduction from genotypes of single-nucleotide polymorphism combinations to a binary variable with a value of high risk or low risk. This method has been widely expanded to own a specific objective. Among those expansions, fuzzy-MDR uses the fuzzy set theory for the membership of high risk or low risk and increases the detection rates of gene-gene interactions. Fuzzy-MDR is expanded by a maximum likelihood estimator as a new membership function in empirical fuzzy MDR (EFMDR). However, EFMDR is relatively slow, because it is implemented by R script language. Therefore, in this study, we implemented EFMDR using RCPP ($c^{{+}{+}}$ package) for faster executions. Our implementation for faster EFMDR, called EMMDR-Fast, is about 800 times faster than EFMDR written by R script only.

• 한국정보처리학회논문지
• /
• 제4권1호
• /
• pp.149-160
• /
• 1997

본 연구에서는 퍼지사상화(fuzzy mapping)와 FLVQ(fuzzy learning vector quantization)에 의한 사상된(mapped)코드북을 사용하는 화자적용 음성합성 알고리즘 을 제안하고, 기존의 음성합성결과와 비교한다. 입력화자와 기준화자의 코드북은 FLVQ 방법으로 작성한다. 사상된 코드북은 퍼지 히스토그램을 작성하여 이들을 선형 결합함으로써 얻어지는 퍼지 사상화에 의하여 작성된다. 대응 코드벡터의 퍼지 히스 토그램은 동일 입력벡터에 대해 선택된 입력화자의 코드벡터와 기준화자의 코드벡터 사이의 DTW(dynamic time warping)을 행하여 대응하는 코드벡터들의 소속값 (membership value)을 누적하여 얻는다. 음성합성시에는 사상된 코드북을 사용하여 입력화자의 음성을 퍼지벡터 양자화한 다음, FCM(fuzzy c means) 합성규칙을 사용하 여 사상된 코드북내의 코드벡터가 아닌 새로운 하나의 합성벡터를 얻게 되어 좀 더 입력화자에 적응된 합성음을 얻게 된다. 이 기술의 성능평가는 성별이 서로 다른 화 자를 입력화자 및 기준화자로 선정하여 입력화자의 음성에 가까운 정도로 평가하였으 며 그 결과 기존의 음성합성보다 입력화자에 더 적용된 합성음을 얻었다.

• 한국안전학회지
• /
• 제15권4호
• /
• pp.69-77
• /
• 2000

This paper presents a study for decision method of human body safety according to power load fluctuation of the Chechon area. To cope with substantial electromagnetics analysis, the safety assessments were analyzed the double 154 [kV] T/L, 345 [kV] T/L, Chechon-Ichon, Chechon-Youngju, respectively. As the results of case study, in case of 345 [kV] T/L, the electric field value was 11.4927 [kV/m], magnetic field value was 0.4622 [G] at the point about 7 [m] away from the line in severest case. The results showed that the membership of a proposition fuzzy '11.4937 [kV/m] is hazardous' is 0.7063 for the severest case in case of 345 [kV] T/L. As the analytic results using the fuzzy qualifier term, the membership in case of very false is 0.0862 and fairly false is 0.5419, respectively.

• 한국융합학회논문지
• /
• 제6권5호
• /
• pp.227-232
• /
• 2015

퍼지 선형계획법은 불확실성하에서의 문제들을 해결하는데 유용한 의사결정 모형이다. 본 연구에서는 목적함수 값이 퍼지수이고 우변 상수도 퍼지수인 융합 등식 제약식을 갖는 퍼지 선형계획법 문제를 다룬다. 연구의 목적은 퍼지 해를 정의하고 그것을 구하는 절차를 모색하는 것이다. 목적함수 값에 대한 소속 함수로 부분 선형함수를, 제약식의 소속 함수로는 사다리꼴 함수를 도입한다. 사다리꼴 함수는 구간별 선형 함수 들로 나누어 나타낼 수 있다. 따라서 모든 소속 함수들을 선형식 들로 대체함으로써 퍼지 선형계획 모형을 Zimmermann의 대칭 선형 모형으로 바꿀 수 있다. 여기에 최대-최소 기준을 적용하여 일반 선형계획법 문제를 도출해 내고, 이 문제의 최적해로부터 원 문제의 퍼지 해를 얻게 된다. 본 논문에서는 사다리꼴 소속 함수에 대해 살펴보았는데 앞으로는 오목 부분 선형함수와 같은 좀 더 일반화된 소속 함수에 대한 연구가 필요하다.