• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.3004-3006
• /
• 2000

In this paper, we propose an optimal design method of Fuzzy-Neural Networks model with parallel structure for complex and nonlinear systems. The proposed model is consists of a multiple number of FNN connected in parallel. The proposed FNNs with parallel structure is based on Yamakawa's FNN and it uses simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rules. We use a HCM clustering and GAs to identify the structure and the parameters of the proposed model. Also, a performance index with a weighting factor is presented to achieve a sound balance between approximation and generalization abilities of the model. To evaluate the performance of the proposed model. we use the time series data for gas furnace and the numerical data of nonlinear function.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부 B
• /
• pp.654-656
• /
• 1998

This paper presents an optimal identification method of nonlinear and complex system that is based on fuzzy-neural network(FNN). The FNN used simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM Algorithm to find initial parameters of membership function. And then to obtain optimal parameters, we use the genetic algorithm. Genetic algorithm is a random search algorithm which can find the global optimum without converging to local optimum. The parameters such as membership functions, learning rates and momentum coefficients are easily adjusted using the genetic algorithms. Also, the performance index with weighted value is introduced to achieve a meaningful balance between approximation and generalization abilities of the model. To evaluate the performance of the FNN, we use the time series data for 9as furnace and the sewage treatment process.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1993년도 Fifth International Fuzzy Systems Association World Congress 93
• /
• pp.834-837
• /
• 1993

This paper describes the implementation of an autonomous fuzzy logic controller. The controller is endowed with basic control principles and learning constructs which enable it to autonomously modify its control policy based on system performance. The controller lies dormant when system response is satisfactory but if rapidly initiates adaptation in real time when adverse performance is observed. The autonomous fuzzy controller is implemented on an Intel MCS-51 series micro-controller board using an inexpensive 8-bit Intel 8031 processor. The 11.06 MHz micro-controller operates at a rate exceeding 200 "global" look-up table reinforcements per second. This is important when developing practical on-line adaptive controllers for fast systems. It is also significant because an initial controller look-up table could be incorrect or non-existent. The relatively high learning rate enables the controller to learn to control a system even while it is controlling it.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2099-2101
• /
• 2001

In general, the thermal control system has nonlinearity and the time delay, futhermore, it is difficult to design the free size controller, because the external environmental disturbances, such as rapid temperature change. Many researchers in this field are preferring to adapt the fuzzy logic control methods. But it is noted that the actuator identification of M.F.'s used in FLC is very difficult. Therefore in this paper, an implementation technique of thermal control system using RVEGA based optimal fuzzy control was proposed. It's superiority and exaction in controller design processes hardware in implementation were proved through a series of simulations and experimentations.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권3호
• /
• pp.186-199
• /
• 2013

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance. We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels. Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1161-1164
• /
• 1996

In this paper, a parameter optimization technique for a mobile robot navigation is discussed. Authors already have proposed a navigation algorithm for mobile robots with sonar sensors using fuzzy decision making theory. Fuzzy decision making selects the optimal via-point utilizing membership values of each via-point candidate for fuzzy navigation goals. However, to make a robot successfully navigate through an unknown and cluttered environment, one needs to adjust parameters of membership function, thus changing shape of MF, for each fuzzy goal. Furthermore, the change in robot configuration, like change in sensor arrangement or sensing range, invokes another adjusting of MFs. To accomplish an intelligent way to adjust these parameters, we adopted a genetic algorithm, which does not require any formulation of the problem, thus more appropriate for robot navigation. Genetic algorithm generates the fittest parameter set through crossover and mutation operation of its string representation. The fitness of a parameter set is assigned after a simulation run according to its time of travel, accumulated heading angle change and collision. A series of simulations for several different environments is carried out to verify the proposed method. The results show the optimal parameters can be acquired with this method.

• 대한전기학회논문지:전력기술부문A
• /
• 제53권10호
• /
• pp.529-535
• /
• 2004

Load forecasting is essential in the electricity market for the participants to manage the market efficiently and stably. A wide variety of techniques/algorithms for load forecasting has been reported in many literatures. These techniques are as follows: multiple linear regression, stochastic time series, general exponential smoothing, state space and Kalman filter, knowledge-based expert system approach (fuzzy method and artificial neural network). These techniques have improved the accuracy of the load forecasting. In recent 10 years, many researchers have focused on artificial neural network and fuzzy method for the load forecasting. In this paper, we propose an algorithm of a hybrid load forecasting method using fuzzy linear regression and general exponential smoothing and considering the sensitivities of the temperature. In order to consider the lower load of weekends and Monday than weekdays, fuzzy linear regression method is proposed. The temperature sensitivity is used to improve the accuracy of the load forecasting through the relation of the daily load and temperature. And the normal load of weekdays is easily forecasted by general exponential smoothing method. Test results show that the proposed algorithm improves the accuracy of the load forecasting in 1996.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 추계학술대회 학술발표 논문집
• /
• pp.99-104
• /
• 1998

본 논문은 효과적인 시계열 예측을 위한 새로운 퍼지 학습방법을 제안한다. 기존의 학습방법에서는 입력 데이터를 F(y(t),y(t-1),y(t-2)..)의 형태로 주어 예측을 수행했으나 본 논문에서 제안한 방법에서는 입력 데이터를 F(y(t)-y(t-1),y(t-1)-y(t-2)..)로 설정한다. 이것은 각 입력값의 차이를 새로운 입력으로 사용함으로써 유사한 시계열 분포를 좀더 능동적인 퍼지 규칙으로 만들기 때문에 Non-stationary한 데이터뿐만 아니라 기존의 시계열 데이터 예측방법 보다 나은 결과를 나타낸다. 알고리즘의 수행능력을 살펴보기 위해 Mackey-Glass time series와 Lorenz data를 사용하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.514-518
• /
• 1996

This paper focuses on automated generation of the knowledge base for a fuzzy network performance manager in order to satisfy delay constraints imposed on time-critical messages while maintaining as much network capacity as possible for non-time-critical messages. Therefore, the bowlegs base is formulated to minimize a certain penalty function by using a type of genetic algorithm. The efficacy of the formulation method has been demonstrated by a series of simulation experiments.

• 대한교통학회지
• /
• 제22권3호
• /
• pp.179-201
• /
• 2004

지능형 교통체계(intelligent transport systems)의 구축이 점차 널리 확대됨에 따라 교통류의 실시간 모형화(online traffic flow modeling)의 중요성이 증대되고 있다. 교통량-밀도 관계는 주어진 교통량, 밀도 상황에서 교통류의 행태를 나타낼 뿐만 아니라 거시 교통류 모형의 결과에 많은 영향을 미친다. 현재까지 교통량-밀도관계에 관한 대부분의 연구는 그 관계식을 규명하는데 그치고 있다. 상류부와 하류부의 교통 상태에 따른 교통량-밀도관계의 시간적 변화는 교통류의 모형화에 반드시 고려되어야 할 특성이지만, 현재까지 그에 대한 연구가 폭넓게 이루어지지 않고 있는 실정이다. 본 논문에서는 한 지점에서의 교통량-밀도관계가 시간의 흐름에 따라 분석되었고 states diagram으로 표현되었다. 동적 교통량-밀도관계 (dynamic flow-density relation)는 states diagram으로부터 fuzzy-logic을 이용하여 유추되었고, 거시 교통류모형을 실시간으로 응용할 수 있는 기초를 제공하였다. 동적 교통량-밀도관계를 거시 교통류 모형에 이용함으로써 교통류의 실시간 모형화 과정에서 발생하는 모수추정 (parameter calibration) 문제를 완화하였다.