• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 B
• /
• pp.578-580
• /
• 1998

The paper presents an approach to the structure identification based on the CART (Classification And Regression Tree) algorithm and to the parameter identification by hybrid learning method in neuro-fuzzy system. By using the CART algorithm, the proposed method can roughly estimate the numbers of membership function and fuzzy rule using the centers of decision regions. Then the parameter identification is carried out by the hybrid learning scheme using BP (Back-propagation) and RLSE (Recursive Least Square Estimation) from the numerical data. Finally, we will show it's usefulness for fuzzy modeling to truck backer upper control.

• 인터넷정보학회논문지
• /
• 제16권6호
• /
• pp.69-78
• /
• 2015

다중 에이전트 시스템에 대한 연구는 최근 다양한 분야에서 활성화 되고 있으며, 복잡한 시스템의 제어 및 최적화에 관한 연구가 진행되어 왔다. 본 논문에서는 게임 환경에서의 NPC(Non-Player Character) 시뮬레이션을 위한 다중 에이전트 시스템을 개발한다. 시스템 개발의 목적은 동적 이산사건 영역의 상황을 추론하여 신속하고 정확한 판단을 제공하고 에이전트 시스템의 최적화 과정을 보다 손쉽게 도와주는데 있다. 이를 위한 에이전트 시스템의 기본 모델은 페트리넷을 활용하여 구조를 단순화 하고 퍼지 추론엔진을 사용하여 다양한 상황을 결정할 수 있도록 하였다. 본 연구 시스템의 실험은 NPC간의 가상 전장 상황을 묘사하며, 퍼지 규칙이 적용된 에이전트와 유한 상태 기계로 구현된 NPC를 시뮬레이션 하여 에이전트의 승률과 생존율을 산출하였다. 실험 결과 퍼지 규칙 기반 에이전트의 승률과 생존율이 유한 상태 기계로 구현된 NPC보다 더 높은 것으로 나타났다.

• 한국지능시스템학회논문지
• /
• 제11권4호
• /
• pp.354-358
• /
• 2001

본 논문에서는 underutilization 문제를 해결한 퍼지 신경회로망 모델을 제시한다. 이 퍼지 신경 회로망은 ART-1 신경회로망과 유사한 제어 구조를 가지고 있어 유연성이 있으면서도 안정성이 있다. 또한 연결강도의 초기화가 필요 없고 ART-1 신경회로망에 비하여 잡음에 민감하지 않다. 이 퍼지 신경회로망의 학습법칙은 코호넨의 학습법칙을 변형하고 퍼지화 하였으며 누설 경쟁학습의 퍼지화와 조건 확률의 퍼지화에 기반을 두고 있다. 출력 뉴런 중에서 승자를 정한 후에 행해지는 점검 테스트에서는 유사척도로 상대적 거리를 사용하였다. 이 상대적 거리는 유클리디안 거리와 함께 데이터와 클러스터들의 대푯값들 간의 상대적인 위치를 고려한 것이다. 본 논문에서 제안한 퍼지 신경회로망과 코호넨 자기 조직화 특징 지도의 성능을 비교하기 위하여 널리 사용되어온 IRIS 데이터와 가우시안 분포 데이터를 사용하였다.

• 한국지능시스템학회논문지
• /
• 제15권2호
• /
• pp.236-244
• /
• 2005

본 연구에서는 퍼지관계 및 진화론적 최적 다층 퍼셉트론에 기초한 퍼지다항식 뉴럴네트워크(FPNN)의 새로운 구조를 소개하고, 포괄적인 설계방법론을 토의하며, 그리고 일련의 수치적인 실험이 수행된다. 진화론적 최적 FPNN(EFPNN)의 구축을 위해 컴퓨터지능(CI)의 기반 기술을 이용한다. EFPNN의 구조는 규칙베이스 퍼지뉴럴네트워크와 다항식 뉴럴네트워크의 결합에 의한 유전자 최적 구동 하이브리드 시스템의 시너지 이용으로 얻어진다. 퍼지뉴럴네트워크는 EFPNN의 전체규칙 구조의 전반부에 기여하고, EFPNN의 후반부는 다항식 뉴럴네트워크를 사용하여 설계된다. EFPNN의 후반부를 위한 유전론적 최적 다항식 뉴럴네트워크의 개발은 두 최적화 기법에 의해 수행된다. 즉 구조적 최적화는 유전자알고리즘에 의해 수행되고, 파라미터 최적화는 최소자승법 기반의 학습을 통해 행하여진다. EFPNN의 성능 평가를 위해, 모델은 몇 가지 수치 예제를 이용한다. 비교에 의한 해석은 제안된 EFPNN이 이전에 제시된 다른 지능형 모델보다 높은 정확도 뿐만 아니라 좀 더 우수한 예측능력을 가지는 모델임을 보여준다.

• 전기학회논문지
• /
• 제59권5호
• /
• pp.981-989
• /
• 2010

In this study, we introduce a new architecture of fuzzy inference system. In the fuzzy inference system, we use Fuzzy C-Means clustering algorithm to form the premise part of the rules. The membership functions standing in the premise part of fuzzy rules do not assume any explicit functional forms, but for any input the resulting activation levels of such radial basis functions directly depend upon the distance between data points by means of the Fuzzy C-Means clustering. As the consequent part of fuzzy rules of the fuzzy inference system (being the local model representing input output relation in the corresponding sub-space), four types of polynomial are considered, namely constant, linear, quadratic and modified quadratic. This offers a significant level of design flexibility as each rule could come with a different type of the local model in its consequence. Either the Least Square Estimator (LSE) or the weighted Least Square Estimator (WLSE)-based learning is exploited to estimate the coefficients of the consequent polynomial of fuzzy rules. In fuzzy modeling, complexity and interpretability (or simplicity) as well as accuracy of the obtained model are essential design criteria. The performance of the fuzzy inference system is directly affected by some parameters such as e.g., the fuzzification coefficient used in the FCM, the number of rules(clusters) and the order of polynomial in the consequent part of the rules. Accordingly we can obtain preferred model structure through an adjustment of such parameters of the fuzzy inference system. Moreover the comparative experimental study between WLSE and LSE is analyzed according to the change of the number of clusters(rules) as well as polynomial type. The superiority of the proposed model is illustrated and also demonstrated with the use of Automobile Miles per Gallon(MPG), Boston housing called Machine Learning dataset, and Mackey-glass time series dataset.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 추계학술대회 학술발표 논문집
• /
• pp.252-256
• /
• 1997

The success of a fuzzy neural network(FNN) control system solving any given problem critically depends on the architecture of the network. Various attempts have been made in optimizing its structure using genetic algorithm automated designs. This paper presents a new approach to structurally optimized designs of FNN models. A messy genetic algorithm is used to obtain structurally optimized FNN models. Structural optimization is regarded important before neural networks based learning is switched into. We have applied the method to the problem of a numerical approximation

• 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.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2762-2764
• /
• 2001

In this study, we introduce and investigate a class of dynamic perceptron architectures, discuss a comprehensive design methodology and carry out a series of numeric experiments. The proposed dynamic perceptron architectures are called as Polynomial Neural Networks(PNN). PNN is a flexible neural architecture whose topology is developed through learning. In particular, the number of layers of the PNN is not fixed in advance but is generated on the fly. In this sense, PNN is a self-organizing network. PNN has two kinds of networks, Polynomial Neuron(FPN)-based and Fuzzy Polynomial Neuron(FPN)-based networks, according to a polynomial structure. The essence of the design procedure of PN-based Self-organizing Polynomial Neural Networks(SOPNN) dwells on the Group Method of Data Handling (GMDH) [1]. Each node of the SOPNN exhibits a high level of flexibility and realizes a polynomial type of mapping (linear, quadratic, and cubic) between input and output variables. FPN-based SOPNN dwells on the ideas of fuzzy rule-based computing and neural networks. Simulations involve a series of synthetic as well as experimental data used across various neurofuzzy systems. A detailed comparative analysis is included as well.

• International Journal of Control, Automation, and Systems
• /
• 제1권2호
• /
• pp.194-202
• /
• 2003

In this paper, we introduce a category of Multi-FNN (Fuzzy-Neural Networks) models, analyze the underlying architectures and propose a comprehensive identification framework. The proposed Multi-FNNs dwell on a concept of fuzzy rule-based FNNs based on HCM clustering and evolutionary fuzzy granulation, and exploit linear inference being treated as a generic inference mechanism. By this nature, this FNN model is geared toward capturing relationships between information granules known as fuzzy sets. The form of the information granules themselves (in particular their distribution and a type of membership function) becomes an important design feature of the FNN model contributing to its structural as well as parametric optimization. The identification environment uses clustering techniques (Hard C - Means, HCM) and exploits genetic optimization as a vehicle of global optimization. The global optimization is augmented by more refined gradient-based learning mechanisms such as standard back-propagation. The HCM algorithm, whose role is to carry out preprocessing of the process data for system modeling, is utilized to determine the structure of Multi-FNNs. The detailed parameters of the Multi-FNN (such as apexes of membership functions, learning rates and momentum coefficients) are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. To evaluate the performance of the proposed model, two numeric data sets are experimented with. One is the numerical data coming from a description of a certain nonlinear function and the other is NOx emission process data from a gas turbine power plant.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권3호
• /
• pp.135-144
• /
• 2004

In this paper, we propose competitive fuzzy polynomial neurons-based advanced Self-Organizing Neural Networks(SONN) architecture for optimal model identification and discuss a comprehensive design methodology supporting its development. The proposed SONN dwells on the ideas of fuzzy rule-based computing and neural networks. And it consists of layers with activation nodes based on fuzzy inference rules and regression polynomial. Each activation node is presented as Fuzzy Polynomial Neuron(FPN) which includes either the simplified or regression polynomial fuzzy inference rules. As the form of the conclusion part of the rules, especially the regression polynomial uses several types of high-order polynomials such as linear, quadratic, and modified quadratic. As the premise part of the rules, both triangular and Gaussian-like membership (unction are studied and the number of the premise input variables used in the rules depends on that of the inputs of its node in each layer. We introduce two kinds of SONN architectures, that is, the basic and modified one with both the generic and the advanced type. Here the basic and modified architecture depend on the number of input variables and the order of polynomial in each layer. The number of the layers and the nodes in each layer of the SONN are not predetermined, unlike in the case of the popular multi-layer perceptron structure, but these are generated in a dynamic way. The superiority and effectiveness of the Proposed SONN architecture is demonstrated through two representative numerical examples.