• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.302-305
• /
• 1996

In this paper, an optimal identification method using fuzzy-neural networks is proposed for modeling of nonlinear complex systems. The proposed fuzzy-neural modeling implements system structure and parameter identification using the intelligent schemes together with optimization theory, linguistic fuzzy implication rules, and neural networks(NNs) from input and output data of processes. Inference type for this fuzzy-neural modeling is presented as simplified inference. To obtain optimal model, the learning rates and momentum coefficients of fuzz-neural networks(FNNs) and parameters of membership function are tuned using genetic algorithm(GAs). For the purpose of its application to nonlinear processes, data for route choice of traffic problems and those for activated sludge process of sewage treatment system are used for the purpose of evaluating the performance of the proposed fuzzy-neural network modeling. The show that the proposed method can produce the intelligence model w th higher accuracy than other works achieved previously.

• 한국정보처리학회논문지
• /
• 제4권8호
• /
• pp.2070-2079
• /
• 1997

본 논문에서는 호 수락 제어 문제를 해결하기 위해 퍼지 논리 제어기의 장점과 신경망의 학습 능력을 이용한 ATM 망의 호 수락 제어 시스템을 제안하였다. ATM 망의 새로운 호는 현재 서비스 중인 호의 서비스 품질(QoS : quality of service)이 영향을 받지 않을 경우 망에 접속이 된다. 신경망 호 수락 제어 시스템은 입/출력 패턴의 학습으로 예측성 잇게 호 수락/거절을 하는 시스템이다. 본 논문의 퍼지 신경망 호 수락 제어 시스템에서는 학습 속도 개선을 위해 학습율과 모맨텀 상수에 퍼지 추론을 적용하였다. 이 시스템은 시뮬레이션을 통해 기존의 신경망 방법과 퍼지 신경망 방법에서의 학습 횟수 측정으로 제안 알고리즘의 우수성을 검증하였다. 시뮬레이션 결과 퍼지 학습 규칙에 근거한 퍼지 신경망 CAC(call admission control) 방식이 종래의 신경망 이론에 근거한 CAC 방식보다 학습 속도면에서 약 5배의 속도 향상이 있었다.

• 한국지능시스템학회논문지
• /
• 제7권1호
• /
• pp.45-53
• /
• 1997

This paper presents hybrid expert system for diagnosis of electric power transformer faults. The expert system diagnose and detect faults in oil-filled power transformers based on dissolved gas analysis. As the preprocessing stage, fuzzy information theory is used to manage the uncertainty in transformer fault diagnosis using dissolved gas analysis. The Kohonen neural network takes the interim results by applying fuzzy informations theory as inputs, and performs the transformer fault diagnosis. The Proposed system tested gas records of power transformers from Korea Electric Power Corporation to verify the diagnosis performance of transformer faults.

• 전자공학회논문지CI
• /
• 제38권6호
• /
• pp.18-25
• /
• 2001

분석적학습과 귀납적학습을 결합한 지식기반신경망은 다른 기계학습모델보다 우수한 성능을 나타내고 있다. 그러나 지식기반신경망에서는 신경망이 형성된 후 그 구조를 동적으로 변경할 수 없어서 영역이론정련화 기능을 제공하지 못한다. 이러한 단점을 갖고 있는 지식기반신경망을 보완하기 위하여 TopGen 알고리즘이 제안되었지만 부분적인 문제점을 안고 있다. 본 논문에서는 TopGen의 문제점을 해소하면서 지식기반 신경망을 확장하여 영역이론정련기능을 부여하는 방안 2가지를 제시하고 이를 평가하였다.

• 대한기계학회논문집
• /
• 제18권8호
• /
• pp.1939-1951
• /
• 1994

This study introduces a new tool breakage detecting technology comprised of an unsupervised neural network combined with adaptive time series autoregressive(AR) model where parameters are estimated recursively at each sampling instant using a parameter adaptation algorithm based on an RLS(Recursive Least Square). Experiment indicates that AR parameters are good features for tool breakage, therefore it can be detected by tracking the evolution of the AR parameters during milling process. an ART 2(Adaptive Resonance Theory 2) neural network is used for clustering of tool states using these parameters and the network is capable of self organizing without supervised learning. This system operates successfully under the wide range of cutting conditions without a priori knowledge of the process, with fast monitoring time.

• 한국컴퓨터정보학회논문지
• /
• 제10권4호
• /
• pp.17-27
• /
• 2005

본 논문에서는 문자인식과 복원을 위한 신경회로망 시스템을 제안한다. 제안하는 시스템은 인식부와 연상부로 구성되었다. 인식부에서는 ART 신경회로망의 인식성능을 개선하기 위해 불필요한 하향틀의 생성과 변화를 제한하여 효과적인 패턴인식이 가능한 모델을 제안하였다. 또한, 한글의 구조적인 특징을 능동적으로 적용할 수 있게 구성된 위치특징 추출 알고리즘을 적용하였다. 연상부에서는 Hopfield 신경회로망으로, 입력된 이미지 패턴의 복원이 가능한 모델을 구성하였다. 제안하는 시스템은 그 성능을 확인하기 위해 각 부분별 실험을 하였다. 그 결과 인식율이 개선되고 복원이 가능함을 보였다.

• 제어로봇시스템학회논문지
• /
• 제10권12호
• /
• pp.1249-1255
• /
• 2004

A new parameter adaptation scheme for RBF Neural Network (NN) has been developed in this paper. Even though the RBF Neural Network (NN) based controllers are robust against both un-modeled dynamics and external disturbances, the performance is not satisfactory for a fast and precise mobile robot. To improve the tracking performance as well as robustness, all the parameters of RBF NN are updated in real time. The stability of this control law is rigorously proved by following the Lyapunov stability theory and shown by the experimental simulations. The fact that all of the weighting factors, width and center of RBF NN have been updated implies that this scheme utilizes all the possibilities in RBF NN to make the controller robust and precise while the mobile robot is following un-known trajectories. The performance of this new algorithm has been compared to the conventional RBF NN controller where some of the parameters are adjusted for robustness.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 V
• /
• pp.2775-2778
• /
• 2003

Neural network-based fault diagnosis algorithm to detect and isolate faults in the nonlinear systems is proposed. In the proposed method, the fault is detected when the errors between the system output and the neural network nominal system output cross a predetermined threshold. Once a fault in the system is detected, the system outputs are transferred to the fault classifier by ART2 NN (adaptive resonance theory 2 neural network) for fault isolation. From the computer simulation results, it is verified that the proposed fault diagonal method can be performed successfully to detect and isolate faults in a nonlinear system.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권12호
• /
• pp.1498-1506
• /
• 1999

In this paper, a hybridization of Evolutionary Strategy (ES) and a Two-Phase Neural Network(TPNN) is applied to the optimal environmental and economic operation. As the evolutionary computation, ES is to search for the global optimum based on natural selection and genetics but it shows a defect of reducing the convergence rate in the latter part of search, and often does not search the exact solution. Also, neural network theory as a local search technique can be used to search a more exact solution. But it also has the defect that a solution frequently sticks to the local region. So, new algorithm is presented as hybrid methods by combining merits of two methods. The hybrid algorithm has been tested on Emission Constrained Economic Dispatch (ECED) problem and Weighted Emission Economic Dispatch (WEED) problem for optimal environmental and economic operation. The result indicated that the hybrid approach can outperform the other computational efficiency and accuracy.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.173.4-173
• /
• 2001

This paper presents a stable nonlinear adaptive control for AUV(Autonomous Underwater Vehicle) by using neural network. AUV's dynamics are highly nonlinear, and their hydrodynamic coefficients vary with different operational conditions. In this paper, the nonlinear uncertainties of the AUV's dynamics are approximated by using LPNN(Linearly parameterized Neural Network). The presented controller is consist of three parallel terms; linear feedback control, sliding mode control, and adaptive control(LPNN). Lyapunov theory is used to guarantee the stability of tracking errors and neural network´s weights errors. Numerical simulations for nonlinear control of the AUV show the effectiveness of the proposed techniques.