• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.11a
• /
• pp.267-270
• /
• 1997

In this paper, we proposed an optimazation method using Genetic Algorithm for nonlinear system modeling. Fuzzy Neural Network(FNNs) was used as basic model of nonlinear system. FNNs was fused of Fuzzy Inference which has linguistic property and Neural Network which has learning ability and high tolerence level. This paper, We used FNNs which was proposed by Yamakawa. The FNNs was composed Simple Inference and Error Back Propagation Algorithm. To obtain optimal model, parameter of membership function, learning rate and momentum coefficient of FNNs are tuned using genetic algorithm. And we used simplex algorithm additionaly to overcome limit of genetic algorithm. For the purpose of evaluation of proposed method, we applied proposed method to traffic choice process and waste water treatment process, and then obtained more precise model than other previous optimization methods and objective model.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.118-121
• /
• 2000

본 논문에서는 FNNs(feedforwad neural networks)구현을 위한 새로운 학습 방안을 제안하였다. 제안된 방식은 온 칩 학습이 가능하도록 FNNs와 학습회로 사이에 스위칭 회로를 추가하여 단일패턴과 다중패턴 학습이 가능하도록 구현하였다. 학습 회로는 MEBP(modified error back-propagation) 학습 규칙을 적용하였고 간단한 비선형 시냅스 회로를 이용하여 구현하였다. 제안된 방식은 표준 CMOS 공정으로 구현되었고, MOSIS AMI $1.5\mu\textrm{m}$공정 HSPICE 파라메터를 이용하여 그 동작을 검증하였다. 제안된 학습방안 및 비선형 회로는 향후 학습 기능을 가진 대규모의 FNNs 구현에 매우 적합하리라 예상된다.

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.420-423
• /
• 2001

In this paper, the Multi-FNNs(Fuzzy-Neural Networks) architecture is identified and optimized using HCM(Hard C-Means) clustering method and genetic algorithms. The proposed Multi-FNNs architecture uses simplified inference and linear inference as fuzzy inference method and error back propagation algorithm as learning rules. Here, HCM clustering method, which is carried out for the process data preprocessing of system modeling, is utilized to determine the structure of Multi-FNNs according to the divisions of input-output space using I/O process data. Also, the parameters of Multi-FNNs model such as apexes of membership function, learning rates and momentum coefficients are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is used 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 NOx emission process data of gas turbine power plant.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.3
• /
• pp.621-632
• /
• 2013

In this paper, we introduce advanced architectures of genetically-oriented Fuzzy Neural Networks (FNNs) based on fuzzy set and fuzzy relation and discuss a comprehensive design methodology. The proposed FNNs are based on 'if-then' rule-based networks with the extended structure of the premise and the consequence parts of the fuzzy rules. We consider two types of the FNNs topologies, called here FSNN and FRNN, depending upon the usage of inputs in the premise of fuzzy rules. Three different type of polynomials function (namely, constant, linear, and quadratic) are used to construct the consequence of the rules. In order to improve the accuracy of FNNs, the structure and the parameters are optimized by making use of genetic algorithms (GAs). We enhance the search capabilities of the GAs by introducing the dynamic variants of genetic optimization. It fully exploits the processing capabilities of the FNNs by supporting their structural and parametric optimization. To evaluate the performance of the proposed FNNs, we exploit a suite of several representative numerical examples and its experimental results are compared with those reported in the previous studies.

• International Journal of Contents
• /
• v.8 no.4
• /
• pp.30-35
• /
• 2012

Error surfaces provide us with very important information for training of feed-forward neural networks (FNNs). In this paper, we draw the contour plots of various error or objective functions for training of FNNs. Firstly, when applying FNNs to classifications, the weakness of mean-squared error is explained with the viewpoint of error contour plot. And the classification figure of merit, mean log-square error, cross-entropy error, and n-th order extension of cross-entropy error objective functions are considered for the contour plots. Also, the recently proposed target node method is explained with the viewpoint of contour plot. Based on the contour plots, we can explain characteristics of various error or objective functions when training of FNNs proceeds.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.8
• /
• pp.1615-1623
• /
• 2009

In order to develop reliable on-site partial discharge (PD) pattern recognition algorithm, we introduce Type-2 Fuzzy Neural Networks (T2FNNs) optimized by means of Particle Swarm Optimization(PSO). T2FNNs exploit Type-2 fuzzy sets which have a characteristic of robustness in the diverse area of intelligence systems. Considering the on-site situation where it is not easy to obtain voltage phases to be used for PRPDA (Phase Resolved Partial Discharge Analysis), the PD data sets measured in the laboratory were artificially changed into data sets with shifted voltage phases and added noise in order to test the proposed algorithm. Also, the results obtained by the proposed algorithm were compared with that of conventional Neural Networks(NNs) as well as the existing Radial Basis Function Neural Networks (RBFNNs). The T2FNNs proposed in this study were appeared to have better performance when compared to conventional NNs and RBFNNs.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.63-67
• /
• 2004

The conventional typical techniques as fast codebook generation methods are PDS, FNNS, and FC. In this paper, we propose FCNNPDS integrated the conventional methods to generate a codebook. The results of simulations show that the computational magnitude of FCNNPDS is reduced to 40-95% lower than conventional techniques. But comparison computation has no relation with k dimension of vectors, that is, because the computational magnitude of comparison is smaller than others, therefore FCNNPDS may be the best method than the conventional methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.172.4-172
• /
• 2001

In this paper, the Generalized Predictive Control(GPC) method based on Fuzzy Neural Networks(FNNs) is presented for the control of chaotic nonlinear systems without precise mathematical models. In our method, FNNs is used as the predictor whose parameters are tuned by the error between the actual output of nonlinear chaotic system and that of FNNs model. The parameters of GPC controller are adjusted via the gradient descent method where the difference between the actual output and the reference signal is used as a control error. Finally, computer simulation on the representative continuous-time chaotic system(Duffing system) is presented to demonstrate the effectiveness of our chaos control method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.03a
• /
• pp.99-102
• /
• 1998

본 논문에서는 복잡하고 비선형적인 시스템의 최적 모델링을 우해서 지능형 퍼지-뉴럴네트워크의 최적 모델 구축을 위한 방법을 제안한다. 기본 모델은 퍼지 추론 시스템의 언어적인 규칙생성의 장점과 뉴럴 네트워크의 학습기능을 결합한 FNNs 모델을 사용한다. FNNs 모델의 퍼지 추론부는 간략추론이 사용되고, 학습은 요류 역전파 알고리즘을 사용하여 다른 모델들에 비해 학습속도가 빠르고 수렴능력이 우수하다. 그러나 기본 모델은 주어진 시스템에 대하여 퍼지 공간을 균등하게 분할하여 퍼지 소속을 정의한다. 이것은 비선형 시스템의 모델링에 있어어서 성능을 저하시켜 최적의 모델을 얻기가 어렵다. 논문에서는 주어진 데이터의 특성을 부여한 공간을 설정하기 위하여 클러스터링 알고리즘을 사용한다. 클러스터링 알고리즘은 주어진 시스템에 대하여 상호 연관성이 있는 데이터들끼리 특성을 나누어 몇 개의 클래스를 이룬다. 클러스터링 알고리즘을 사용하여 초기 FNNs 모델의 퍼지 공간을 나누고 소속함수를 정의한다. 또한, 최적화 기법중의 하나로 자연선택과 자연계의 유전자 메카니즘에 바탕을 둔 탐색 알고리즘인 유전자 알고리즘을 사용하여 주\ulcorner 진 모델에 대하여 최적화를 수행한다. 또한 본 연구에서는 학습 및 테스트 데이터의 성능 결과의 상호 균형을 얻기 위한 하중값을 가긴 성능지수가 제시된다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.17-20
• /
• 2000

본 논문에서는 간단한 비선형 시냅스 회로를 이용하여 온 칩 학습기능을 포함한 모듈 칩을 구현하였다. 학습 회로는 MEBP(modified error back-propagation) 학습 규칙을 적용하여 구현하였으며, 제안된 회로는 표준 CMOS 공정으로 구현되었고, MOSIS AMI $1.5\mu\textrm{m}$공정 HSPICE 파라메터를 이용하여 그 동작을 검증하였다. 구현된 모듈 칩은 온 칩 학습기능을 가진 확장 가능한 신경회로망 칩으로 대규모의 FNNs(feedforwad neural networks) 구현에 매우 적합하리라 예상된다.