• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.05a
• /
• pp.329-332
• /
• 2003

본 논문에서는 새로이 제안된 자기 구조화하는(Self-structuring) 새로운 Radial-Basis Function Network(RBFN)에 대해서 실험적인 검증을 했다. 이 자기 구조화하는 새로운 RBFN은 기존의 RBFN과 비교해서 여러 장점이 있다. Lyapunov 이론에 기초해서 새로운 학습 규칙을 선정하였기 때문에 시스템의 안정도를 보장할 수 있다. 그리고, 자기 구조화의 과정 즉, 생성과 병합을 통해 은닉층에서 적정수의 뉴런을 결정할 수 있다. 기존의 RBFN과 성능을 비교하기 위하여, 실제 비선형 시스템인 2축 암로봇에 대해 실험한 결과를 보였다. 결과적으로, 우리는 실험결과를 통해 자기 구조화하는 RBFN의 효율적인 구조와 시스템에 대한 안정도를 보장함을 볼 수 있다.

• ETRI Journal
• /
• v.28 no.1
• /
• pp.59-66
• /
• 2006

Learning behaviors of a radial basis function network (RBFN) using a singular value decomposition (SVD) and stochastic gradient (SG) algorithm, together named RBF-SVD-SG, for odor sensing systems are analyzed, and a fast training method is proposed. RBF input data is from a conducting polymer sensor array. It is revealed in this paper that the SG algorithm for the fine-tuning of centers and widths still shows ill-behaving learning results when a sufficiently small convergence coefficient is not used. Since the tuning of centers in RBFN plays a dominant role in the performance of RBFN odor sensing systems, our analysis is focused on the center-gradient variance of the RBFN-SVD-SG algorithm. We found analytically that the steadystate weight fluctuation and large values of a convergence coefficient can lead to an increase in variance of the center-gradient estimate. Based on this analysis, we propose to use the least mean square algorithm instead of SVD in adjusting the weight for stable steady-state weight behavior. Experimental results of the proposed algorithm have shown faster learning speed and better classification performance.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.26.6-26
• /
• 2001

In this paper we introduce a new algorithm that enables radial basis function network(RBFN) to be structured automatically and guarantees the stability of the RBFN. Because this new algorithm is efficient and also have the advantage of fast computational speed we adopt this algorithm as online learning scheme for uncertain nonlinear dynamical systems. Based on the fact that a 3-layered RBFN can represent a specific nonlinear function reasonably well by linearly combining a set of nonlinear and localized basis functions, we show that this RBFN can identify the nonlinear system very well without knowing the information of the system in advance.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.36 no.4
• /
• pp.59-63
• /
• 2013

In biomass gasification, efficiency of energy quantification is a difficult part without finishing the process. In this article, a radial basis function neural network (RBFN) is proposed to predict biomass efficiency before gasification. RBFN will be compared with a principal component regression (PCR) and a multilayer perceptron neural network (MLPN). Due to the high dimensionality of data, principal component transform is first used in PCR and afterwards, ordinary regression is applied to selected principal components for modeling. Multilayer perceptron neural network (MLPN) is also used without any preprocessing. For this research, 3 wood samples and 3 other feedstock are used and they are near infrared (NIR) spectrum data with high-dimensionality. Ash and char are used as response variables. The comparison results of two responses will be shown.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.10a
• /
• pp.368-373
• /
• 2003

A major concern at present is the simultaneous control of transverse thickness profile and flatness in the finishing stages of hot rolling process. The mathematical modeling of hot rolling process has long been recognized to be a desirable approach to investigate rolling operating practice and the design of mill equipment to improve productivity and quality. However, many factors make the mathematical analysis of the rolling process very complex and time-consuming. In order to overcome these problems and to obtain an accurate rolling force, the predicted model of rolling force using neural networks has widely been employed. In this paper, Radial Basis Function Network(RBFN) is applied to improve the accuracy of rolling force prediction in hot rolling mill. In order to verify and analysis the performance of applied neural network, the comparison with the measured rolling force and the predicted results using two different neural networks - RBFN, MLP, has respectively been carried out. The results obtained using RBFN neural network are much more accurate those obtained the MLP.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.13 no.6
• /
• pp.29-33
• /
• 2004

A major concern at present is the simultaneous control of transverse thickness profile and flatness in the finishing stages of hot rolling process. The mathematical modeling of hot rolling process has long been recognized to be a desirable approach to investigate rolling operating practice and the design of mill equipment to improve productivity and quality. However, many factors make the mathematical analysis of the rolling process very complex and time-consuming. In order to overcome these problems and to obtain an accurate rolling force, the predicted model of rolling force using neural networks has widely been employed. In this paper, Radial Basis Function Network(RBFN) is applied to improve the accuracy of rolling force prediction in hot rolling mill. In order to verify and analyze the performance of applied neural network the comparison with the measured rolling force and the predicted results using two different neural networks-RBFN, MLP, has respectively been carried out. The results obtained using RBFN neural network are much more accurate those obtained the MLP.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2004.07b
• /
• pp.1129-1133
• /
• 2004

반도체공정 최적화에 소요되는 시간과 경비를 줄이기 위해 신경망 모델이 개발되고 있다. 주로 역전파 신경망을 이용하여 모델이 개발되고 있으며, 본 연구에서는 Radial Basis Function Network (RBFN)을 이용하여 플라즈마 식각공정 모델을 개발한다. 실험데이터는 유도결합형 플라즈마를 이용한 Silicon Carbide 박막의 식각공정으로부터 수집되었다. 모델개발을 위해 $2^4$ 전인자 (full factorial) 실험계획법이 적용되었으며, 모델에 이용된 식각응답은 식각률과 atomic force microscopy로 측정한 식각표면 거칠기이다. 모델검증을 위해 추가적으로 16번의 실험을 수행하였다. RBFN의 예측성능은 세 학습인자, 즉 뉴런수, width, 초기 웨이트 분포 (initial weight distribution-IWD) 크기에 의해 결정된다. 본 연구에서는 각 학습인자의 영향을 최적화하였으며, IWD의 불규칙성을 고려하여 주어진 학습인자에 대해서 100개의 모델을 발생하고, 이중 최소의 IWD를 갖는 모델을 선택하였다. 최적화한 식각률과 표면거칠기 모델의 RMSE는 각기 26 nm/min과 0.103 nm이었다. 통계적인 회귀모델과 비교하여, 식각률과 표면거칠기 모델은 각기 52%와 24%의 향상된 예측정확도를 보였다. 이로써 RBFN이 플라즈마 공정을 효과적으로 모델링 할 수 있음을 확인하였다.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.1
• /
• pp.38-44
• /
• 2001

This paper investigates the direct adaptive control of nonlinear systems using RBFN(radial basis function networks). The structure of the controller consists of a fixed PD controller and a RBFN controller in parallel. An adaptation law for the parameters of RBFN is developed based on the Lyapunov stability theory to guarantee the stability of the overall control system. The filtered tracking error between the system output and the desired output is shown to be UUB(uniformly ultimately bounded). To evaluate the performance of the controller, the proposed method is applied to the trajectory contro of the two-link manipulator.

• Journal of Korea Water Resources Association
• /
• v.33 no.4
• /
• pp.505-515
• /
• 2000

THe purpose of this study is to forecast of runoff hydrographs according to rainfall event in a stream. The neural network theory as a hydrologic blackbox model is used to solve hydrological problems. The Back-Propagation(BP) algorithm by the Levenberg-Marquardt(LM) techniques and Radial Basis Function(RBF) network in Neural Network(NN) models are used. Runoff hydrograph is forecasted in Bocheongstream basin which is a IHP the representative basin. The possibility of a simulation for runoff hydrographs about unlearned stations is considered. The results show that NN models are performed to effective learning for rainfall-runoff process of hydrologic system which involves a complexity and nonliner relationships. The RBF networks consist of 2 learning steps. The first step is an unsupervised learning in hidden layer and the next step is a supervised learning in output layer. Therefore, the RBF networks could provide rather time saved in the learning step than the BP algorithm. The peak discharge both BP algorithm and RBF network model in the estimation of an unlearned are a is trended to observed values.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.175-176
• /
• 2007

Radial Basis Function Network (RBFN)을 이용하여 플라즈마 전위의 예측 모델을 개발하였다. RBFN의 예측성능은 Genetic Algorithm (GA)를 이용하여 최적화 하였다. 체계적인 모델링을 위해 통계적인 실험계획법이 적용되었으며, 실험은 반구형 유도 결합형 플라즈마 장비를 이용하여 수행이 되었다. $Cl_2$ 플라즈마에서의 데이터 측정에는 Langmuir probe가 이용되었다. 최적화된 GA-RBFN 모델을 일반 RBFN 모델과 비교하였으며, 15%정도 모델의 예측성능을 향상시켰다.