• 한국시뮬레이션학회논문지
• /
• 제14권4호
• /
• pp.55-68
• /
• 2005

The advancement in semiconductor technology is leading toward smaller critical dimension designs and larger wafer manufactures. Due to such phenomena, semiconductor industry is in need of an accurate control of the process. Photolithography is one of the key processes where the pattern of each layer is formed. In this process, precise superposition of the current layer to the previous layer is critical. Therefore overlay parameters of the semiconductor photolithography process is targeted for this research. The complex relationship among the input parameters and the output metrologies is difficult to understand and harder yet to model. Because of the superiority in modeling multi-nonlinear relationships, neural networks is used for the simulator modeling. For training the neural networks, conjugate gradient method is employed. An experiment is performed to evaluate the performance among the proposed neural network simulator, stepwise regression model, and the currently practiced prediction model from the test site.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.372-372
• /
• 2000

In this paper, we propose Neuro Fuzzy Polynomial Networks(NFPN) based on Polynomial Neural Network(PNN) and Neuro-Fuzzy(NF) for model identification of complex and nonlinear systems. The proposed NFPN is generated from the mutually combined structure of both NF and PNN. The one and the other are considered as the premise part and consequence part of NFPN structure respectively. As the premise part of NFPN, NF uses both the simplified fuzzy inference as fuzzy inference method and error back-propagation algorithm as learning rule. The parameters such as parameters of membership functions, learning rates and momentum coefficients are adjusted using genetic algorithms. As the consequence part of NFPN, PNN is based on Group Method of Data Handling(GMDH) method and its structure is similar to Neural Networks. But the structure of PNN is not fixed like in conventional Neural Networks and self-organizing networks that can be generated. NFPN is available effectively for multi-input variables and high-order polynomial according to the combination of NF with PNN. Accordingly it is possible to consider the nonlinearity characteristics of process and to get better output performance with superb predictive ability. In order to evaluate the performance of proposed models, we use the nonlinear function. The results show that the proposed FPNN can produce the model with higher accuracy and more robustness than any other method presented previously.

• 한국정보통신학회논문지
• /
• 제10권6호
• /
• pp.1032-1037
• /
• 2006

채널은 무선망에 있어서 한정된 주요 자원 중의 하나이다. 다양한 채널 관리 기법들이 제시되어 왔으며, 최근 들어 가드채널의 최적화 문제가 부각되고 있다. 본 논문에서는 신경망을 이용한 지능적인 채널 관리 기법을 제안한다. 신경망의 학습 데이터 생성과 성능분석을 위하여 SRN(Stochastic Reward Net) 채널 할당 모델이 개발된다. 제안된 기법에서 신경망은 지도학습 방법인 역전파 알고리즘을 이용하여 최적의 가드채널 값 g를 계산하도록 학습한다. 학습된 신경망을 이용하여 최적의 g를 계산하고, 이를 SRM모델에서 구해진 결과와 비교한다. 실험 결과는 신경망에서 구한 가드채널 수와 SRM모델로부터 구한 가드채널 수의 상대적 차이가 없음을 보여준다.

• Computers and Concrete
• /
• 제25권5호
• /
• pp.401-409
• /
• 2020

Artificial neural networks are used as a useful tool in distinct fields of civil engineering these days. In order to control long-term quality of Self-Compacting Semi-Lightweight Concrete (SCSLC), the 90 days compressive strength is considered as a key issue in this paper. In fact, combined artificial neural networks are used to predict the compressive strength of SCSLC at 28 and 90 days. These networks are able to re-establish non-linear and complex relationships straightforwardly. In this study, two types of neural networks, including Radial Basis and Multilayer Perceptron, were used. Four groups of concrete mix designs also were made with two water to cement ratios (W/C) of 0.35 and 0.4, as well as 10% of cement weight was replaced with silica fume in half of the mixes, and different amounts of superplasticizer were used. With the help of rheology test and compressive strength results at 7 and 14 days as inputs, the neural networks were used to estimate the 28 and 90 days compressive strengths of above-mentioned mixes. It was necessary to add the 14 days compressive strength in the input layer to gain acceptable results for 90 days compressive strength. Then proper neural networks were prepared for each mix, following which four existing networks were combined, and the combinatorial neural network model properly predicted the compressive strength of different mix designs.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2949-2952
• /
• 2000

A new neural networks and learning algorithm are proposed in order to measure nonlinear heights of complexed road environments in realtime without pre-information. This new neural networks is Error Self Recurrent Neural Networks(ESRN), The structure of it is similar to recurrent neural networks: a delayed output as the input and a delayed error between the output of plant and neural networks as a bias input. In addition, we compute the desired value of hidden layer by an optimal method instead of transfering desired values by back-propagation and each weights are updated by RLS(Recursive Least Square). Consequently. this neural networks are not sensitive to initial weights and a learning rate, and have a faster convergence rate than conventional neural networks. We can estimate nonlinear models in realtime by ESRN and learning algorithm and control nonlinear models. To show the performance of this one. we control 7 degree of freedom full car model with several control method. From this simulation. this estimation and controller were proved to be effective to the measurements of nonlinear road environment systems.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제1권1호
• /
• pp.29-34
• /
• 2001

This paper is devoted to investigating direct adaptive neural control of nonlinear systems with uncertain or unknown dynamic models. In the direct adaptive neural networks control area, theoretical issues of the existing backpropagation-based adaptive neural networks control schemes. The major contribution is proposing the variable index control approach, which is of great significance in the control field, and applying it to derive new stable robust adaptive neural network control schemes. This new schemes possess inherent robustness to system model uncertainty, which is not required to satisfy any matching condition. To demonstrate the feasibility of the proposed leaning algorithms and direct adaptive neural networks control schemes, intensive computer simulations were conducted based on the flexible joint robot systems and functions.

• 제어로봇시스템학회논문지
• /
• 제8권2호
• /
• pp.126-135
• /
• 2002

The study is concerned with an approach to the design of new architectures of fuzzy neural networks and the discussion of comprehensive design methodology supporting their development. We propose an Adaptive Fuzzy Polynomial Neural Networks(APFNN) based on Fuzzy Neural Networks(FNN) and Self-organizing Networks(SON) for model identification of complex and nonlinear systems. The proposed AFPNN is generated from the mutually combined structure of both FNN and SON. The one and the other are considered as the premise and the consequence part of AFPNN, respectively. As the premise structure of AFPNN, FNN uses both the simplified fuzzy inference and error back-propagation teaming rule. The parameters of FNN are refined(optimized) using genetic algorithms(GAs). As the consequence structure of AFPNN, SON is realized by a polynomial type of mapping(linear, quadratic and modified quadratic) between input and output variables. In this study, we introduce two kinds of AFPNN architectures, namely the basic and the modified one. The basic and the modified architectures depend on the number of input variables and the order of polynomial in each layer of consequence structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the AFPNN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed AFPNN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• 전기학회논문지
• /
• 제60권3호
• /
• pp.639-647
• /
• 2011

In this paper, we introduce a design methodology of data-centroid Radial Basis Function neural networks with extended polynomial function. The two underlying design mechanisms of such networks involve K-means clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on K-means clustering method for efficient processing of data and the optimization of model was carried out using PSO. In this paper, as the connection weight of RBF neural networks, we are able to use four types of polynomials such as simplified, linear, quadratic, and modified quadratic. Using K-means clustering, the center values of Gaussian function as activation function are selected. And the PSO-based RBF neural networks results in a structurally optimized structure and comes with a higher level of flexibility than the one encountered in the conventional RBF neural networks. The PSO-based design procedure being applied at each node of RBF neural networks leads to the selection of preferred parameters with specific local characteristics (such as the number of input variables, a specific set of input variables, and the distribution constant value in activation function) available within the RBF neural networks. To evaluate the performance of the proposed data-centroid RBF neural network with extended polynomial function, the model is experimented with using the nonlinear process data(2-Dimensional synthetic data and Mackey-Glass time series process data) and the Machine Learning dataset(NOx emission process data in gas turbine plant, Automobile Miles per Gallon(MPG) data, and Boston housing data). For the characteristic analysis of the given entire dataset with non-linearity as well as the efficient construction and evaluation of the dynamic network model, the partition of the given entire dataset distinguishes between two cases of Division I(training dataset and testing dataset) and Division II(training dataset, validation dataset, and testing dataset). A comparative analysis shows that the proposed RBF neural networks produces model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• 한국국방경영분석학회지
• /
• 제34권1호
• /
• pp.1-11
• /
• 2008

본 연구에서는 다양하고 복잡한 변수간의 비선형적인 관계를 분석할 수 있는 신경망의 특성을 이용하여 곡사화기를 사용하는 포병의 모의훈련체계를 향상시킬 수 있는 방안을 제시하였다. 신경망 모델은 Conjugate Gradient 학습알고리즘을 사용하였고, 모델의 신뢰성은 모의실험을 통해 수학적 회귀분석모델과 신경망 모델의 예측오차를 비교하여 입증하였다. 신경망모델을 곡사화기 모의훈련체계 개선에 활용한다면, 보다 실전적인 모의훈련을 가능하게 하여 전투력 향상 및 예산절감에도 크게 기여할 것이다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권4호
• /
• pp.264-269
• /
• 2008

In this paper, two neural networks as a material model, which are based on the state-space method, have been proposed. One outputs the rates of inelastic strain and material internal variables whereas the outputs of the other are the next state of the inelastic strain and material internal variables. Both the neural networks were trained using input-output data generated from Chaboche's model and successfully converged. The former neural network could reproduce the original stress-strain curve. The neural network also demonstrated its ability of interpolation by generating untrained curve. It was also found that the neural network can extrapolate in close proximity to the training data.