• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.11
• /
• pp.56-65
• /
• 1993

This paper presents the design consideration of the MFNNs(Multilayer Feed forward Neural Networks) based on the distribution of the given teching patterns. By extracting the feature points from the given teaching patterns, the structure of a network including the netowrk size and interconnection weights of a network is initialized. This network is trained based on the modified version of the EBP(Error Back Propagation) algorithm. As a result, the proposed method has the advantage of learning speed compared to the conventional learning of the MFNNs with randomly chosen initial weights. To show the effectiveness of the suggested approach, the simulation result on the approximation of a two demensional continuous function is shown.

• KIEE International Transaction on Systems and Control
• /
• v.2D no.2
• /
• pp.53-58
• /
• 2002

In this paper, various aspects in magnetoencephalography (MEG) source localization are studied. To minimize the errors in experimental data, an approximation technique using a polynomial function is proposed. The simulation shows that the proposed technique yields more accurate results. To improve the convergence characteristics in the optimization algorithm, a hybrid algorithm of evolution strategy and sensitivity analysis is applied to the neuromagnetic inverse problem. The effectiveness of the hybrid algorithm is verified by comparison with conventional algorithms. In addition, an artificial neural network (ANN) is applied to find an initial source location quickly and accurately. The simulation indicates that the proposed technique yields more accurate results effectively.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• 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.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.18 no.1
• /
• pp.22-28
• /
• 2009

In order to optimize the polishing process, Run to Run control scheme has been applied to the micro mold polishing in this study. Also, to fully understand the effect of parameters on the surface roughness a design of experiment is performed. By linear approximation of main factors such as gap and rotational speed of micro quill, EWMA (Exponential Weighted Moving Average) gradual mode controller is adopted as a optimizing tool. Consequently, the process converged quickly at a target value of surface roughness Ra 10nm and Rmax 50nm, and was hardly affected by unwanted process noises like initial surface quality and wear of magnetic abrasives.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.24 no.62
• /
• pp.11-20
• /
• 2001

At the initial deployment of B-ISDN, the heterogeneity of services and the bandwidth requirement make dimensioning of B-ISDN much more complicated than the dimensioning of homogeneous circuit or packet switched networks. Therefore B-ISDN must be extended gradually according to future telecommunication technology or service demands. In this paper, we propose a mathematical formulation for dimensioning problem of B-ISDN extension, considering the characteristics of ATM(Asynchronous Transfer Mode) such as various quality of services, the statistical multiplexing effects of VPCs(Virtual Path Connections) and the modularity of transmission links allocated when new B-ISDN nodes are given. The algorithm based on the simultaneous linear approximation technique and Lagrangian relaxation method and some numerical results are also presented.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2911-2913
• /
• 1999

This paper suggest an optimal identification method to complex and nonlinear system modeling that is based on Fuzzy-Neural Network(FNN). The FNN modeling implements parameter identification using HCM algorithm and optimal identification algorithm structure combined with two types of optimization theories for nonlinear systems, we use a HCM Clustering Algorithm to find initial parameters of membership function. The parameters such as parameters of membership functions, learning rates and momentum coefficients are adjusted using optimal identification algorithm. The proposed optimal identification algorithm is carried out using both a genetic algorithm and the improved complex method. Also, an aggregate objective function(performance index) with weighted value is proposed 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, the data of sewage treatment process and traffic route choice process.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.9
• /
• pp.127-134
• /
• 2004

It is difficult to have the existing structural optimization using coupled field analysis from CFD to structure analysis when the structure is influenced of fluid. Therefore in an initial model of this study after doing parameter design from the background of shape using topology optimization. and it is making a approximation formula using by the CFD-structure coupled-field analysis and design of experiment. By using this result, we conducted multi-objective optimization. We could confirm efficiency of stochastic method applicable in the scene of structure reliability design to be needed multi-objective optimization. And we presented a way of design that could overcome the time and space restriction in structural design such as the butterfly valve with the less experiment.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.23 no.3
• /
• pp.253-266
• /
• 2019

In this article, a systematic solution based on the sequence of expansion method is planned to solve the time-fractional diffusion equation, time-fractional telegraphic equation and time-fractional wave equation in three dimensions using a current and valid approximate method, namely the ADM, VIM, and the NIM subject to the estimate initial condition. By using these three methods it is likely to find the exact solutions or a nearby approximate solution of fractional partial differential equations. The exactness, efficiency, and convergence of the method are demonstrated through the three numerical examples.

• Bulletin of the Korean Chemical Society
• /
• v.7 no.3
• /
• pp.224-228
• /
• 1986

A nonlinear theory is presented for the fluctuations of intermediates in the Brusselator near the critical point caused by diffusion. The method used is the two time scaling method different from the conventional method in the sense that a slight nonlinear effect is included in the initial time region where the linear approximation is conventionally valid. The result obtained by the nonlinear theory shows that fluctuations close to the critical point approach the value of a stable steady state or deviate infinitely from an unstable steady state, as time goes to infinity, while the linear theory gives approximately time-independent fluctuations. A brief discussion is given for the correlation at a time between fluctuating intermediates when the system approaches a stable steady state.

• Honam Mathematical Journal
• /
• v.43 no.1
• /
• pp.130-140
• /
• 2021

In this paper, we introduce and study the concept of trigonometrically quasi-convex function. We prove Hermite-Hadamard type inequalities for the newly introduced class of functions and obtain some new Hermite-Hadamard inequalities for functions whose first derivative in absolute value, raised to a certain power which is greater than one, respectively at least one, is trigonometrically quasi-convex convex. We also extend our initial results to functions of several variables. Next, we point out some applications of our results to give estimates for the approximation error of the integral the function in the trapezoidal formula.