• IE interfaces
• /
• v.16 no.spc
• /
• pp.28-32
• /
• 2003

In this paper, the fuzzy group method data handling-type(GMDH) neural networks and their application to the forecasting of mobile communication system are described. At present, GMDH family of modeling algorithms discovers the structure of empirical models and it gives only the way to get the most accurate identification and demand forecasts in case of noised and short input sampling. In distinction to neural networks, the results are explicit mathematical models, obtained in a relative short time. In this paper, an adaptive learning network is proposed as a kind of neuro-fuzzy GMDH. The proposed method can be reinterpreted as a multi-stage fuzzy decision rule which is called as the neuro-fuzzy GMDH. The GMDH-type neural networks have several advantages compared with conventional multi-layered GMDH models. Therefore, many types of nonlinear systems can be automatically modeled by using the neuro-fuzzy GMDH. The computer program is developed and successful applications are shown in the field of estimating problem of mobile communication with the number of factors considered.

• Computers and Concrete
• /
• v.15 no.4
• /
• pp.547-562
• /
• 2015

A two-dimensional multi-layered finite elements modeling of reinforced concrete structures at non-linear behaviour under monotonic and cyclical loading is presented. The non-linearity material is characterized by several phenomena such as: the physical non-linearity of the concrete and steels materials, the behaviour of cracked concrete and the interaction effect between materials represented by the post-cracking filled. These parameters are taken into consideration in this paper to examine the response of the reinforced concrete structures at the non-linear behaviour. Four examples of application are presented. The numerical results obtained, are in a very good agreement with available experimental data and other numerical models of the literature.

• Journal of Korean Vacuum Science & Technology
• /
• v.2 no.1
• /
• pp.43-48
• /
• 1998

The effect of mcirostructure on the electrical properties of yttria stabilized zirconia (YSZ) was analyzed by modeling layer arrangements and mixed phase structure. The YSZ thin films were deposited by RF magnetron sputtering using 30mol% YSZ and 8 mol% YSZ targets with yttrium pellets on porous alumina substrates. The structure, composition and electricla properties of the YSZ films were investigated as functions of sputtering conditons and layer arrangements by XRD, TEM, XPS and acimpedance spectroscopy. The results showed that the triple palyered YSZ films had highermicrohardness, lower compressive stress state and higher ionic conductivity by one order than single and double layered YSZ films. However, sputtered YSZ films have low conductivity compared to YSZ pellets or doctor bladed YSZ thin plates. These results were probably due to the influence of insulating alumina substrates, impractical for most stacking geometries and inductance induced by relatively long platinum, lead wire on YSZ conductivity.

• Journal of Intelligence and Information Systems
• /
• v.1 no.1
• /
• pp.135-148
• /
• 1995

Neural Networks have been advocated as an alternative to statistical forecasting methods. However, the empirical evidences are not consistent. In the present experiments, multi-layered perceptron (MLP) are adopted as approximator to the time series generating processes. To prevent the MLP from being overfitted to the given time series, the information obtained from ARMA modeling is used to determine the architecture of MLP. The proposed approach was tested empirically using the subsamples of the 111 time series used in the first Markridakis Competition. The forecasting results were analyzed to find out the factors that affect the performance of MLP. The experimental results show that the proposed approach outperforms ARMA models in terms of fitting and forecasting accuracy. In addition, it is found that the use of deseasonalized data improves the forecasting accuracy of MLP.

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.224-226
• /
• 1993

Multi-layered neural networks offer an exciting alternative for modelling complex non-liner systems. This paper investigates the identification of continuous time nonliner system using neural networks with a single hidden layer. The digital low - pass filter are introduced to avoid direct approximation of system derivatives from sampled data. Using a pre-designed digital low pass filter, an approximated discrete-time estimation model is constructed easily. A continuous approximation liner model is first estimated from sampled input-out signals. Then the modeling error due to the nonlinearity is decreased by a compensator using neural network. Simulation results are given to demonstrate the effective of the proposed method.

• The Journal of Engineering Geology
• /
• v.6 no.3
• /
• pp.119-130
• /
• 1996

The theory and numerical technique using boundary elements method (BEM) are given to solve 2-dimensional resistivity problems. Potential distributions from homogeneous resistivity model and layered model are calculated by using BEM for a point source of current injection. The potential distributions of BEM are compared with those of finite difference method (FDM) and finite elements method (FEM). Among the three numerical methods to solve 2-dimensional resistivity problem, it is proved that BEM is more efficient tool than FDM and FEM in consideration of computing storage and time as weU as the accuracy of solutions.

• Journal of Korean Institute of Industrial Engineers
• /
• v.26 no.4
• /
• pp.296-305
• /
• 2000

Virtual wavelength path (VWP) is the optical path when a wavelength conversion is possible in a wavelength division multiplexing (WDM) network that is transmission infrastructure for the next generation high speed backbone networks. To achieve efficient design for VWP networks, we must consider VWP routing, wavelength assignment, and wavelength conversion while satisfying many technical constraints of the WDM networks. In this study we propose an integrated model for efficient VWP design in WDM networks. We also develope a 3-phase algorithm, each of which deals with routing, wavelength assignment and route and wavelength reassignment, respectively. In our computational experiments, phase 1 algorithm can solve the problem to the optimality for medium size test networks. Phase 2 algorithm is an efficient heuristic based on a reduced layered network and can give us an effective wavelength assignment. Finally, phase 3 algorithm reconfigure VWP routing and its wavelength assignment to concentrate wavelength conversion nodes in the suggested VWP network.

• Asia pacific journal of information systems
• /
• v.1 no.1
• /
• pp.63-86
• /
• 1991

In this paper, our concern is the artificial neural network-based patten classification, when can resolve the difficulties in the Autoregressive Moving Average(ARMA) model identification problem To effectively classify a time series into an approriate ARMA model, we adopt the Multi-layered Backpropagation Network (MLBPN) as a pattern classifier, and Extended Sample Autocorrelation Function (ESACF) as a feature extractor. To improve the classification power of MLBPN's we suggest an integrated neural network system which consists of an AR Network and many small-sized MA Networks. The output of AR Network which will gives the MA order. A step-by-step training strategy is also suggested so that the learned MLBPN's can effectively ESACF patterns contaminated by the high level of noises. The experiment with the artificially generated test data and real world data showed the promising results. Our approach, combined with a statistical parameter estimation method, will provide a way to the automation of ARMA modeling.

• Textile Coloration and Finishing
• /
• v.31 no.3
• /
• pp.214-224
• /
• 2019

This investigation reported the simulation technologies to design the 3-dimensional fabrics such as 3 dimensional multi-layered fabric, 3 dimensional braided fabric and spacer fabric. The simulation system or software has been actively used to develop products of 3 dimensional fabric which can be reduced development costs and time. Thus, many countries such as Japan, Germany, China, and U.K. show great interests on simulation technologies for developing new materials and processes including 3 dimensional fabric field. In this study, simulation systems have been reviewed for the 3 dimensional fabric design system from Mikawa Textile Research Center, Japan; ProCad and ProFab from Karl Mayer and Texion, Germany; xComposites from China; TexGen from Nottingham University, U.K.; TexPro from Young Woo CnI, Korea, respectively.

• ETRI Journal
• /
• v.43 no.2
• /
• pp.288-298
• /
• 2021

Abstractive text summarization is a process of making a summary of a given text by paraphrasing the facts of the text while keeping the meaning intact. The manmade summary generation process is laborious and time-consuming. We present here a summary generation model that is based on multilayered attentional peephole convolutional long short-term memory (MAPCoL; LSTM) in order to extract abstractive summaries of large text in an automated manner. We added the concept of attention in a peephole convolutional LSTM to improve the overall quality of a summary by giving weights to important parts of the source text during training. We evaluated the performance with regard to semantic coherence of our MAPCoL model over a popular dataset named CNN/Daily Mail, and found that MAPCoL outperformed other traditional LSTM-based models. We found improvements in the performance of MAPCoL in different internal settings when compared to state-of-the-art models of abstractive text summarization.