• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.553-558
• /
• 2003

The boundary estimation problem is used to estimate the shape of organic depend on the phase of the cardiac cycle or interested in the detection of the location and size of anomalies with resistivity values different from the background tissues such as nuclear reactor. And we can use the method to solve the optimal solution such as modified Newton raphson, kalman filter, extended kalman filter, etc. But, this method consumes much time and is sensitive to the initial value and noise in the estimation of the unknown shape. In the paper, we propose that multi-layer neural networks estimate the boundary of the unknown object using Fourier coefficient. This method can be used at the real time estimation and have strong characteristics at the noise and initial value. It uses voltage change; difference the homogeneous voltage to the non-homogeneous voltage, and change of Fourier coefficient change to train multi-layer neural network. After train, we can have real time estimation using this method.

• 한국해양공학회지
• /
• 제31권1호
• /
• pp.1-7
• /
• 2017

There are two wave modes induced by an oscillating body on the free surface of a two-layer fluid: the barotropic and baroclinic modes. To investigate the generated waves composed of two modes, a radiation problem involving a heaving rectangular body was solved in a numerical wave tank. A new artificial damping zone scheme was developed and applied in the frequency-domain analysis. The performance of this damping scheme was compared with given radiation boundary conditions for various conditions. The added mass and radiation damping coefficients for the heaving rectangular body were also calculated for various fluid-density ratios.

• Wind and Structures
• /
• 제5권2_3_4호
• /
• pp.379-392
• /
• 2002

Numerical flow computations around an aeroelastic 3D square cylinder immersed in the turbulent boundary layer are shown. Present computational code can be characterized by three numerical aspects which are 1) the method of artificial compressibility is adopted for the incompressible flow computations, 2) the domain decomposition technique is used to get better grid point distributions, and 3) to achieve the conservation law both in time and space when the flow is computed a with moving and transformed grid, the time derivatives of metrics are evaluated using the time-and-space volume. To provide time-dependant inflow boundary conditions satisfying prescribed time-averaged velocity profiles, a convenient way for generating inflow turbulence is proposed. The square cylinder is modeled as a 4-lumped-mass system and it vibrates with two-degree of freedom of heaving motion. Those blocks which surround the cylinder are deformed according to the cylinder's motion. Vigorous oscillations occur as the vortex shedding frequency approaches cylinder's natural frequencies.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2002년도 학술대회지
• /
• pp.779-782
• /
• 2002

We examine the problem in which porous/viscoelastic compliant thin plates are subject to pressure fluctuations under transitional or turbulent boundary layer. Measurements are presented of the frequency spectra of the near-field pressure and radiated sound by compliant surface. A porous plate consisting of 5mm thick, open-cell foam with fabric covering and a viscoelastic painted plate of 1mm thick over an acoustic board of 4m thick were placed over a rigid surface in an anechoic wind tunnel. Streamwise velocity and wall pressure measurements were shown to highly attenuate the convective wall pressure energy when the convective wavenumber ($k_{ch}$) was 3.0 or more. The sound source localization on the compliant walls is applied to the measurement of radiated sound by using an acoustic mirror system.

• Structural Engineering and Mechanics
• /
• 제12권2호
• /
• pp.215-230
• /
• 2001

In this paper, an asymptotic two-scale method is developed for solving vibration problem of long periodic structures. Such eigenmodes appear as a slow modulations of a periodic one. For those, the present method splits the vibration problem into two small problems at each order. The first one is a periodic problem and is posed on a few basic cells. The second is an amplitude equation to be satisfied by the envelope of the eigenmode. In this way, one can avoid the discretisation of the whole structure. Applying the Floquet method, the boundary conditions of the global problem are determined for any order of the asymptotic expansions.

• Journal of applied mathematics & informatics
• /
• 제33권5_6호
• /
• pp.485-502
• /
• 2015

A computational method for solving singularly perturbed boundary value problem of differential equation with shift arguments of mixed type is presented. When shift arguments are sufficiently small (o(ε)), most of the existing method in the literature used Taylor's expansion to approximate the shift term. This procedure may lead to a bad approximation when the delay argument is of O(ε). The main idea for this work is to deal with constant shift arguments, which are independent of ε. In the present method, we construct the formally asymptotic solution of the problem using the method of composite expansion. The reduced problem is solved numerically by using operator compact implicit method, and the second problem is solved analytically. Error estimate is derived by using the maximum norm. Numerical examples are provided to support the theoretical results and to show the efficiency of the proposed method.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권5호
• /
• pp.277-282
• /
• 2003

In this paper we propose Using Higher Order Neuron on the Supervised Learning Machine of the Kohonen Feature Map. The architecture of proposed model adopts the higher order neuron in the input layer of Kohonen Feature Map as a Supervised Learning Machine. It is able to estimate boundary on input pattern space because or the higher order neuron. However, it suffers from a problem that the number of neuron weight increases because of the higher order neuron in the input layer. In this time, we solved this problem by placing the second order neuron among the higher order neuron. The feature of the higher order neuron can be mapped similar inputs on the Kohonen Feature Map. It also is the network with topological mapping. We have simulated the proposed model in respect of the recognition rate by XOR problem, discrimination of 20 alphabet patterns, Mirror Symmetry problem, and numerical letters Pattern Problem.

• International journal of advanced smart convergence
• /
• 제7권4호
• /
• pp.57-65
• /
• 2018

Artificial neural network is inspired by the biological neural network. For simplicity, in computer science, it is represented as a set of layers. Many research has been made in evaluating the number of neurons in the hidden layer but still, none was accurate. Several methods are used until now which do not provide the exact formula for calculating the number of thehidden layer as well as the number of neurons in each hidden layer. In this paper model selection approach was presented. Proposed model is based on geographical analysis of decision boundary. Proposed model selection method is useful when we know the distribution of the training data set. To evaluate the performance of the proposed method we compare it to the traditional architecture on IRIS classification problem. According to the experimental result on Iris data proposed method is turned out to be a powerful one.

• Asian Journal of Atmospheric Environment
• /
• 제4권3호
• /
• pp.141-149
• /
• 2010

Aerosol size and number concentration were observed in the atmospheric boundary layer over Beijing (from near the ground to 1,200 m) on March 15 (a clear day) and 16 (a dusty day), 2005. The results were further compared with lidar measurements in order to understand the dependency of extinction on the particle size distribution and their vertical changes. The boundary layer atmosphere was composed of several sub-layers, and a dry air layer appeared between 400 and 1,000 m under the influence of dust event. In this dry air layer, the concentration of the fine-mode particles (diameter smaller than $1.0\;{\mu}m$) was slightly lower than the value on the clear day, while the concentration of coarse-mode particles (diameter larger than $1.0\;{\mu}m$) was remarkably higher than that on the clear day. This situation was attributed to the inflow of an air mass containing large amounts of Asian dust particles and a smaller amount of fine-mode particles. The results strongly suggest that the fine-mode particles affect light extinction even in the dusty atmosphere. However, quantitatively the relation between extinction and particle concentration is not satisfied under the dusty atmospheric conditions since laser beam attenuates in the atmosphere with high concentration of particles. Laser beam attenuation effect becomes larger in the relation between extinction and coarse particle content comparing the relation between extinction and fine particle content. To clarify this problem technically, future in situ measurements such as balloon-borne lidar are suggested. Here extinction was measured at 532 nm wavelength. Measurements of extinction at other wavelengths are desired in the future.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.118-121
• /
• 2003

In the conventional boundary estimation in EIT (Electrical Impedance Tomography), the interface between anomalies and background is expressed in usual as Fourier series and the boundary is reconstructed by obtaining the Fourier coefficients. This paper proposes a method for the boundary estimation, where the boundary of anomaly is approximated as the interpolation of front points located discretely along the boundary and is imaged by tracking the points in the region of interest. In the solution to the inverse problem to estimate the front points, the multi-layer neural network is introduced. For the verification of the proposed method, numerical experiments are conducted and the results indicate a good performance.