• Structural Engineering and Mechanics
• /
• v.54 no.5
• /
• pp.891-908
• /
• 2015

For a systematic study on wind-induced vibration characteristics of large hyperbolic cooling towers with different feature sizes, the pressure measurement tests are finished on the rigid body models of three representative cooling towers with the height of 155 m, 177 m and 215 m respectively. Combining the refined frequency-domain algorithm of wind-induced responses, the wind-induced average response, resonant response, background response, coupling response and wind vibration coefficients of large cooling towers with different feature sizes are obtained. Based on the calculating results, the parametric analysis on wind-induced vibration of cooling towers is carried out, e.g. the feature sizes, damping ratio and the interference effect of surrounding buildings. The discussion shows that the increase of feature sizes makes wind-induced average response and fluctuating response larger correspondingly, and the proportion of resonant response also gradually increased, but it has little effect on the wind vibration coefficient. The increase of damping ratio makes resonant response and the wind vibration coefficient decreases obviously, which brings about no effect on average response and background response. The interference effect of surrounding buildings makes the fluctuating response and wind vibration coefficient increased significantly, furthermore, the increase ranges of resonant response is greater than background response.

• Structural Engineering and Mechanics
• /
• v.41 no.1
• /
• pp.113-137
• /
• 2012

Frame structures with viscoelastic (VE) dampers mounted on them are considered in this paper. It is the aim of this paper to compare the dynamic characteristics of frame structures with VE dampers when the dampers are modelled by means of different models. The classical rheological models, the model with the fractional order derivative, and the complex modulus model are used. A relatively large structure with VE dampers is considered in order to make the results of comparison more representative. The formulae for dissipation energy are derived. The finite element method is used to derive the equations of motion of the structure with dampers and such equations are written in terms of both physical and state-space variables. The solution to motion equations in the frequency domain is given and the dynamic properties of the structure with VE dampers are determined as a solution to the appropriately defined eigenvalue problem. Several conclusions concerning the applicability of a family of models of VE dampers are formulated on the basis of results of an extensive numerical analysis.

• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2001.05a
• /
• pp.40-45
• /
• 2001

Recently, the use of numerical simulation has been increased rapidly because of the development of high performance computer systems. The present study is aimed to investigate flow characteristics of a two-dimensional sharp plane. Unsteady calculation by FDM(Finite Difference Method) based upon SOLA scheme which was performed at $Re=2{\times}10^4$in viscous incompressible flow within a finite domain on the irregular grid formation. Total numbers of irregular grids are $8{\times}10^4$. The minimum grid size is 1/100 of the plane length L which is the representative length. The inclined angles of every objects are $15^{\circ}, \;30^{\circ}\;and\; 45^{\circ}.$ And, the edge angle of the plane is $30^{\circ}.$ This study discussed the flow characteristics in term of the turbulent intensity, vorticity and frequency analysis. Developed flows show that the periodic Karman vortices occur at the back of the plane.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.7
• /
• pp.247-256
• /
• 2008

This research is to develop product retrieval system through simplified communication by applying intelligent agent technology based on automatically created domain ontology to present solution on problems with e-commerce system which searches in the web documents with a simple keyword. Ontology development extracts representative term based on classification information of international product classification code(UNSPSC) and jewelry websites that is applied to analogy relationship thesaurus to establish standardized ontology. The intelligent agent technology is applied to retrieval stage to support efficiency of information collection for users by designing and developing e-commerce system supported with semantic web. Moreover, it designs user profile to personalized search environment and provide personalized retrieval agent and retrieval environment with inference function to make available with fast information collection and accurate information search.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.4
• /
• pp.310-317
• /
• 2013

In this paper an ESPRIT-based super-resolution time delay estimation algorithm is developed for real-time locating system (RTLS) and its performance is analyzed in various multipath environments. The performance of the existing correlation method for time delay estimation seriously degrades in multipath environments where the relative time delays of multipath signals are less than a PN chip. To solve the problem we shall develop a frequency domain super-resolution time delay estimation algorithm using the ESPRIT, the most representative super-resolution direction-of-arrival (DOA) estimation algorithm, and analyze its performance in various multipath environments.

• Earthquakes and Structures
• /
• v.9 no.2
• /
• pp.321-337
• /
• 2015

Existing building structures can easily present material mechanical properties which can largely vary even within a single structure. The current European Technical Code, Eurocode 8, does not provide specific instructions to account for high variability in mechanical properties. As a consequence of the high strength variability, at the occurrence of seismic events, the structure may evidence unexpected phenomena, like torsional effects, with larger experienced deformations and, in turn, with reduced seismic performance. This work is focused on the reduction in seismic performance due to the concrete strength variability. The analysis has been performed on a case-study, i.e., a 3D RC framed 4 storey building. A Normal distribution, compatible to a large available database, has been taken to represent the concrete strength domain. Different plan layouts, representative of realistic strength distributions, have been considered, and a statistical analysis has been performed on the induced reduction in seismic performance. The obtained results have been compared to the standard analysis as provided by Eurocode 8 for existing buildings. The comparison has shown that the Eurocode 8 provisions are not conservative for existing buildings having a large variability in concrete strength.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.420-429
• /
• 2019

Recently, researches using Convolutional Neural Network (CNN)-based approaches have been actively conducted to improve the reduced quality of compressed video using block-based video coding standards such as H.265/HEVC. This paper aims to summarize and analyze the network models in these quality enhancement studies. At first the detailed components of CNN for quality enhancement are overviewed and then we summarize prior studies in the image domain. Next, related studies are summarized in three aspects of network structure, dataset, and training methods, and present representative models implementation and experimental results for performance comparison.

• Clinical and Experimental Pediatrics
• /
• v.65 no.5
• /
• pp.239-249
• /
• 2022

Cells survive and proliferate through complex interactions among diverse molecules across multiomics layers. Conventional experimental approaches for identifying these interactions have built a firm foundation for molecular biology, but their scalability is gradually becoming inadequate compared to the rapid accumulation of multiomics data measured by high-throughput technologies. Therefore, the need for data-driven computational modeling of interactions within cells has been highlighted in recent years. The complexity of multiomics interactions is primarily due to their nonlinearity. That is, their accurate modeling requires intricate conditional dependencies, synergies, or antagonisms between considered genes or proteins, which retard experimental validations. Artificial intelligence (AI) technologies, including deep learning models, are optimal choices for handling complex nonlinear relationships between features that are scalable and produce large amounts of data. Thus, they have great potential for modeling multiomics interactions. Although there exist many AI-driven models for computational biology applications, relatively few explicitly incorporate the prior knowledge within model architectures or training procedures. Such guidance of models by domain knowledge will greatly reduce the amount of data needed to train models and constrain their vast expressive powers to focus on the biologically relevant space. Therefore, it can enhance a model's interpretability, reduce spurious interactions, and prove its validity and utility. Thus, to facilitate further development of knowledge-guided AI technologies for the modeling of multiomics interactions, here we review representative bioinformatics applications of deep learning models for multiomics interactions developed to date by categorizing them by guidance mode.

• Wind and Structures
• /
• v.35 no.2
• /
• pp.131-146
• /
• 2022

Tornadic wind flow is inherently turbulent. A turbulent wind flow is characterized by fluctuation of the velocity in the flow field with time, and it is a dynamic process that consists of eddy formation, eddy transportation, and eddy dissipation due to viscosity. Properly modeling turbulence significantly increases the accuracy of numerical simulations. The lack of a clear and detailed comparison between turbulence models used in tornadic wind flows and their effects on tornado induced pressure demonstrates a significant research gap. To bridge this research gap, in this study, two representative turbulence modeling approaches are applied in simulating real-world tornadoes to investigate how the selection of turbulence models affects the simulated tornadic wind flow and the induced pressure on structural surface. To be specific, LES with Smagorinsky-Lilly Subgrid and k-ω are chosen to simulate the 3D full-scale tornado and the tornado-structure interaction with a building present in the computational domain. To investigate the influence of turbulence modeling, comparisons are made of velocity field and pressure field of the simulated wind field and of the pressure distribution on building surface between the cases with different turbulence modeling.

• Journal of Korea Multimedia Society
• /
• v.24 no.11
• /
• pp.1552-1559
• /
• 2021

Convolution neural networks (CNNs) show notable performance in image processing and are used as representative core models. CNNs extract and learn features from large amounts of train dataset. In general, it has a structure in which a convolution layer and a fully connected layer are stacked. The core of CNN is the convolution layer. The size of the kernel used for feature extraction and the number that affect the depth of the feature map determine the amount of weight parameters of the CNN that can be learned. These parameters are the main causes of increasing the computational complexity and memory usage of the entire neural network. The most computationally expensive components in CNNs are fully connected and spatial convolution computations. In this paper, we propose a Fourier Convolution Neural Network that performs the operation of the convolution layer in the Fourier domain. We work on modifying and improving the amount of computation by applying the fast fourier transform method. Using the MNIST dataset, the performance was similar to that of the general CNN in terms of accuracy. In terms of operation speed, 7.2% faster operation speed was achieved. An average of 19% faster speed was achieved in experiments using 1024x1024 images and various sizes of kernels.