• Journal of the Mineralogical Society of Korea
• /
• v.27 no.4
• /
• pp.271-281
• /
• 2014

The physicochemical properties of clay minerals have been investigated at the atomistic to nano scale. The microscopic studies are often challenging to perform by using experimental approaches alone. In particular, hydroxyl groups of octahedral sheets in 2:1 clay minerals have been hypothesized to impact the sorption process of metal cations; however, X-ray based techniques alone, a common tool for mineral structure examination, cannot properly test the hypothesis. The current study has examined whether computational mineralogy techniques can be applied to examine the hydroxyl structures of clay minerals. Based on quantum-mechanics and molecular-mechanics computational methods, geometry optimizations were carried out for representative dioctahedral and trioctahedral phyllosilicate minerals. Both methods well reproduced the experimental lattice parameters; however, for structural distortion occurring in the tetrahedral or octahedral sheets, molecular mechanics showed significant deviations from experimental data. The orientation angle of the hydroxyl with respect to (001) basal plane is determined by the balance of repulsion between the hydroxyl proton and Si cations of tetrahedral sites; the quantum-mechanics method predicted $25-26^{\circ}$ for the angle, whereas the angle predicted by the molecular-mechanics method was much higher by $10^{\circ}$ (i.e., $35^{\circ}$). These results demonstrate that computational mineralogy techniques are a reliable tool for clay mineral studies and can be used to further elucidate the roles of hydroxyls in metal sorption process.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.5
• /
• pp.713-723
• /
• 2003

In this study, a state of art review of existing residual stress analysis techniques and representative solutions is presented in order to develope the residual stress analysis procedure for fitness-for-service (FFS) assessment of welded structure. Critical issues associated with existing residual stress solutions and their treatments in performing FFS are discussed. It should be recognized that detailed residual stress evolution is an extremely complicated phenomenon that typically involves material-specific ther-momechanical/metallurgical response, welding process physics, and structural interactions within a component being welded. As a result, computational procedures can vary significantly from highly complicated numerical techniques intended only to elucidate a small part of the process physics to cost-effective procedures that are deemed adequate for capturing some of the important features in a final residual stress distribution. Residual stress analysis procedure for FFS purposes belongs to the latter category. With this in mind, both residual stress analysis techniques and their adequacy for FFS are assessed based on both literature data and analyses performed in this investigation.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.11
• /
• pp.2131-2138
• /
• 2011

GPS is vulnerable to jamming because of extremely low signal power. Many anti-jamming techniques are studied for complement this vulnerability. Anti-jamming techniques using array antenna are most effective technique and these techniques are required the DOA estimates. MUSIC algorithm and Root-MUSIC Algorithm are typical algorithms used in DOA estimation. Two algorithms have different characteristics, so the choice of an algorithm may depends on many factors such as the environment and the system requirements. The analysis and performance comparison of both algorithms is necessary to choose the best method to apply. This paper summarizes the theory of MUSIC and Root-MUSIC algorithms. And this paper extends both algorithm to estimate two-dimensional angles. The software simulator of both algorithms are implemented to evaluate the performance. Root-MUSIC algorithm has the computational advantage on ULA. MUSIC algorithm is applicable to any antenna array. MUSIC shows better estimation performance when number of array element is small while the computational load of MUSIC is much higher than Root-MUSIC.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.5
• /
• pp.314-326
• /
• 2016

The purpose of variable selection techniques is to select a subset of relevant variables for a particular learning algorithm in order to improve the accuracy of prediction model and improve the efficiency of the model. We conduct an empirical analysis to evaluate and compare seven well-known variable selection techniques for multiple linear regression model, which is one of the most commonly used regression model in practice. The variable selection techniques we apply are forward selection, backward elimination, stepwise selection, genetic algorithm (GA), ridge regression, lasso (Least Absolute Shrinkage and Selection Operator) and elastic net. Based on the experiment with 49 regression data sets, it is found that GA resulted in the lowest error rates while lasso most significantly reduces the number of variables. In terms of computational efficiency, forward/backward elimination and lasso requires less time than the other techniques.

• Journal of the Korean Society for information Management
• /
• v.16 no.4
• /
• pp.53-74
• /
• 1999

This paper describes concept mapping techniques for eliciting and representing knowledge. Concept mapping techniques range from very informal to very formal. Informal concept mapping techniques are usually very easy to use and understand for humans, but not for computers. Formal concept mapping techniques are computational, but humans usually find them hard to understand and use. A knowledge acquisition and representation tools which handle both kinds, and the transition from informal to formal, would be very useful. It is proposed that concept maps be regarded as basic components of any knowledge-based system, complementing text and image with formal and informl active diagrams.

• Geophysics and Geophysical Exploration
• /
• v.24 no.2
• /
• pp.53-66
• /
• 2021

Velocity model building is an essential procedure in seismic data processing. Conventional techniques, such as traveltime tomography or velocity analysis take longer computational time to predict a single velocity model and the quality of the inversion results is highly dependent on human expertise. Full-waveform inversions also depend on an accurate initial model. Recently, deep neural network techniques are gaining widespread acceptance due to an increase in their integration to solving complex and nonlinear problems. This study investigated cases of seismic velocity model building using deep neural network techniques by classifying items according to the neural networks used in each study. We also included cases of generating training synthetic velocity models. Deep neural networks automatically optimize model parameters by training neural networks from large amounts of data. Thus, less human interaction is involved in the quality of the inversion results compared to that of conventional techniques and the computational cost of predicting a single velocity model after training is negligible. Additionally, unlike full-waveform inversions, the initial velocity model is not required. Several studies have demonstrated that deep neural network techniques achieve outstanding performance not only in computational cost but also in inversion results. Based on the research results, we analyzed and discussed the characteristics of deep neural network techniques for building velocity models.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.15 no.3
• /
• pp.481-490
• /
• 2002

The purpose of this paper is to evaluate the efficiency of various solution techniques and propose new efficient solution techniques for space trusses. Solution techniques used in this study are three load control methods (Newton-Raphson Method, modified Newton-Raphson Method, Secant-Newton Method), two load-displacement control methods(Arc-length Method, Work Increment Control Method) and three combined load-displacement control methods(Combined Arc-length Method I , Combined Arc-length MethodⅡ, Combined Work Increment Control Method). To evaluate the efficiency of these solution techniques, we must examine accuracy of their solutions, convergences and computing times of numerical examples. The combined load-displacement control methods are the most efficient in the geometric nonlinear solution techniques and in tracing post-buckling behavior of space truss. The combined work increment control method is the most efficient in tracing the buckling load of spate trusses with high degrees of freedom.

• Journal of computational fluids engineering
• /
• v.10 no.4 s.31
• /
• pp.24-31
• /
• 2005

A dynamic simulation on the missile staging system is conducted with numerical techniques. Both Euler equations and Navier-Stokes equations are numerically solved respectively. The dynamic simulation of two moving bodies is fully integrated into the computational fluid dynamics solution procedure. The Chimera grid scheme is applied in this simulation for unsteady supersonic flow analysis with dynamic modeling. The objective of the study is to investigate the problem pertaining to possible unstability in missile staging. In addition, the computational comparison between in viscid and viscid flow solvers is also performed in this study.

• Journal of computational fluids engineering
• /
• v.12 no.3
• /
• pp.20-28
• /
• 2007

A fast and robust method of grid generation to multiple functions has been developed for flow analysis in three dimensional space. It is based on the Non-Uniform Rational B-Spline(NURBS) of an approximation method. Many of NURBS intrinsic properties are introduced and much more easily understood. The grid generation method, details of numerical implementation. examples of application, and potential extensions of the current method are illustrated in this paper. The object of this study is to develop the surface grid generation and the grid cluster techniques capable of resolving complex flows with shock waves, expansion waves, shear layers. The knot insert method of Non-Uniform Rational B-Spline seems well worked. In addition, NURBS has been widely utilized to generate grids in the computational fluid dynamics community. Computational examples associated with practical configurations have shown the utilization of the algorithm.

• Journal of the Korean Society of Combustion
• /
• v.8 no.1
• /
• pp.46-56
• /
• 2003

The present paper is concerned with the development of the computational biology in the past half century and its relationship with combustion. The modem computational biology is considered to be initiated by the work of Alan Turing on the morphogenesis in 1952. This paper first touches the life and scientific achievement of Alan Turing and his theory on the morphogenesis based on the reactive-diffusive instability, called the Turing instability. The theory of Turing instability was later extended to the nonlinear realm of the reactive-diffusive systems, which is discussed in the framework of the excitable media by using the Oregonator model. Then, combustion analogies of the Turing instability and excitable media are discussed for the cellular instability, pattern forming combustion phenomena and flame edge. Finally, the recent efforts on numerical simulations of biological systems, employing the detailed bio-chemical knietic mechanism is discussed along with the possibility of applying the numerical combustion techniques to the computational cell biology.