• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1229-1244
• /
• 2017

In recent years, as demand for data-based analytical methodologies increases in various fields, optimization methods have been developed to handle them. In particular, various constraints required for problems in statistics and machine learning can be solved by convex optimization. Alternating direction method of multipliers (ADMM) can effectively deal with linear constraints, and it can be effectively used as a parallel optimization algorithm. ADMM is an approximation algorithm that solves complex original problems by dividing and combining the partial problems that are easier to optimize than original problems. It is useful for optimizing non-smooth or composite objective functions. It is widely used in statistical and machine learning because it can systematically construct algorithms based on dual theory and proximal operator. In this paper, we will examine applications of ADMM algorithm in various fields related to statistics, and focus on two major points: (1) splitting strategy of objective function, and (2) role of the proximal operator in explaining the Lagrangian method and its dual problem. In this case, we introduce methodologies that utilize regularization. Simulation results are presented to demonstrate effectiveness of the lasso.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.12 no.2
• /
• pp.43-54
• /
• 1992

For shallow arches with large dynamic loading, linear analysis is no longer considered as practical and accurate. In this study, a method is presented for the dynamic analysis of shallow arches in which geometric nonlinearity must be considered. A program is developed for the analysis of the nonlinear dynamic behavior and for evaluation of critical buckling loads of shallow arches. Geometric nonlinearity is modeled using Lagrangian description of the motion. The finite element analysis procedure is used to solve the dynamic equation of motion and Newmark method is adopted in the approximation of time integration. A shallow arch subject to radial step loads is analyzed. The results are compared with those from other researches to verify the developed program. The behavior of arches is analyzed using the non-dimensional time, load, and shape parameters. It is shown that geometric nonlinearity should be considered in the analysis of shallow arches and probability of buckling failure is getting higher as arches are getting shallower. It is confirmed that arches with the same shape parameter have the same deflection ratio at the same time parameter when arches are loaded with the same parametric load. In addition, it is proved that buckling of arches with the same shape parameter occurs at the same load parameter. Circular arches, which are under a single or uniform normal load, are analyzed for comparison. A parabolic arch with radial step load is also analyzed. It is verified that the developed program is applicable for those problems.

• Journal of KIISE:Computer Systems and Theory
• /
• v.32 no.11_12
• /
• pp.692-704
• /
• 2005

With the improvement of computer hardware, GPUs(Graphics Processor Units) have tremendous memory bandwidth and computation power. This leads GPUs to use in general purpose computation. Especially, GPU implementation of compute-intensive physics based simulations is actively studied. In the solution of differential equations which are base of physics simulations, tridiagonal matrix systems occur repeatedly by finite-difference approximation. From the point of view of physics based simulations, fast solution of tridiagonal matrix system is important research field. We propose a fast GPU implementation for the solution of tridiagonal matrix systems. In this paper, we implement the cyclic reduction(also known as odd-even reduction) algorithm which is a popular choice for vector processors. We obtained a considerable performance improvement for solving tridiagonal matrix systems over Thomas method and conjugate gradient method. Thomas method is well known as a method for solving tridiagonal matrix systems on CPU and conjugate gradient method has shown good results on GPU. We experimented our proposed method by applying it to heat conduction, advection-diffusion, and shallow water simulations. The results of these simulations have shown a remarkable performance of over 35 frame-per-second on the 1024x1024 grid.

• Korea-Australia Rheology Journal
• /
• v.17 no.3
• /
• pp.131-140
• /
• 2005

In this work we show the results of our most recent Direct Numerical Simulations (DNS) of turbulent viscoelastic channel flow using spectral spatial approximations and a stabilizing artificial diffusion in the viscoelastic constitutive model. The Finite-Elasticity Non-Linear Elastic Dumbbell model with the Peterlin approximation (FENE-P) is used to represent the effect of polymer molecules in solution, The corresponding rheological parameters are chosen so that to get closer to the conditions corresponding to maximum drag reduction: A high extensibility parameter (60) and a moderate solvent viscosity ratio (0.8) are used with two different friction Weissenberg numbers (50 and 100). We then first find that the corresponding achieved drag reduction, in the range of friction Reynolds numbers used in this work (180-590), is insensitive to the Reynolds number (in accordance to previous work). The obtained drag reduction is at the level of $49\%\;and\;63\%$, for the friction Weissenberg numbers 50 and 100, respectively. The largest value is substantially higher than any of our previous simulations, performed at more moderate levels of viscoelasticity (i.e. higher viscosity ratio and smaller extensibility parameter values). Therefore, the maximum extensional viscosity exhibited by the modeled system and the friction Weissenberg number can still be considered as the dominant factors determining the levels of drag reduction. These can reach high values, even for of dilute polymer solution (the system modeled by the FENE-P model), provided the flow viscoelasticity is high, corresponding to a high polymer molecular weight (which translates to a high extensibility parameter) and a high friction Weissenberg number. Based on that and the changes observed in the turbulent structure and in the most prevalent statistics, as presented in this work, we can still rationalize for an increasing extensional resistance-based drag reduction mechanism as the most prevalent mechanism for drag reduction, the same one evidenced in our previous work: As the polymer elasticity increases, so does the resistance offered to extensional deformation. That, in turn, changes the structure of the most energy-containing turbulent eddies (they become wider, more well correlated, and weaker in intensity) so that they become less efficient in transferring momentum, thus leading to drag reduction. Such a continuum, rheology-based, mechanism has first been proposed in the early 70s independently by Metzner and Lamley and is to be contrasted against any molecularly based explanations.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.14 no.8
• /
• pp.845-854
• /
• 2003

In this paper, the characteristics of compensation for WDM channel signal distortion due to both chromatic dispersion and Ken effect in 1,000 km 200 Gbps(5${\times}$40 Gbps) WDM systems was investigated. The WDM system has a path-averaged intensity approximation(PAIA) mid-span spectral inversion(MSSI) as a compensation method. This system has a highly nonlinear dispersion shifted fiber(HNL-DSF) optical phase conjugator(OPC) in the mid-way of transmission line. In order to evaluate the degree of compensation, 1 dB eye opening penalty(EOP), bit error rate(BER) characteristics and power penalty of 10$\^$-9/ BER are used. It is confirmed that HNL-DSF is an useful nonlinear medium in OPC fur wideband WDM system with PAIA MSSI and that the optimal compensation for WDM channel distortion is achieved by the selection of pump light power of OPC, which equalize the conjugated light power into the second half fiber section with the input WDM signal light power depending on total transmission length, dispersion coefficient of fiber, OPC pump light wavelength, conversion efficiency of WDM channel in OPC.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.43 no.3
• /
• pp.152-159
• /
• 2017

Objectives: The aim of this study was to investigate the effects of advancement magnitude and changes in mandibular plane angle on the stability of mandibular advancement. Materials and Methods: This retrospective cohort study evaluated the postoperative stability of mandibular advancement in class II skeletal subjects who underwent bilateral sagittal split osteotomy. Radiographs taken preoperatively, immediately postoperatively and 1 year postoperatively were traced and analyzed using linear and angular measurements. To determine horizontal and vertical relapse, an X-Y coordinate system was established in which the X-axis was constructed by rotating S-N downward by $7^{\circ}$ (approximation of the Frankfort horizontal plane) and the Y-axis was defined as a line perpendicular to the X-axis and passing through the point Sella. For certain reference points including point A, point B, pogonion and menton, the perpendicular distance between each point and both axes was determined and cephalometric variables were recorded as X and Y coordinates. Results: Twenty-five subjects were studied. A significant correlation between the amount of mandibular advancement and relapse in the B point (vertical and horizontal) and the pogonion point was observed (vertical and horizontal, P<0.001). Evaluation of data demonstrated a positive correlation between the mandibular plane angle (SN/ML) change and vertical relapse in the B point (P<0.05). A simple regression model demonstrated that 74% of horizontal relapse and 42.3% of vertical relapse in the B point was related to the amount of mandibular advancement. The receiver operating characteristic test showed that 8.5 mm mandibular advancement is related to a relapse rate of 1 mm or more in the pogonion, vertically or horizontally. Conclusion: The magnitude of mandibular advancement is a stronger surgical predictor for horizontal rather than vertical relapse at the B point. Changes in mandibular plane angle (SN/ML) during surgery affect vertical, but not horizontal relapse at the B point.

• Journal of the Korean Institute of Gas
• /
• v.24 no.2
• /
• pp.44-49
• /
• 2020

In this paper, a measuring circuit is designed through analyzing manufacture specification of the sensor based on MOS. And the best input-output polynomial are induced that really gas sensors are used in gas safety management industrial fields. Response characteristics of a MOS gas sensor is analysed by through sensor's output voltages are measured after standard gases with six kinds of concentrations are manufactured and are injected to the sensor. A lookup table is created by relations of sensor's output voltages by injecting gases with other concentrations. Because data of the formed lookup table are equal interval, a polynomial can be induced of method of approximation function. So the 5th polynomial of input-output for a sensor is defined, coefficients are calculated by using least squares method, and the 5th polynomial is completed for representing characteristics of the sensor. If the proposed polynomial is applied to gas leak detectors, an inverse transformation of polynomial and programing of array codes are recreated. In this research, polynomial is implemented with array types that intervals of values of a lookup table are one-fifth sampled and interpolated. The performance of proposed 5th calibration equation is verified that errors are reduced than a linear expression when tests are performed by measurement of concentrations against injection of standard gases.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.8 no.2
• /
• pp.123-136
• /
• 1996

A three-dimensional Galerkin-FEM model which can handle the temporal and spatial variation of density is presented. The hydrostatic approximation is used and density effects are included by means of conservation equation of heat and the equation of state. The finite difference grids are used in the horizontal plane and a set of linear-shape functions is used for the vertical expansion. The similarity transform is introduced to solve resultant matrix equations. The proposed model was first applied to the density-driven circulation in an idealized basin in the presence of the heat exchange between the air and the sea. The advection terms in the momentum equation were ignored, while the convection terms were retained in the heat equation. Coefficients of the vertical eddy viscosity and diffusivity were fixed to be constant. Calculation in a non-rotating idealized basin shows that the difference in heat capacity with depth gives rise to the horizontal gradient of temperature. Consequently, there is a steady new in the upper layer in the direction of increasing depth with compensatory counter flow .in the lower layer. With Coriolis force, geostrophic flow was predominant due to the balance between the pressure gradient and the Coriolis force. As a test in region of irregular topography, the model is applied to the Yellow Sea. Although the resultant flow was very complex, the character of the flow Showed to be geostrophic on the whole.

• Korean Journal of Remote Sensing
• /
• v.24 no.2
• /
• pp.165-177
• /
• 2008

Small baseline subset (SBAS) algorithm has been recently developed using an appropriate combination of differential interferograms, which are characterized by a small baseline in order to minimize the spatial decorrelation. This algorithm uses the singular value decomposition (SVD) to measure the time-series surface deformation from the differential interferograms which are not temporally connected. And it mitigates the atmospheric effect in the time-series surface deformation by using spatially low-pass and temporally high-pass filter. Nevertheless, it is not easy to correct the phase unwrapping error of each interferogram and to mitigate the time-varying noise component of the surface deformation from this algorithm due to the assumption of the linear surface deformation in the beginning of the observation. In this paper, we present an improved SBAS technique to complement these problems. Our improved SBAS algorithm uses an iterative approach to minimize the phase unwrapping error of each differential interferogram. This algorithm also uses finite difference method to suppress the time-varying noise component of the surface deformation. We tested our improved SBAS algorithm and evaluated its performance using 26 images of ERS-1/2 data and 21 images of RADARSAT-1 fine beam (F5) data at each different locations. Maximum deformation amount of 40cm in the radar line of sight (LOS) was estimated from ERS-l/2 datasets during about 13 years, whereas 3 cm deformation was estimated from RADARSAT-1 ones during about two years.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.2
• /
• pp.73-84
• /
• 1993

In this research, configuration design optimization of plane truss structure has been tested by using decomposition technique. In the first level, the problem of transferring the nonlinear programming problem to linear programming problem has been effectively solved and the number of the structural analysis necessary for doing the sensitivity analysis can be decreased by developing stress constraint into member stress approximation according to the design space approach which has been proved to be efficient to the sensitivity analysis. And the weight function has been adopted as cost function in order to minimize structures. For the design constraint, allowable stress, buckling stress, displacement constraint under multi-condition and upper and lower constraints of the design variable are considered. In the second level, the nodal point coordinates of the truss structure are used as coordinating variable and the objective function has been taken as the weight function. By treating the nodal point coordinates as design variable, unconstrained optimal design problems are easy to solve. The decomposition method which optimize the section areas in the first level and optimize configuration variables in the second level was applied to the plane truss structures. The numerical comparisons with results which are obtained from numerical test for several truss structures with various shapes and any design criteria show that convergence rate is very fast regardless of constraint types and configuration of truss structures. And the optimal configuration of the truss structures obtained in this study is almost the identical one from other results. The total weight couldbe decreased by 5.4% - 15.4% when optimal configuration was accomplished, though there is some difference.