• The Korean Journal of Applied Statistics
• /
• v.15 no.1
• /
• pp.57-71
• /
• 2002

We introduce three types of exponential survival models, such as simple model, change-point model and finite mixture model in this paper. Among these models, in order to choose the best model, the model choice method is proposed using Gelfand and Ghosh(1998)'s idea. Then to avoid the computational difficulties, data augmentation method (Tanner and Wong, 1987) and Gibbs sampler (Gelfand and Smith, 1990) are employed. Our methodology is applied to both simulated data and Stangl (1991)'s On-impramint Hydrochloride data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.11
• /
• pp.1012-1016
• /
• 2004

In the present paper, equivalent beam and equivalent bimorph beam models for IPMC(ionic Polymer-Metal Composite) actuators are described. Physical properties of an IPMC, such as Young's modulus and electro-mechanical coupling coefficient. are determined from the rule of mixture, bimorph beam equations, and measured force-displacement data of a cantilevered IPMC actuator. By using a beam equation with estimated physical properties, actuation displacements of a cantilevered IPMC actuator was calculated and a good agreement between the computed tip displacements and the measured data was observed. Finite element analysis(FEA) combined with the estimated physical properties was used to reproduce the force-displacement relationship of an IPMC actuator. Results from the FEA agreed well with the measure data. The proposed models might be used for modeling of IPMC actuators with complicated shapes and boundary conditions.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.57-68
• /
• 2017

Clustering algorithms attempt to find a partition of a finite set of objects in to a potentially predetermined number of nonempty subsets. Gibbs sampling of a normal mixture of linear mixed regressions with a Dirichlet prior distribution calculates posterior probabilities when the number of clusters was known. Our approach provides simultaneous partitioning and parameter estimation with the computation of classification probabilities. A Monte Carlo study of curve estimation results showed that the model was useful for function estimation. Examples are given to show how these models perform on real data.

• Advances in nano research
• /
• v.14 no.5
• /
• pp.443-450
• /
• 2023

Carbon nanotubes (CNTs) have received increased interest in reinforcing research for polymer matrix composites due to their exceptional mechanical characteristics. Its high surface area/volume ratio and aspect ratio enable polymer-based composites to make the most of its features. This study focuses on the experimental tensile testing and fabrication of carbon nanotube reinforced composite (CNTRC) beams, exploring various micromechanical models. By examining the performance of these models alongside experimental results, the research aims to better understand and optimize the mechanical properties of CNTRC materials. Tensile properties of neat epoxy and 0.3%; 0.4% and 0.5% by CNT reinforced laminated single layer (0°/90°) carbon fiber composite beams were investigated. The composite plates were produced in accordance with ASTM D7264 standard. The tensile test was performed in order to see the mechanical properties of the composite beams. The results showed that the optimum amount of CNT was 0.3% based on the tensile capacity. The capacity was significantly reduced when 0.4% CNT was utilized. Moreover, the experimental results are compared with Finite Element Models using ABAQUS. Hashin Failure Criteria was utilized to predict the tensile capacity. Good conformance was observed between experimental and numerical models. More importantly is that Young' Moduli of the specimens is compared with the prediction Halpin-Tsai and Mixture-Rule. Although Halpin-Tsai can accurately predict the Young's Moduli of the specimens, the accuracy of Mixture-Rule was significantly low.

• Communications of the Korean Mathematical Society
• /
• v.14 no.2
• /
• pp.405-423
• /
• 1999

We consider a statistical multiplexer model with finite buffer capacity and finite number of independent identical 3-state bursty voice sources. The burstiness of the sources is modeled by describing both two different active periods (at the rate of one packet perslot) and the passive periods during which no packets are generated. Assuming a mixture of two geometric distributions for active period and a geometric distribution for passive period and geometric distribution for passive period, we derive the recursive algorithm for the probability mass function of the buffer contents (in packets). We also obtain loss probability and the distribution of packet delay. Numerical results show that the system performance deteriorates considerably as the variance of the active period increases. Also, we see that the loss probability of 2-state Markov models is less than that of 3-state Markov models.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.8
• /
• pp.718-722
• /
• 2015

This paper proposes a fractional-order PID (Proportional-Integral-Derivative) control used electromagnetic strip stabilization controller in a continuous galvanizing line. Compared to a conventional PID controller, a fractional-order PID controller has integration-fractional-order and derivation-fractional-order as additional control parameters. Thanks to increased control parameters, more precise controller adjustment is available. In addition, accurate transfer function of a real system generally has a fractional-order form. Therefore, it is more adequate to use a fractional-order PID controller than a conventional PID controller for a real world system. Finite element models of a $1200{\times}2000{\times}0.8mm$ strip, which were extracted using a commercial software ANSYS were used as simulation plants, and Gaussian mixture models were used to find optimized control parameters that can reduce the strip vibrations to the lowest amplitude. Simulation results show that a fractional-order PID controller significantly reduces strip vibration and transient response time than a conventional PID controller.

• The Korean Journal of Applied Statistics
• /
• v.32 no.2
• /
• pp.199-211
• /
• 2019

This paper is concerned with issues in the finite mixture of regression modeling as well as the simultaneous selection of the number of mixing components and relevant predictors. We propose a penalized likelihood method for both mixture components and regression coefficients that enable the simultaneous identification of significant variables and the determination of important mixture components in mixture of regression models. To avoid over-fitting and bias problems, we applied smoothly clipped absolute deviation (SCAD) penalties on the logarithm of component probabilities suggested by Huang et al. (Statistical Sinica, 27, 147-169, 2013) as well as several well-known penalty functions for coefficients in regression models. Simulation studies reveal that our method is satisfactory with well-known penalties such as SCAD, MCP, and adaptive lasso.

• Computers and Concrete
• /
• v.16 no.5
• /
• pp.759-774
• /
• 2015

This study simulates the flexural behavior of ultra-high-performance fiber-reinforced concrete (UHPFRC) beams reinforced with steel and glass fiber-reinforced polymer (GFRP) rebars. For this, micromechanics-based modeling was first carried out on the basis of single fiber pullout models considering inclination angle. Two different tension-softening curves (TSCs) with the assumptions of 2-dimensional (2-D) and 3-dimensional (3-D) random fiber orientations were obtained from the micromechanics-based modeling, and linear elastic compressive and tensile models before the occurrence of cracks were obtained from the mechanical tests and rule of mixture. Finite element analysis incorporating smeared crack model was used due to the multiple cracking behaviors of structural UHPFRC beams, and the characteristic length of two times the element width (or two times the average crack spacing at the peak load) was suggested as a result of parametric study. Analytical results showed that the assumption of 2-D random fiber orientation is appropriate to a non-reinforced UHPFRC beam, whereas the assumption of 3-D random fiber orientation is suitable for UHPFRC beams reinforced with steel and GFRP rebars due to disorder of fiber alignment from the internal reinforcements. The micromechanics-based finite element analysis also well predicted the serviceability deflections of UHPFRC beams with GFRP rebars and hybrid reinforcements.

• Transactions of the Society of Information Storage Systems
• /
• v.7 no.2
• /
• pp.60-64
• /
• 2011

Chemical mechanical planarization is one of the core processes in fabrication of semiconductors, which are increasingly used for information storage devices like solid state drives. For higher data capacity in storage devices, CMP process is required to show ultimate precision and accuracy. In this work, 2-dimensional finite element models were developed to investigate the effects of the slurry particle impact on microscratch generation and the phenomena generated at pad-particle-wafer contact interface. The results revealed that no plastic deformation and corresponding material removal could be generated by simple impact of slurry particles under real CMP conditions. From the results of finite element simulations, it could be concluded that the pad-particle mixture formed in CMP process would be one of major factors leading to microscratch generation.

• Journal of the Korean Data and Information Science Society
• /
• v.9 no.2
• /
• pp.289-298
• /
• 1998

We investigate the distribution of likelihood ratio test(LRT) of null hypothesis a sample is from single gamma with unknown shape and scale against the alternative hypothesis a sample is from a mixture of two gammas, each with unknown scale and unknown (but equal) scale. To obtain stable maximum likelihood estimates(MLE) of a mixture of two gamma distributions, the EM(Dempster, Laird, and Robin(1977))and Modified Newton(Jensen and Johansen(1991)) algorithms were implemented. Based on EM, we made a simple structure likelihood equation for each parameter and could obtain stable solution by Modified Newton Algorithms. Simulation study was conducted to investigate the distribution of LRT for sample size n = 25, 50, 75, 100, 50, 200, 300, 400, 500 with 2500 replications. To determine the small sample distribution of LRT, I considered the model of a gamma distribution with shape parameter equal to 1 + f(n) and scale parameter equal to 2. The simulation results indicate that the null distribution is essentially invariant to the value of the shape parameter. Modeling of the null distribution indicates that it is well approximated by a gamma distribution with shape parameter equal to the quantity $0.927+1.18/\sqrt{n}$ and scale parameter equal to 2.16.