• Communications for Statistical Applications and Methods
• /
• v.21 no.4
• /
• pp.297-307
• /
• 2014

We derive a test statistic for the general linear test in the ridge regression model. The exact distribution for the test statistic is too difficult to derive; therefore, we suggest an approximate reference distribution. We use numerical studies to verify that the suggested distribution for the test statistic is appropriate. A asymptotic result for the test statistic also is considered.

• The Korean Journal of Applied Statistics
• /
• v.7 no.1
• /
• pp.75-82
• /
• 1994

It is known that collinearity among the explanatory variables in generalized linear models inflates the variance of maximum likelihood estimators. A ridge-type estimator is presented using penalized likelihood. A method for choosing a shrinkage parameter is discussed and this method is based on a prediction-oriented criterion, which is Mallow's $C_L$ statistic in a linear regression setting.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.1 s.68
• /
• pp.135-144
• /
• 2004

Since Modern bridges have a tendency to make the spans continuous and longer, the effect of concrete shrinkage and creep is very important and must be evaluated appropriately for the durability and safety of steel-concrete composite bridges. However, highway design specification in current use prescribes $180^{1\;2}$ as the final shrinkage strain. which is for less value than one resulted from many experimental researches and cause some problems in the construction of composite bridges due to the understimation of shrinkage strain. Thus, in this paper nonlinear analysis with time-steps applying the CEB-FIP(90) provision have been conducted for plate girder bridge, box girder bridge and Preflex beam bridge and the linear equivalent shrinkage strain for the design of composite bridges. which produces the stress equal to the values from the nonlinear analysis, has been calculated by comparing the results with the values following highway design specification. The results yield appropriately double values than $180^{1\;2}$ which highway design specification prescribes.

• Journal of Technologic Dentistry
• /
• v.42 no.2
• /
• pp.113-120
• /
• 2020

Purpose: This study is to compare and analyze the shrinkage of the specimen after UV irradiation of UV cured resin at 5, 15, and 30 minutes. Methods: A cylindrical UV cured specimen was produced using a stainless steel mold. UV cured resin specimens were prepared in three groups: 5 minutes cured (5M), 15 minutes cured (15M), and 30 minutes cured (30M). The measurement was made in total 3rd. The measurement was made in total 3rd. The primary measurement was made after 24 hours using a digital measuring instrument. The 2nd and 3rd measurements were deposited in a constant temperature water bath and the shrinkage was measured. The measured data was calculated by referring to the ASTM C326 linear measurement calculation method. T-test and One-way ANOVA were performed to test the significance between groups. The post-test was conducted with Tukey (α=0.05). Results: When the inner diameter and the outer diameter of the three groups not placed in the water bath were compared and analyzed, the contraction was the smallest at 6.8% in the 5M group, and the contraction was the largest at 7.3% in the 30M group. In the outer diameter, the contraction of the 5M group was the smallest at 3.5%, and the contraction of the 30M group was the largest at 4.5%. Shrinkage decreased in all three groups immersed in a water bath for 3-7 days. Conclusion: In the UV cured resin specimen, the shrinkage increased as the amount of UV irradiation increased.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.305-312
• /
• 2023

Physics-Informed Neural Network (PINN) is used to invert bubble size distributions from attenuation losses. By considering a linear system for the bubble population inversion, Adaptive Learned Iterative Shrinkage Thresholding Algorithm (Ada-LISTA), which has been solved linear systems in image processing, is used as a neural network architecture in PINN. Furthermore, a regularization based on the linear system is added to a loss function of PINN and it makes a PINN have better generalization by a solution satisfying the bubble physics. To evaluate an uncertainty of bubble estimation, deep ensemble is adopted. 20 Ada-LISTAs with different initial values are trained using the same training dataset. During test with attenuation losses different from those in the training dataset, the bubble size distribution and corresponding uncertainty are indicated by average and variance of 20 estimations, respectively. Deep ensemble Ada-LISTA demonstrate superior performance in inverting bubble size distributions than the conventional convex optimization solver of CVX.

• Fibers and Polymers
• /
• v.5 no.1
• /
• pp.19-24
• /
• 2004

PET yarns textured at different texturing conditions were treated with superheated steam or dry heat at different temperatures for different times. The effects of the treatment conditions on the thermomechanical and structural changes of the yarn were examined by shrinkage, X-ray diffraction and birefringence measurements. With increase in superheated steam temperature, the crystalline orientation factor and birefringence decreased, whereas crystal size increased. Dry heat treatment had a smaller effect on shrinkage and structural properties in comparison with superheated steam treatment. The additional shrinkage after texturing process was investigated. The effect of heat-setting in both media was more significant at $200^{\circ}C$. The time dependence of the properties was not linear.

• Steel and Composite Structures
• /
• v.2 no.2
• /
• pp.85-98
• /
• 2002

The aim of the paper is to present some results about the influence of rheological phenomena on steel-concrete composite beams. Both the cases of slab with normal and high performance concrete for one and two-span beams are analysed. A new finite element model that allows taking into account creep, shrinkage and cracking in tensile zones for concrete, along with non-linear behaviour of connection, steel beam and reinforcement, has been used. The main parameters that affect the response of the composite beam under the service load are highlighted. The influence of shrinkage on the slip over the supports is analysed, together with the cracking along the beam. At last, by performing a collapse analysis after a long-term analysis, the influence of rheological phenomena on the ductility demand of connection and reinforcement is analysed.

• Journal of the Korea Concrete Institute
• /
• v.21 no.4
• /
• pp.481-488
• /
• 2009

It has been reported that the magnitude and the development rate of autogenous shrinkage of cement paste, mortar and concrete were affected by history and magnitude of inner temperature at an early age. But it was not enough to explain the relation between hydration heat and autogenous shrinkage at an early age, because there was no certain analysis on histories of hydration heat and autogenous shrinkage in previous studies. In our prior study, to understand the relationship between hydration heat and autogenous shrinkage of concrete at an early age, the analysis method for histories of hydration heat and autogenous shrinkage was suggested. Based on this method, early age properties of hydration heat and autogenous shrinkage of high strength concrete with different sizes and hydration retardation were investigated in this study. As a result of the study, properties of hydration temperature and autogenous shrinkage were different according to specimen size and hydration retardation. However, there was a close relationship between hydration temperature and autogenous shrinkage at an early age, especially between HHV and ASV as linear slopes of the sections where hydration temperature and autogenous shrinkage increase rapidly; the higher HHV, the higher ASV and the greater ultimate autogenous shrinkage. And it was found that, among the setting time, bend point and temperature increasing point, they were close relationship each other on cement hydration process.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.31 no.3
• /
• pp.516-526
• /
• 2004

Polymerization shrinkage of photoinitiation type composite resin cause several clinical problems. The purpose of this study was to evaluate the shrinkage strain stress, linear polymerization shrinkage, compressive strength and microhardness of recently developed composite resins. The composite resins were divided into four groups according to the contents of matrix and filler type. Group I : $Denfil^{TM}$(Vericom, Korea) with conventional matrix, Group II : $Charmfil^{(R)}$(Dentkist, Korea) with microfiller and nanofller mixture, Group III : $Filtek^{TM}$ Z250(3M-ESPE, USA) TEGDMA replaced by UDMA and Bis-EMA(6) in the matrix, and Group IV : $Filtek^{TM}$ Supreme(3M-ESPE, USA) using pure nanofiller. Preparation of acrylic molds were followed by filling and curing with light gun. Strain gauges were attached to each sample and the leads were connected to a strainmeter. With strainmeter shrinkage strain stress and linear polymerization shrinkage was measured for 10 minutes. The data detected at 1 minute and 10 minutes were analysed statistically with ONE-way ANOVA test. To evaluate the mechanical properties of tested materials, compressive hardness test and microhardness test were also rendered. The results can be summarized as follows : 1. Filling materials in acrylic molds showed initial temporary expansion in the early phase of polymerization. This was followed by contraction with the rapid increase in strain stress during the first 1 minute and gradually decreased during post-gel shrinkage phase. After 1 minute, there's no statistical differences of strain stress between groups. The highest strain stress was found in group IV and followed by group III, I, II at 10 minutes-measurement(p>.05). In regression analysis of strain stress, group III showed minimal inclination and followed by group II, I, IV during 1 minute. 2. In linear polymerization shrinkage test, the composite resins in every group showed initial increase of shrinkage velocity during the first 1 minute, followed by gradually decrease of shrinkage velocity. After 1 minute, group IV and group III showed statistical difference(p<.05). After 10 minutes, there were statistical differences between group IV and group I, III(p<.05) and between group II and group III(p<.05). In regression analysis of linear polymerization shrinkage, group II showed minimal inclination and followed by group IV, III, I during 1 minute. 3. In compressive strength test, group III showed the highest strength and followed by group II, IV, I. There were statistical differences between group III and group IV, I(p<.05). 4. In microhardness test, upper surfaces showed higher value than lower surfaces in every group(p<.05).

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2002.10a
• /
• pp.36-40
• /
• 2002

The development of feed drive system with high speed and accuracy has been a major issue in the machine tool industry. Linear motors can be used as the efficient tool to achieve fast feed mechanism and high accuracy. However. a high speed feed drive system with linear motors can generate heat problems such as the variation of temperature distribution and the resultant thermal stress. In this paper, the important heat sources and the resultant thermal errors are presented. The thermal deformation characteristics of the machine tool with linear motors were identified, which are thermal expansion of linear scale, shrinkage, expansion and bending in the machine tool structure.