• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.121-127
• /
• 2018

A system for diagnosing surface defects of long and large pipe inner overlay welds, 1m in diameter and 6m in length, was developed using a Liquid Penetrant Test (PT). First, CATIA was used to model all major units and PT machines in 3-dimensions. They were used for structural strength analysis and strain analysis, and to check the motion interference phenomenon of each unit to produce two-dimensional production drawings. Structural strength analysis and deformation analysis using the ANSYS results in a maximum equivalent stress of 44.901 MPa, which is less than the yield tensile strength of SS400 (200 MPa), a material of the PT Machine. An examination of the performance of the developed equipment revealed a maximum travel speed of 7.2 m/min., maximum rotational speed of 9 rpm, repeatable position accuracy of 1.2 mm, and inspection speed of $1.65m^2/min$. The results of the automatic PT-inspection system developed to check for surface defects, such as cracks, porosity, and undercut, were in accordance with the method of ASME SEC. V&VIII. In addition, the results of corrosion testing of the overlay weld layer in accordance with the ferric chloride fitting test by the method of ASME G48-11 indicated that the weight loss was $0.3g/m^2$, and met the specifications. Furthermore, the chemical composition of the overlay welds was analyzed according to the method described in ASTM A375-14, and all components met the specifications.

• Wind and Structures
• /
• v.26 no.4
• /
• pp.191-204
• /
• 2018

This article presents a study of the largest-ever (height = 220 m) cooling tower using the large eddy simulation (LES) method. Information about fluid fields around the tower and 3D aerodynamic time history in full construction process were obtained, and the wind pressure distribution along the entire tower predicted by the developed model was compared with standard curves and measured curves to validate the effectiveness of the simulating method. Based on that, average wind pressure distribution and characteristics of fluid fields in the construction process of ultra-large cooling tower were investigated. The characteristics of fluid fields in full construction process and their working principles were investigated based on wind speeds and vorticities under different construction conditions. Then, time domain characteristics of ultra-large cooling towers in full construction process, including fluctuating wind loads, extreme wind loads, lift and drag coefficients, and relationship of measuring points, were studied and fitting formula of extreme wind load as a function of height was developed based on the nonlinear least square method. Additionally, the frequency domain characteristics of wind loads on the constructing tower, including wind pressure power spectrum at typical measuring points, lift and drag power spectrum, circumferential correlations between typical measuring points, and vertical correlations of lift coefficient and drag coefficient, were analyzed. The results revealed that the random characteristics of fluctuating wind loads, as well as corresponding extreme wind pressure and power spectra curves, varied significantly and in real time with the height of the constructing tower. This study provides references for design of wind loads during construction period of ultra-large cooling towers.

• Korean Journal of Optics and Photonics
• /
• v.14 no.5
• /
• pp.504-508
• /
• 2003

We present a new point-diffraction interferometer, which has been devised for the three-dimensional profile measurement of light scattering rough surfaces. The interferometer system has multiple sources of two-point-diffraction and a CCD camera composed of an array of two-dimensional photodetectors. Each diffraction source is an independent two-point-diffraction interferometer made of a pair of single-mode optical fibers, which are housed in a ceramic ferrule to emit two spherical wave fronts by means of diffraction at their free ends. The two spherical wave fronts then interfere with each other and subsequently generate a unique fringe pattern on the test surface. A He-Ne source provides coherent light to the two fibers through a 2${\times}$l optical coupler, and one of the fibers is elongated by use of a piezoelectric tube to produce phase shifting. The xyz coordinates of the target surface are determined by fitting the measured phase data into a global model of multilateration. Measurement has been performed for the warpage inspection of chip scale packages (CSPs) that are tape-mounted on ball grid arrays (BGAs) and backside profile of a silicon wafer in the middle of integrated-circuit fabrication process. When a diagonal profile is measured across the wafer, the maximum discrepancy turns out to be 5.6 ${\mu}{\textrm}{m}$ with a standard deviation of 1.5 ${\mu}{\textrm}{m}$.

• Journal of Chest Surgery
• /
• v.28 no.6
• /
• pp.533-541
• /
• 1995

We recently developed a new model of moving actuator type totally implantable artificial heart[TIAH , based on the reverse position of the aortic and pulmonary conduits. This concept was proposed by one of surgeons in our team[Joon-Ryang Rho, M.D. to facilitate anatomical fitting of TIAHs. The moving actuator type electromechanical TIAH consisted of the left and right blood sacs, and the moving actuator including a motor. The inverted umbrella type polyurethane valves were used in the blood pumps. The aortic conduit was positioned anterior to the pulmonary conduit, which was the opposite relation to the conventional configuration of other total artificial hearts. We also adapted slip-in connectors for the aortic and pulmonary conduits. Two sheep , weighing 60-69 kg, were used for implantation. After small cervical incision and trans-sternal bilateral thoracotomy, cardiopulmonary bypass [CPB was administered using an American Optical 5-head pump and a membrane oxygenator[Univox-IC, Bentley . The anterior and posterior vena cavae were drained separately for venous return. An arterial return cannula was inserted into the right common carotid artery. During CPB, almost all of the ventricular myocardium was excised down to the atrioventricular groove and the artificial heart was implanted. We achieved 3-day survival in the first sheep and 2-day survival in the second. The day after operation the first sheep was successfully extubated and the second sheep was weaned from a respirator with good condition. After extubation, the first sheep walked around in the cage and fed herself. Serial laboratory and hemodynamic examinations were done during the experiments. In both sheep, pulmonary dysfunction was gradually developed, which was accompanied by acute renal failure. The animals were sacrificed and autopsy was done. Unexpected pregnnacy was incidentally found in both sheep. To our knowledge this is the first report of significant survival cases in the orthotopic implantation of electric TIAH using sheep.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.41 no.4
• /
• pp.296-305
• /
• 2005

Acoustic target strength (TS) of 12 commercially important fish species caught in the Korean waters had been investigated and their results were presented. Laboratory measurements of target strength on 12 dominant fish species were carried out at a frequencies of 75 kHz by single beam method under the controlled condition of the water tank with the 241 samples of dead and live fishes. The target strength pattern on individual fish of each species was measured as a function of tilt angle, ranging from $-45^{\circ}$ (head down aspect) to $45^{\circ}$ (head up aspect) in $0.2^{\circ}$ intervals, and the averaged target strength was estimated by assuming the tilt angle distribution as N ($-5.0^{\circ}$, $^15.0{\circ}$). The 75 to fish length relationship for each species was independently derived by a least - squares fitting procedure. Also, a linear regression analysis for all species was performed to reduce the data to a set of empirical equations showing the variation of target strength to fish length and fish species. An empirical model for fish target strength(TS, dB) averaged over the dorsal aspect of 158 fishes of 7 species and which spans the fish length(L, m) to wavelength(${\lambda}$, m) ratio between 6.2 and 21.3 was derived: TS: 27.03 Log(L)-7.7Log(${\kanbda}$)-17.21, ($r^2$=0.59).

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.185-202
• /
• 2012

.Since the value of information has been realized in the information society, the usage and collection of information has become important. A facial expression that contains thousands of information as an artistic painting can be described in thousands of words. Followed by the idea, there has recently been a number of attempts to provide customers and companies with an intelligent service, which enables the perception of human emotions through one's facial expressions. For example, MIT Media Lab, the leading organization in this research area, has developed the human emotion prediction model, and has applied their studies to the commercial business. In the academic area, a number of the conventional methods such as Multiple Regression Analysis (MRA) or Artificial Neural Networks (ANN) have been applied to predict human emotion in prior studies. However, MRA is generally criticized because of its low prediction accuracy. This is inevitable since MRA can only explain the linear relationship between the dependent variables and the independent variable. To mitigate the limitations of MRA, some studies like Jung and Kim (2012) have used ANN as the alternative, and they reported that ANN generated more accurate prediction than the statistical methods like MRA. However, it has also been criticized due to over fitting and the difficulty of the network design (e.g. setting the number of the layers and the number of the nodes in the hidden layers). Under this background, we propose a novel model using Support Vector Regression (SVR) in order to increase the prediction accuracy. SVR is an extensive version of Support Vector Machine (SVM) designated to solve the regression problems. The model produced by SVR only depends on a subset of the training data, because the cost function for building the model ignores any training data that is close (within a threshold ${\varepsilon}$) to the model prediction. Using SVR, we tried to build a model that can measure the level of arousal and valence from the facial features. To validate the usefulness of the proposed model, we collected the data of facial reactions when providing appropriate visual stimulating contents, and extracted the features from the data. Next, the steps of the preprocessing were taken to choose statistically significant variables. In total, 297 cases were used for the experiment. As the comparative models, we also applied MRA and ANN to the same data set. For SVR, we adopted '${\varepsilon}$-insensitive loss function', and 'grid search' technique to find the optimal values of the parameters like C, d, ${\sigma}^2$, and ${\varepsilon}$. In the case of ANN, we adopted a standard three-layer backpropagation network, which has a single hidden layer. The learning rate and momentum rate of ANN were set to 10%, and we used sigmoid function as the transfer function of hidden and output nodes. We performed the experiments repeatedly by varying the number of nodes in the hidden layer to n/2, n, 3n/2, and 2n, where n is the number of the input variables. The stopping condition for ANN was set to 50,000 learning events. And, we used MAE (Mean Absolute Error) as the measure for performance comparison. From the experiment, we found that SVR achieved the highest prediction accuracy for the hold-out data set compared to MRA and ANN. Regardless of the target variables (the level of arousal, or the level of positive / negative valence), SVR showed the best performance for the hold-out data set. ANN also outperformed MRA, however, it showed the considerably lower prediction accuracy than SVR for both target variables. The findings of our research are expected to be useful to the researchers or practitioners who are willing to build the models for recognizing human emotions.

• Journal of fish pathology
• /
• v.22 no.2
• /
• pp.125-135
• /
• 2009

The pharmacokinetic properties of oxolinic acid (OA) were studied after oral administration, intraperitoneal injection and dipping to cultured olive flounder, Paralichthys olivaceus (average 90 g, $23{\pm}1{^{\circ}C}$). Plasma samples were taken at 3, 5, 10, 15, 24, 30, 48, 96 and 144 h post-dose. In oral dosage at 15, 30 and 60 ㎎/㎏, the peak plasma concentrations of OA, which attained at 10~15 h post-dose, were 1.92, 2.45 and 3.72 $\mu{g}/m\ell$, respectively. In intraperitoneal injection with 10 and 20 ㎎/㎏, the peak plasma concentrations of OA, which attained at 10 h post-dose, were 4.1 and 4.8 $\mu{g}/m\ell$, respectively. In dipping in 30 and 50 ppm for 1 h, peak concentrations were observed at 5 h and 30 h post-dose, were 0.22 and 0.38 $\mu{g}/m\ell$, respectively. The kinetic profile of absorption, distribution and elimination of OA in plasma were analyzed fitting to a one-compartment model by WinNonlin program. Calculated parameters for a single oral dosage of 15, 30 and 60 ㎎/㎏, respectively, were: AUC (the area under the concentration-time curve)=70.93, 120.0 and 141.86 $\mu{g}$ $h/m\ell$ $T_{max}$ (time for maximum concentration)=16.22, 20.39 and 17.33 h; $C_{max}$ (maximum concentration)=���D1.61, 2.40 and 3.01 $\mu{g}/m\ell$. Following intraperitoneal injection of 10 and 20 ㎎/㎏, these parameters were AUC=184.7 and 315.92 $\mu{g}$ $h/m\ell$ $T_{max}$=5.91 and 6.26 h; $C_{max}$=4.19 and 4.45 $\mu{g}/m\ell$. Following dipping at 30 and 50 ppm, these parameters were AUC=17.58 and 21.69 $\mu{g}$ $h/m\ell$ $T_{max}$=19.08 and 31.43 h; $C_{max}$x=0.22 and 0.25 $\mu{g}/m\ell$.

• Asia pacific journal of information systems
• /
• v.19 no.2
• /
• pp.157-178
• /
• 2009

Corporate credit rating is a very important factor in the market for corporate debt. Information concerning corporate operations is often disseminated to market participants through the changes in credit ratings that are published by professional rating agencies, such as Standard and Poor's (S&P) and Moody's Investor Service. Since these agencies generally require a large fee for the service, and the periodically provided ratings sometimes do not reflect the default risk of the company at the time, it may be advantageous for bond-market participants to be able to classify credit ratings before the agencies actually publish them. As a result, it is very important for companies (especially, financial companies) to develop a proper model of credit rating. From a technical perspective, the credit rating constitutes a typical, multiclass, classification problem because rating agencies generally have ten or more categories of ratings. For example, S&P's ratings range from AAA for the highest-quality bonds to D for the lowest-quality bonds. The professional rating agencies emphasize the importance of analysts' subjective judgments in the determination of credit ratings. However, in practice, a mathematical model that uses the financial variables of companies plays an important role in determining credit ratings, since it is convenient to apply and cost efficient. These financial variables include the ratios that represent a company's leverage status, liquidity status, and profitability status. Several statistical and artificial intelligence (AI) techniques have been applied as tools for predicting credit ratings. Among them, artificial neural networks are most prevalent in the area of finance because of their broad applicability to many business problems and their preeminent ability to adapt. However, artificial neural networks also have many defects, including the difficulty in determining the values of the control parameters and the number of processing elements in the layer as well as the risk of over-fitting. Of late, because of their robustness and high accuracy, support vector machines (SVMs) have become popular as a solution for problems with generating accurate prediction. An SVM's solution may be globally optimal because SVMs seek to minimize structural risk. On the other hand, artificial neural network models may tend to find locally optimal solutions because they seek to minimize empirical risk. In addition, no parameters need to be tuned in SVMs, barring the upper bound for non-separable cases in linear SVMs. Since SVMs were originally devised for binary classification, however they are not intrinsically geared for multiclass classifications as in credit ratings. Thus, researchers have tried to extend the original SVM to multiclass classification. Hitherto, a variety of techniques to extend standard SVMs to multiclass SVMs (MSVMs) has been proposed in the literature Only a few types of MSVM are, however, tested using prior studies that apply MSVMs to credit ratings studies. In this study, we examined six different techniques of MSVMs: (1) One-Against-One, (2) One-Against-AIL (3) DAGSVM, (4) ECOC, (5) Method of Weston and Watkins, and (6) Method of Crammer and Singer. In addition, we examined the prediction accuracy of some modified version of conventional MSVM techniques. To find the most appropriate technique of MSVMs for corporate bond rating, we applied all the techniques of MSVMs to a real-world case of credit rating in Korea. The best application is in corporate bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. For our study the research data were collected from National Information and Credit Evaluation, Inc., a major bond-rating company in Korea. The data set is comprised of the bond-ratings for the year 2002 and various financial variables for 1,295 companies from the manufacturing industry in Korea. We compared the results of these techniques with one another, and with those of traditional methods for credit ratings, such as multiple discriminant analysis (MDA), multinomial logistic regression (MLOGIT), and artificial neural networks (ANNs). As a result, we found that DAGSVM with an ordered list was the best approach for the prediction of bond rating. In addition, we found that the modified version of ECOC approach can yield higher prediction accuracy for the cases showing clear patterns.