• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.1
• /
• pp.114-123
• /
• 2018

This paper concerns a design and learning method of softmax function neural networks based on K-means clustering. The partial discharge data Information is preliminarily processed through simulation using an Epoxy Mica Coupling sensor and an internal Phase Resolved Partial Discharge Analysis algorithm. The obtained information is processed according to the characteristics of the pattern using a Motor Insulation Monitoring System program. At this time, the processed data are total 4 types that void discharge, corona discharge, surface discharge and slot discharge. The partial discharge data with high dimensional input variables are secondarily processed by principal component analysis method and reduced with keeping the characteristics of pattern as low dimensional input variables. And therefore, the pattern classifier processing speed exhibits improved effects. In addition, in the process of extracting the partial discharge data through the MIMS program, the magnitude of amplitude is divided into the maximum value and the average value, and two pattern characteristics are set and compared and analyzed. In the first half of the proposed partial discharge pattern classifier, the input and hidden layers are classified by using the K-means clustering method and the output of the hidden layer is obtained. In the latter part, the cross entropy error function is used for parameter learning between the hidden layer and the output layer. The final output layer is output as a normalized probability value between 0 and 1 using the softmax function. The advantage of using the softmax function is that it allows access and application of multiple class problems and stochastic interpretation. First of all, there is an advantage that one output value affects the remaining output value and its accompanying learning is accelerated. Also, to solve the overfitting problem, L2-normalization is applied. To prove the superiority of the proposed pattern classifier, we compare and analyze the classification rate with conventional radial basis function neural networks.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.19 no.4
• /
• pp.52-60
• /
• 2015

Operation area of corps was expanded under military reformation, and extending range of 155 mm howitzer became important issue. New approach is needed to extend range to 80 kim. Ramjet engine is air breathing engine, and it can provide specific impulse several times more than solid rocket motor so that range is extended using same weight of propellant. If the ramjet engine is gun-launched system, it does not require any other booster because muzzle velocity is near Mach 3. Especially solid fuel ramjet (SFRJ) does not have any moving part so that it is favorable for gun-launching system which is under high stress during launching. In this paper, we design air intake, combustion chamber, and nozzle of 155 mm gun launched ramjet propelled artillery shell with inviscid flow assumption. We conduct parameter study to have range more than 80 km, and maximum high explosive volume.

• The Journal of the Korea Contents Association
• /
• v.4 no.4
• /
• pp.123-131
• /
• 2004

Internet is absolutely contributed to information technology revolution nowadays. Internet services such as voice and data, etc. are provided home or small office via home gateway. The development of communication equipment via home gateway is implemented rapidly, and its product various. This paper presents the design and implementation of the VoIP equipment including the telephone function based on the embedded environment and being connected to the home gateway and the PC because of taking 2-ethernet LAN ports. As developing environment, the STLC1502 developed at ST Microelectronics as single chip solution, VxWorks as RTOS, and C language as coding mechanism are used. The verification of the developed systems for the voice test is carried out for the gatekeeper via Internet. The performance parameter is considered as the call processing capacity measuring the time of the call setup and clearing, and the data processing capacity for the file transfer. As a call setup and clearing is about 95ms, the call processing capacity is about 10 calls per second. The data processing capacity is 5.7Mbps in case of file transfer of server and client environment. Therefore the performance result is satisfied in the aspect of the call processing time and the data transfer time in Internet.

• Journal of the Korean Magnetic Resonance Society
• /
• v.9 no.1
• /
• pp.48-60
• /
• 2005

Purpose: The purpose of this study was to design and build an optimized birdcage resonator configuration with a low pass filter, which would facilitate the acquisition of high-resolution 3D-image of small animals at 3T MRI system. Methods and Materials: The birdcage resonator with 12-element structures was built, in order to ensure B1 homogeneity over the image volume and maximum filling factor, and hence to maximize the signal to noise ratio (SNR) and resolution of the 3-dimensional images. The diameter and length of each element of a birdcage resonator were as follows: (1) diameter 13 cm, length 22 cm, (2) diameter 15 cm, length 22 cm, (3) diameter 17 cm, length 25 cm. Spin echo pulse sequence and fast spin echo pulse sequence were employed in obtaining MR images. The quality of the manufactured birdcage resonators wes evaluated on the basis of the return loss following matching and tuning process. Results: The experimental MR image of phantoms by the various manufactured birdcage resonators were obtained to compare the SNR in accordance with the size of objects. The size of an object to that of coil was identified by parameters that were estimated from the image of a phantom. First, the diameter of the birdcage resonator was 15cm, and the ratio of the tangerine to the birdcage resonator accounted for approximately 27%. The Q factor was 53.2 and the SNR was 150.7. Second, at the same birdcage resonator, the ratio of the orange was approximately 53%. The SNR and the Q parameter was 212.8 and 91.2, respectively. Conclusion: The present study demonstrated that if birdcage resonators have the same forms, SNR could be different depending on the size of an object, especially when the size of an object to that of coil is approximately 40~80%, the former is bigger than the latter. Therefore, when the size of an object to be observed is smaller than that of coil, the coil should be manufactured in accordance with the size of an object in order to obtain much more excellent images.

• Journal of the Korean Society for Railway
• /
• v.12 no.1
• /
• pp.31-38
• /
• 2009

The selection of the support system is an important design parameter in design and construction of the tunnel using the new Australian tunnel method. It is a common practice to select the support based on the rock mass grade, in which the rock mass is classified into five rock groups. The method is applicable if the characteristics of the rock mass are uniform in the direction of tunnel excavation. However, such case is seldom encountered in practice and not applicable when the properties vary along the longitudinal direction. This study performs comprehensive three dimensional finite difference analyses to investigate the ground deformation pattern for cases in which the rock mass properties change in the direction of the tunnel axis. The numerically calculated displacements at the tunnel crown show that the displacement is highly dependent on the stiffness contrast of the rock masses. The results strongly indicate the need to select the support type $0.5{\sim}1.0D$ before the rock mass boundary. The paper proposes a new guideline for selecting the support type based the results of the analyses.

• Journal of the Microelectronics and Packaging Society
• /
• v.22 no.1
• /
• pp.75-81
• /
• 2015

In this paper, numerical analysis by finite element method and parameter design by the Taguchi method were used to reduce warpage of a two passive components embedded double side substrate for PoP(Package on Package). The effect of thickness of circuit layers (L1, L2) and thickness of solder resist (SR_top, SR_BTM) were analyzed with 4 variations and 3 levels(minimum, average and maximum thickness) to find optimized thickness conditions. Also, paste effect of solder resist on unit area of top surface was analyzed. Finally, experiments was carried out to prove numerical analysis and the Taguchi method. Based on the numerical and experimental results, it was known that circuit layer in ball side of substrate was the most severe determining deviation for reducing warpage. Buried circuit layer in chip side, solder resist and were insignificant effects on warpage relatively. However, warpage decreased as circuit layer in ball side thickness increased but effect of solder resist and circuit layer in chip side thickness were conversely.

• Journal of the Korean Geotechnical Society
• /
• v.18 no.2
• /
• pp.51-64
• /
• 2002

Observational results of ground movement during the construction were very different from those predicted during the analysis of design step because of the uncertainty of the numerical analysis modelling, the soil parameter, and the condition of a construction field, etc., however accurately numerical analysis method was applied for prediction of ground movement per the excavation step. Therefore, the management system through the construction field measurement should be achieved for grasping the situation during the excavation. Until now, the measurement system restricted by 'Absolute Value Management system'analyzing only the stability of present step has been executed. So, it was difficult to expect the prediction of ground movement fur the next excavation step. In this situation, this study developed 'The Management system TOMAS-EXCAV'consisted of 'Absolute value management system'analyzing the stability of present step and 'Prediction management system'expecting the ground movement of next excavation step and analyzing the stability of next excavation step by 'Back Analysis'. TOMAS-EXCAV could be applied to all the uncertainty of earth retaining structures analysis by connecting 'Forward analysis program'and 'Back analysis program'and optimizing the main design variables using SQP-MMFD optimization method through measurement results. The applicability of TOMAS-EXCAV was confirmed by back analysis selecting two earth retains construction fields.

• Journal of the Korean Society of Tobacco Science
• /
• v.21 no.2
• /
• pp.109-118
• /
• 1999

A model predicted describing the effect of cigarette making materials on the level of filter ventilation was developed and evaluated. The developed model was expressed in terms of a linear and quadratic relationship which was validated with experimental measurements for different porosity of plug wrap and tipping paper, unencapsulated pressure drop of filter plug and cigarette column and vent position. Forty-six experimental frequencies were determined as a result of using three levels with five factors Box-Behnken design and analyzed by the multiple regression analysis with backward stepwise in STATISTICA/PC under restricted conditions. The four factors, except filter pressure drop variable, were statistically significant at the level of 0.05 but most of all linear by linear interactions were comparatively lower significant. By the analysis of linear and quadratic regression coefficient, filter ventilation of the cigarette was affected by porosity of plugwrap (5.87, -4.25), porosity of tip paper (5.68, -1.00), vent position (-3.87, 3.08), tobacco column pressure drop (2.56, 0.66), and filter pressure drop (1.50, 0.58) in the decreasing order. It should be emphasized that the major conclusion of this study was not that any particular parameter was linear or quadratic on any limit scale, but that there were highly significant relationships among factors involving linear, quadratic and their interaction and perhaps even linearity between and within factors. While, there is also quite strong evidence that vent position from mouth end and cigarette making materials are reverse relationship on this experimental model. On the basis of the result, it can be concluded that the porosity of the plug wrap and tipping paper has a marked effect on degree of filter ventilation rate. The F-value of plug wrap and tipping paper porosity among five factors were 39.2 and 36.8 respectively with P-value of 0.000 indicating higher significant for both factors. According to the analysis of variance, the model fitted for filter ventilation was significant at 5% confidence level and the coefficient of determination ($R^2$=0.84) was the proportion to variability in the data well fitted for by the model.

• Journal of Korean Society of Steel Construction
• /
• v.22 no.4
• /
• pp.325-334
• /
• 2010

In this study, an unbraced five-story steel-framed structure was designed in accordance with KBC2005 to understand the features of structural behavior for the arrangement of semi-rigid connections. A pushover analysis of the structural models was performed, wherein all the connections were idealized as fully rigid and semi-rigid. Additionally, horizontal and vertical arrangements of the semi-rigid connection were adopted for the models. A fiber model was utilized for the moment-curvature relationship of the steel beam and the column, and a three-parameter power model was adopted for the moment-rotation angle of the semi-rigid connection. The top displacement, base-shear force, required ductility for the connection, sequence of the plastic hinge, and design factors such as the overstrength factor, ductility factor, and response modification coefficient were investigated using the pushover analysis of a 2D structure subjected to the equivalent static lateral force of KBC2005. The partial arrangement of the semi-rigid connection was found to have secured higher strength and lateral stiffness than that of the A-Semi frame, and greater ductility than the A-Rigid frame. The TSD connection was found suitable for use for economy and safety in the sample structure.

• The Korean Journal of Food And Nutrition
• /
• v.26 no.4
• /
• pp.831-840
• /
• 2013

In this study, an attempt is being made to extract surface-active substances from defatted rapeseed cakes by supercritical carbon dioxide fluid. Independent variables for the extraction process, being formulated by D-optimal design, are pressure (150~350 bar), temperature ($33{\sim}65^{\circ}C$ and co-solvent (ethanol, 50~250 g). The dependent variables of the extraction yield, the content of neutral lipids, phospholipids and glycolipids in the extracts were analyzed upon the results through the response surface methodology. As for the extraction yield, it was found to increase with increasing independent variables, among which the co-solvent proved to be a major influencing parameter. Similar trends were found for the content of surface-active substances (i.e, phospholipids and glycolipids) in the extracts, except for the content of neutral lipids. Regression equations were suggested to coincide well with the results from the experiments. Extraction conditions are being optimized to maximize the extraction yields, the content of phospholipids, and glycolipids were 350 bar (pressure), $65^{\circ}C$ (temperature) and 228.55 g (co-solvent), respectively.