• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2002.07a
• /
• pp.135-138
• /
• 2002

VCO(Voltage Controlled Oscillator) is one of the main components governing the size, performance and power consumption of telecommunication devices. As the devices become much smaller, VCO need to have much smaller size with better characteristics. Buried type passive components of L,C,R were developed previously and the structure of these components are good for minimizing the size of VCO. Our own library of passive components is used in simulation and fabrication of VCO circuit, and surface mounted components like varactor diode are analysed using the measurement circuit designed by ourselves. Two-Dimensional simulation of VCO circuit and local three-Dimensional structure simulation are performed and their relation is obtained. In structure of multi-layered VCO, some components governing the characteristics of VCO are selected and placed on the top of oscillator for the good tuning process. In resonator part, the stripline structure and low loss glass/ceramic material are used to get higher Q value. In our research, a VCO oscillates in the 2.3∼2.36 GHz band is developed.

• BMB Reports
• /
• v.37 no.5
• /
• pp.533-537
• /
• 2004

Previously published kinetic data on the interactions of seventeen different enzymes with their physiological substrates are re-examined in order to understand the connection between ground state binding energy and transition state stabilization of the enzyme-catalyzed reactions. When the substrate ground state binding energies are normalized by the substrate molar volumes, binding of the substrate to the enzyme active site may be thought of as an energy concentration interaction; that is, binding of the substrate ground state brings in a certain concentration of energy. When kinetic data of the enzyme/substrate interactions are analyzed from this point of view, the following relationships are discovered: 1) smaller substrates possess more binding energy concentrations than do larger substrates with the effect dropping off exponentially, 2) larger enzymes (relative to substrate size) bind both the ground and transition states more tightly than smaller enzymes, and 3) high substrate ground state binding energy concentration is associated with greater reaction transition state stabilization. It is proposed that these observations are inconsistent with the conventional (Haldane) view of enzyme catalysis and are better reconciled with the shifting specificity model for enzyme catalysis.

• Korean Journal of Materials Research
• /
• v.27 no.2
• /
• pp.119-126
• /
• 2017

WC-CrC-Ni coatings were prepared by nine processes of the Taguchi program with three levels for the four spray parameters: spray distance, flow rates of hydrogen and oxygen, and powder feed rate. The optimal coating process (OCP) was oxygen flow rate of 38 FMR, hydrogen flow rate of 53 FMR, powder feed rate of 25 g/min, and spray distance of 7 inches. Hardness of 1150 Hv and porosity of 1.2 %, were obtained by OCP; these are better results compared with the highest 1033 Hv and the lowest 1.5 % porosity obtained by nine processes of the Taguchi program. Friction coefficient of the WC-CrC-Ni coating decreased from $0.36{\pm}0.07$ at $25^{\circ}C$ to $0.23{\pm}0.07$ at $450^{\circ}C$. These values were smaller than those of the EHC (electrolytic hard chrome) plating at both temperatures due to lubrication from the oxide debris. The wear trace and wear depth of the coating are smaller than those of the EHC at both temperatures. Pitting was not found in the WC-CrC-Ni coating sample, while it did appear in the EHC sample.

• Food Science of Animal Resources
• /
• v.32 no.3
• /
• pp.369-375
• /
• 2012

The current study was carried out to investigate the impact of extraction conditions on the free amino acid level and type in beef longissimus muscle. The sample blocks were chiller aged for 1 d and 7 d at $4^{\circ}C$. There are three homogenization speeds (11,000, 19,000 and 24,000 rpm) for bigger and two speeds (11,000 and 13,000 rpm) for smaller homogenizer's dispersing tools were used for evaluation. Results showed that chiller ageing greatly (p<0.05) increased extractable free amino acids, except cystine. Homogenization with the bigger dispersing tool at 24,000 rpm resulted in the highest free amino acid levels for both 1 and 7 d samples. Significant differences (p<0.05) in the mean values of most amino acids due to the effect of speed and interactions between ageing times. However, the speed effect and interaction between ageing with homogenization speed were not significant (p>0.05) for most of the amino acids except valine and isoleucine when using the smaller dispensing tools. The current data indicated that a standardized method for free amino acid types and levels of aged beef samples. In addition, the results also suggested that utilization of a big dispensing tool at high homogenization speed is a better condition for releasing free amino acid contents in beef samples.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.136-141
• /
• 2022

A very complex task in deep learning such as image classification must be solved with the help of neural networks and activation functions. The backpropagation algorithm advances backward from the output layer towards the input layer, the gradients often get smaller and smaller and approach zero which eventually leaves the weights of the initial or lower layers nearly unchanged, as a result, the gradient descent never converges to the optimum. We propose a two-factor non-saturating activation functions known as Bea-Mish for machine learning applications in deep neural networks. Our method uses two factors, beta (𝛽) and alpha (𝛼), to normalize the area below the boundary in the Mish activation function and we regard these elements as Bea. Bea-Mish provide a clear understanding of the behaviors and conditions governing this regularization term can lead to a more principled approach for constructing better performing activation functions. We evaluate Bea-Mish results against Mish and Swish activation functions in various models and data sets. Empirical results show that our approach (Bea-Mish) outperforms native Mish using SqueezeNet backbone with an average precision (AP50val) of 2.51% in CIFAR-10 and top-1accuracy in ResNet-50 on ImageNet-1k. shows an improvement of 1.20%.

• Journal of the Ergonomics Society of Korea
• /
• v.4 no.2
• /
• pp.17-24
• /
• 1985

To compare heart rate, $O_2$ uptake, $Vo_2$ ($O_2$ consumption), blood pressure (systolic, diastolic), reaction time, stability, flicker fusion value during 4 load levels with Rs (self-paced respiration) and Rp (paced respiration), 4 subjects participated in this experiment 1 hour/day, 6 days/week for 9 weeks. The cycle of Rp is 6 sec. (inspiration: 3 sec. & expiration: 3 sec.) Implications of the results are discussed in terms of the change in the physiological responses and human performance by the respiratory pattern. The results are as follows, 1. The changing magnitude of heart rate with Rp was larger than with Rs and the variance during load level 4 was significant. 2. The $Vo_2$ with Rp was smaller than with Rs and maximal $O_2$ uptake given load levels with Rp occurred and for two subjects, it significantly moved from low load level to high load level. 3. The changing magnitude of blood pressure was not consistent but the systolic pressure with Rp was smaller at rest than with Rs. 4. The score of reaction time test and stability test with Rp was better than with Rs.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.40 no.5
• /
• pp.894-909
• /
• 2016

This study analyzed a somatotype that was more suitable to a virtual fitting model and to improve the reality of a virtual model size. We analyzed 1,868 women 18-59 years old from the 6th Size Korea data. First, factor analysis was done for abstracting new criteria for dividing the somatotype; subsequently, we selected the waist height proportion to stature (body proportion) and drop (torso shape). Next, the cluster analysis was done with these criteria and 7 body proportion types and 11 torso shapes were distinguished. A virtual model size for the most common somatotype was also developed by a regression analysis of constituting sizes of each factor that was compared with body sizes well as with Clo's virtual model size. The model of this research showed a high similarity in sizes with body as well as improved better realisty than the Clo model which presented size problems such as longer limbs, bigger bust, smaller waist and a smaller arm circumference than the real body.

• Journal of the Microelectronics and Packaging Society
• /
• v.8 no.1
• /
• pp.61-64
• /
• 2001

VCO(Voltage Contolled Oscillator) is one of the main components governing the size, performance and power consumption of telecommunication devices. As the devices become much smaller, YCO need to hove much smaller size with better characteristics. Buried type passive components of L, C, R were developed previously and the structure of these components are good for minimizing the size of VCO. Our own library of passive components is used in simulation and measurement circuit designed by ourselve. In structure of multi-layered VCO, some components governing the characteristics of VCO are selected and placed on the top of oscilltor for the good tuning process. In resonator part, the stripline structure and low loss glass/ceramic material are used to get higher Q value. In our research, a VCO oscillates in the 2.3~2.36 GHz band is developed.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2000.11b
• /
• pp.213-220
• /
• 2000

Numerical modeling of cut slope has some limits in simulating the real slopes. In the case of 2D analysis of slope stability, it is assumed that slope is simply straight even when it is concave or convex in plan view. In this study, 3D analysis in curved shape slopes has been conducted for the comparison with 2D analysis in terms of failure mode and factor of safety. For this, 3D analysis by FLAC3D was compared with 2D analysis in plane strain condition and axi-symmetric model condition by FLAC. It was also observed how safety factors of slopes were affected by the variation of the tensile strength and cohesion, which are important variables to decide whether the slope fails or not. 2D analysis of concave slopes under plane strain condition showed much smaller safety factors by 16-40 % errors depending on the radius of curvature of slopes, compared to the more realistic values from 3D analysis. In case of convex slopes, the lower values by 7-10 % has been reported. 2D analysis of axi-symmetric model showed also smaller safety factors by 6-10 % and by 2-4 %, in case of concave and convex slopes, respectively. Such results are expected to contribute to the better understanding of failure process and could be applied for improved design of slopes.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.20 no.5
• /
• pp.559-563
• /
• 2011

Drop impact reliability assessment of solder joints on the flip chip is one of the critical issues for micro system packaging. Our previous researches have been showing that new solder ball compositions of Sn-3.0Ag-0.5Cu has better mechanical reliability than Sn-1.0Ag-0.5Cu. In this paper, dynamic reliability analysis using Finite Element Analysis (FEA) is carried out to assess the factors affecting flip chip in drop simulation. The design parameters are size and thickness of chip, and size, pitch and array of solder ball with composition of Sn1.0Ag0.5Cu. The board systems by JEDEC standard including 15 chips, solder balls and PCB are modeled with various design parameter combinations, and through these simulations, maximum yield stress and strain at each chip are shown at the solder balls. It is found that larger chip size, smaller chip array, smaller ball diameter, larger pitch, and larger chip thickness have bad effect on maximum yield stress and strain at solder ball of each chip.