• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.413-435
• /
• 2018

Although the accuracy of handwritten character recognition based on deep networks has been shown to be superior to that of the traditional method, the use of an overly deep network significantly increases time consumption during parameter training. For this reason, this paper took the training time and recognition accuracy into consideration and proposed a novel handwritten character recognition algorithm with newly designed network structure, which is based on an extended nonlinear kernel residual network. This network is a non-extremely deep network, and its main design is as follows:(1) Design of an unsupervised apriori algorithm for intra-class clustering, making the subsequent network training more pertinent; (2) presentation of an intermediate convolution model with a pre-processed width level of 2;(3) presentation of a composite residual structure that designs a multi-level quick link; and (4) addition of a Dropout layer after the parameter optimization. The algorithm shows superior results on MNIST and SVHN dataset, which are two character benchmark recognition datasets, and achieves better recognition accuracy and higher recognition efficiency than other deep structures with the same number of layers.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.20 no.4
• /
• pp.337-343
• /
• 2015

The multi-level inverters are highly efficient for high-power and medium-voltage AC driving applications, such as high-speed railway systems and renewable energy resources, because such inverters generate lower total harmonic distortion (THD) and electromagnetic interface (EMI). Lower switching stress occurs on switching devices compared with conventional two-level inverters. Depending on the multi-level inverter topology, the required components and number of switching devices are different, influencing the overall efficiency. Comparative studies of multi-level inverters based on loss analysis and output characteristic are necessary to apply multi-level inverters in high-power AC conversion systems. This paper proposes a theoretical loss analysis method based on piecewise linearization of characteristic curves of power semiconductor devices as well as loss analysis and output performance comparison of five-level neutral-point clamped, flying capacitor inverters, and high-level cascaded H-bridge multi-level inverters.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2003.09a
• /
• pp.565-569
• /
• 2003

One-year-long groundwater-level data have been collected from 18 wells in Cheon-an area. The result of barometric efficiency, autocorrelation, cross-correlation and statistical distribution evaluated from the measurement data shows that groundwater-level measurements from observation wells are the principal source of information about aquifer characteristics. Data from WA-2 has high barometric efficiency as well as steady decreasing auto-correlation coefficient, which means nonleaky confined aquifer, Most aquifers in this study show the unconfined properties so that barometric efficiencies are mostly low and the coefficients of cross-correlation between groundwater-level and precipitation are commonly high. This study showed that the long-term groundwater-level monitoring data without artificial stress such as pumping would give accurate information about aquifer characteristics.

• Journal of Environmental Science International
• /
• v.11 no.6
• /
• pp.577-582
• /
• 2002

Selective catalytic reduction and selective non-catalytic reduction processes are mainly used to treat nitrogen oxidants generated from fossil-fuel combustion. Especially, the selective non-catalytic reduction process can be operated more economical and designed more simply than the selective catalytic reduction. For this reason, many researchers carried out to increase the removal efficiency of nitrogen oxidants in the condition of low oxygen concentration by using the selective non-catalytic reduction process. However, this study was flue gas contained high oxygen concentration of 20(v/v%) with ammonia as a reducing agent. Moreover, it carried out experiment with many factors that are reaction temperature, retention time, initial NO concentration, NSR(normalized stoichiometric ratio). It was determined optimal operating conditions to improve NO removal efficiency with SNCR process. The De-NOx efficiency was increased with NSR, initial NO concentration and retention time increasement. This study has NO removal efficiency over 80% in the high oxygen concentration as well as low oxygen concentration. The injection of reducing agent may be considered for SNCR process and facility operation in 850$\^{C}$ of optimal condition.

• Proceedings of the KSME Conference
• /
• 2001.11b
• /
• pp.112-117
• /
• 2001

An experiment using high preheated and swirled air in the coaxial diffusion flame burner was carried out in order to decrease NOx emission and improve the thermal efficiency. $N_2$ gas was used for diluent and propane was utilized for fuel. Combustion using high preheated air has two remarkable characteristics ; (1) low NOx emission with increasing dilution level, (2) high thermal efficiency in the furnace. Also, swirled air can mix fuel and oxidizer well in condition of diffusion flme and maintain the stable combustion. The color of flame changes from yellow to blue green according to increasing the dilution level of mixture gas. NO emission decreased with increasing dilution level and the swirl number.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.10
• /
• pp.43-46
• /
• 2012

In this paper, high efficiency power amplifier is implemented with high gain amplifier. Two-stage amplifier using adaptive bias control circuit improve efficiency at low input power. Fixed bias circuit and adaptive bias circuit both have about 76 % efficiency at maximum power level. However amplifier using an adaptive bias control circuit has 70 % at 6 dBm input power level when the amplifier using fixed bias circuit has 50%. The proposed power amplifier using the adaptive bias control circuit can have high efficiency at lower power level.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.21 no.6
• /
• pp.946-953
• /
• 2012

When water shortage has been getting worse recently, toilet water-saving is one of the most effective way to solve the water shortage. Therefore, in this study, toilet flushing performance was investigated with VOF(Volume Of Fluid) model to obtain the basic design data for the development of high-efficiency toilets. The result showed that the shorter trapway, the longer the flushing duration time. However, the shorter trapway is found to have disadvantage in exhausting dirt due to relatively weak siphon and higher curvature. Since the remaining water in a trapway appears to help siphon, it is also an important factor in developing high-efficiency toilets. Although higher water level in the water tank shows better performance in flushing, lower water level is preferred to save water. Too low water level may cause dirt to clog the trapway due to relatively weak siphon.

• Journal of the Korean Home Economics Association
• /
• v.8 no.1
• /
• pp.198-215
• /
• 1970

The basic purpose of this study is to explore the household economic effects on home administration. The major goals of home administration is to promote mental and physical welfare of the family members. In order to realize mental and physical welfare, the stability and satisfaction is needed. And to promote stability and satisfaction, the efficiency must be accomplished and the efforts is required. So the four variables to measure home administration situation are as follows. 1) The degree of the stability 2) The degree of the efficiency 3) The degree of the efforts 4) The degree of the satisfaction. The factor of Household economic level is the size of the family's money income. Although size of income is not the bases for either standard of living or Socioeconomic status, it is closely related to both. To solve this problem. Some hypotheses are specified as follows. Hypotheses I The degree of the stability is different among the groups by income level. The group which is high income level will be more stable than low income level. Hypothesis II The degree of the efficiencies different among the groups by income level. The group which is high income level will be more efficient than low income level. Hypothesis III The degree of the efforts is different among the groups by income level. The group which is high income level will more effort than low income level. Hypothesis IV The degree of the satisfaction is different among the groups by income level. The group which is high income level will feel more satisfaction than low income level. In order to test those hypotheses random selection of 190 subjects from the four girls high schools in Seoul was made, and the parents of the students reported their home administration situation. The results of this study show that; 1) the group which is high income level revealed more stability than low income level. The differences of each group are revealed very significant. 2) The differences of the degree of the efficiency among the groups are unsignificant. 3) the group which is high income level revealed more effort than low income level. The differences of each group are revealed significant. 4) The group which is high income level revealed more satisfaction than low income level. The differences of each group are particularly significant. So the conclusion can be drawn from this study. Every group by income level revealed significant differences in home administration. So the household economy greatly effects on home administration.

• Nuclear Engineering and Technology
• /
• v.55 no.4
• /
• pp.1540-1554
• /
• 2023

To use nuclear energy sustainably, spent nuclear fuel, classified as high-level radioactive waste and inevitably discharged after electricity generation by nuclear power plants, must be managed safely and isolated from the human environment. In Korea, the land area is limited and the amount of high-level radioactive waste, including spent nuclear fuels to be disposed, is relatively large. Thus, it is particularly necessary to maximize disposal efficiency. In this study, a high-efficiency deep geological repository concept was developed to enhance disposal efficiency. To this end, design strategies and requirements for a high-efficiency deep geological repository system were established, and engineered barrier modules with a disposal canister for pressurized water reactor (PWR)-type and pressurized heavy water reactor type Canada deuterium uranium (CANDU) plants were developed. Thermal and structural stability assessments were conducted for the repository system; it was confirmed that the system was suitable for the established strategies and requirements. In addition, the results of the nuclear safety assessment showed that the radiological safety of the new system met the Korean safety standards for disposal of high-level radioactive waste in terms of radiological dose. To evaluate disposal efficiency in terms of the disposal area, the layout of the developed disposal areas was assessed in terms of thermal limits. The estimated disposal areas were 2.51 km² and 1.82 km² (existing repository system: 4.57 km²) and the excavated host rock volumes were 2.7 Mm³ and 2.0 Mm³ (existing repository system: 4.5 Mm³) for thermal limits of 100 ℃ and 130 ℃, respectively. These results indicated that the area and the excavated volume of the new repository system were reduced by 40-60% compared to the existing repository system. In addition, methods to further improve the efficiency were derived for the disposal area for deep geological disposal of spent nuclear fuel. The results of this study are expected to be useful in establishing a national high-level radioactive waste management policy, and for the design of a commercial deep geological repository system for spent nuclear fuels.

• Journal of Electrical Engineering and information Science
• /
• v.2 no.3
• /
• pp.123-131
• /
• 1997

Design of single stage AC/DC converter with high power factor for low power level applications is proposed. The proposed converter gives the good power factor correction, low line current harmonic distortions, and tight output voltage regulations. This converter also has a high efficiency by employing an active clamp method and synchronous rectifiers. To verify the performances of the proposed converter, a 90W-converter has been designed. The modelling of this proposed converter is power formed using an averaging technique and based on this model a detailed analysis is carried out. This prototype meets the IEC555-2 requirements satisfactorily with nearly unity power factor and high efficiency.