• The Journal of the Acoustical Society of Korea
• /
• v.21 no.5
• /
• pp.430-437
• /
• 2002

Having a variety of good characteristics against other pattern recognition techniques, the multilayer perceptron (MLP) has been widely used in speech recognition and speaker recognition. But, it is known that the error backpropagation (EBP) algorithm that MLP uses in learning has the defect that requires restricts long learning time, and it restricts severely the applications like speaker recognition and speaker adaptation requiring real time processing. Because the learning data for pattern recognition contain high redundancy, in order to increase the learning speed it is very effective to use the online-based learning methods, which update the weight vector of the MLP by the pattern. A typical online EBP algorithm applies the fixed learning rate for each update of the weight vector. Though a large amount of speedup with the online EBP can be obtained by choosing the appropriate fixed rate, firing the rate leads to the problem that the algorithm cannot respond effectively to different learning phases as the phases change and the number of patterns contributing to learning decreases. To solve this problem, this paper proposes a Changing rate and Omitting patterns in Instant Learning (COIL) method to apply the variable rate and the only patterns necessary to the learning phase when the phases come to change. In this paper, experimentations are conducted for speaker verification and speech recognition, and results are presented to verify the performance of the COIL.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.19 no.2
• /
• pp.111-116
• /
• 1983

In order to furnish the fundamental data for the domestic production of magnetic compass and the prescription of standardization about it in Korea, authors made the helmholtz coil and investigated the characteristics of them. Subsequently, the damping curves of T190 and T165 compasses in the helmholtz coil were measured and analyzed the performance of those compass. The results obtained are as follows; 1. The relation between electric current (I sub I) to flow in the helmholtz coils, that the thickness of coil is 1mm, diameter 1m, winding number 117, and intensity of magnetic field is presented as follows. Vertical magnetic force: Z(Gauss)=0.34+1.506 I sub(i) Horizontal magnetic force: H(Gauss)=0.183+1.506 I sub(i) 2. Period of T190 compass is longer than T165 compass in all horizontal magnetic force. In the amplitude, the former is larger than the latter above 0.08 Gauss, but this phenomenon is opposed to that below 0.08 Gauss. 3. As the porizontal magnetic force is intensive, period of magnetic compass is short, amplitude is large, and damping degree and damping factor are small. The time elapsed to the principal points of damping curve is proportional to the -0.65 power of the horizontal magnetic force.

• Journal of Advanced Marine Engineering and Technology
• /
• v.36 no.7
• /
• pp.869-875
• /
• 2012

In the past decades, extensive studies on convection heat transfer on internal flow have been conducted by using high specific surface area, by increasing heat transfer coefficient and swirl flow, and by improving the transport properties. In this study, we applied a tangential slot swirl generator to improve heat transfer in a horizontal circular copper tube. The Al-Mg particles (approximately $100{\mu}m$ to $130{\mu}m$) were employed for this experimental work. The copper tube was heated uniformly by winding a heating coil with a resistance of 9 ohm per meter for heat transfer. Using Al-Mg particles, experiments were performed in the Reynolds number range of 5,000 to 13,130, with and without swirl. Experimental data transfers or comparisons between Nusselt numbers with and without swirl along the test tube and Reynolds numbers are presented. The Nusselt number is improved by increasing Reynolds numbers or swirl intensities along the test tube.

• Journal of the Korean Society of Visualization
• /
• v.10 no.3
• /
• pp.16-20
• /
• 2012

In the past decades, extensive studies on convection heat transfer on internal flow have been conducted by using high specific surface area, by increasing heat transfer coefficient and swirl flow, and by improving the transport properties. In this study, we applied a tangential slot swirl generator to improve heat transfer in a horizontal circular copper tube. The Al-Mg particles (approximately $100{\mu}m$ to $130{\mu}m$) were employed for this experimental work. The copper tube was heated uniformly by winding a heating coil with a resistance of 9ohm per meter for heat transfer. Using Al-Mg particles, experiments were performed in the Reynolds number range of 5,000 to 13,130, with and without swirl. Experimental data transfers or comparisons between Nusselt numbers with and without swirl along the test tube and Reynolds numbers are presented. The Nusselt number is improved by increasing Reynolds numbers or swirl intensities along the test tube.

• Solar Energy
• /
• v.17 no.4
• /
• pp.57-66
• /
• 1997

The performance of gas-fired boilers were experimentally investigated. The heat exchanger used for the experiments consisted of circular tube with longitudinal fins for the gas side and a spiral coil for the water side. The burner was located at the top of the heat exchanger, and the burned gas flowed down to the exit. The experiments carried out for different water flow rates, the heat capacities of the boiler and the number of baffle. The thermal efficiency of the upward flow was higher than that of the downward flow of the water in the heat exchanger. As the boiler capacity increased, the thermal efficiency decreased. As the number of the baffles increase, the thermal efficiency increased and the increasing rate of the efficiency decreased. The gas-side overall heat transfer coefficient was independent of the flow rate of the water. The effect of the number of the baffles on the heat transfer coefficient was greater than that of the boiler capacity.

• Proceedings of the KIEE Conference
• /
• 2003.07b
• /
• pp.909-911
• /
• 2003

The almost pole transformers are constructed with tank, cover, clamp etc., that contains the insulation oil, core, coil, terminals, bus and the other accessories. If some fault current will be flown by some trouble or accident, interior pressure of the transformer shall be very quickly rise, and mechanical components or insulation oil from the transformer enclosure shall be propelled or dropped from the tank. For the prevention of the above accident, recently the pole transformers should be done 'the Design Tests for Fault Current Capability' according to ANSI C57.12.20(1997) There are two tests method in this standard, Test Number I with a high current arcing fault, without internal fusible elements, shall be conducted on rack enclosure with its minimum designed air space. Test Number II with an internal fusible element, shall be conducted on each enclosure diameter utilizing the internal fusible elements. KEPCO recently request to be done the 'Design Tests for Fault Current Capability' for pole transformers according to KEPCO's standard ES141-$533{\sim}545$, PS141-$482{\sim}518$ and RS141-$611{\sim}628$ that is same with Test Number I of ANSI C57.12.20.

• Progress in Superconductivity and Cryogenics
• /
• v.23 no.2
• /
• pp.24-27
• /
• 2021

This paper presents a design methodology of high-temperature superconducting (HTS) magnets. The magnet consists of several double pancake coils with a variety of wire width. This technique, named Multi-Width, is well known to make efficient use of the superconducting wire. It is common for design of high-temperature superconducting magnets to not only reduce wire consumption used, but also consider the homogeneity of the magnetic field. In this paper, we study a design method that efficiently reduces wire usage while considering magnetic field homogeneity. The design is carried out by calculating the critical current and the critical magnetic field according to the configuration of arranging the thickness of the wire to determine the number of windings. The width of wire comprising the magnet was set to 4 - 12 mm, and the number of double pancake coils was set to an even number to consist of top-down symmetry. To verify the validity of the design, we compared the progress of the design code with a complete enumeration survey. As a case study, we designed a magnet that generates a central magnetic field of 3 T or more in a 240 mm bore in diameter. Optimality can be evaluated by weighing wire consumption and field homogeneity according to the magnet's use or user preference.

• Applied Science and Convergence Technology
• /
• v.24 no.5
• /
• pp.136-141
• /
• 2015

We studied the compact anode design to develop 100 kW rotating anode X-ray tube with large focal spot 1.2 mm, small focal spot 0.6 mm and tube voltage 150 kV for large hospital digital radiography using computer thermal simulation. The larger thermal radiation effect in a high vacuum can reduce the temperature of anode so the method to increase the surface area of anode is investigated. The anode has the multi-tier shape at the back side of TZM body of anode and also bigger diameter of anode. The number of multi-tiers was varied from 6 to 15 and the diameter of anode was also varied from ${\Phi}74$ to ${\Phi}82$. From ANSYS transient thermal simulation result, we could obtain $1056.4^{\circ}C$ anode maximum temperature when applying 100 kW input power at 0.1 second on target focal track which is less than $1091^{\circ}C$ of the conventional 75 kW X-ray tube with reduced anode weight by 15.5% than the conventional anode. The compact anode of reduced anode weight is able to improve the unwanted noise when the rotor is rotating at high-speed and also reduce the rotational torque which the cost effective stator-coil is possible. It is believed that the anode with 15 ea multi-tiers using ${\Phi}82$ can satisfy with the specification of the anode heat capacity. From the results of this paper, it has been confirmed that the proposed compact anode can be used as the anode of 100 kW rotating anode X-ray tube for digital radiography.

• Journal of Sensor Science and Technology
• /
• v.16 no.1
• /
• pp.17-23
• /
• 2007

This paper presents a miniaturized turn-counting sensor (TCS) where the geomagnetism and high-rpm rotation of ammunition are used to detect the turn number of ammunition for applications to small-caliber turn-counting fuzes. The TCS, composed of cores and a coil, has a robust structure with no moving part for increasing the shock survivability in the gunfire environments of ${\sim}30,000$ g's. The TCS is designed on the basis of the simulation results of an electromagnetic analysis tool, $Maxwell^{(R)}$3D. In experimental study, the static TCS test using a solenoid-coil apparatus and the dynamic TCS test (firing test) have been made. The presented TCS has shown that the induction voltage of $6.5{\;}mV_{P-P}$ is generated at the magnetic flux density of 0.05 mT and the rotational velocity of 30,000 rpm. From the measured signal, the TCS has shown the SNR of 44.0 dB, the nonlinearity of 0.59 % and the frequency-normalized sensitivity of $0.26{\pm}0.01{\;}V/T{\cdot}Hz$ in the temperature range of $-30{\sim}+43^{\circ}C$. Firing test has shown that the TCS can be used as a turn-counting sensor for small-caliber ammunition, verifying the shock survivability of TCS in high-g environments.

• Proceedings of the Korea Association of Crystal Growth Conference
• /
• 1996.06a
• /
• pp.508-508
• /
• 1996

;The growth of films have considerable interest in the field of superlattice structured multi-layer epitaxy led to realization of new devices concepts. Molecular beam epitaxy (MBE) with in situ observation by reflection high-energy electron diffraction (RHEED) is a key technology for controlled layered growth on the atomic scale in oxide crystal thin films. Also, the combination of radical oxygen source and MBE will certainly accelerate the progress of applications of oxides. In this study, the growth process of single crystal films using by MBE method is discussed taking the oxide materials of Bi-Sr-Ca-Cu family. Oxidation was provided by a flux density of activated oxygen (oxygen radicals) from an rf-excited discharge. Generation of oxygen radicals is obtained in a specially designed radical sources with different types (coil and electrode types). Molecular oxygen was introduced into a quartz tube through a variable leak valve with mass flowmeter. Corresponding to the oxygen flow rate, the pressure of the system ranged from $1{\;}{\times}{\;}10^{-6}{\;}Torr{\;}to{\;}5{\;}{\times}{\;}10^{-5}$ Torr. The base pressure was $1{\;}{\times}{\;}10^{-10}$ Torr. The growth of Bi-oxides was achieved by coevaporation of metal elements and oxygen. In this way a Bi-oxide multilayer structure was prepared on a basal-plane MgO or $SrTiO_3$ substrate. The grown films compiled using RHEED patterns during and after the growth. Futher, the exact observation of oxygen radicals with MBE is an important technology for a approach of growth conditions on stoichiometry and perfection on the atomic scale in oxide. The oxidization degree, which is determined and controlled by the number of activated oxygen when using radical sources of two types, are utilized by voltage locked loop (VLL) method. Coil type is suitable for oxygen radical source than electrode type. The relationship between the flux of oxygen radical and the rf power or oxygen partial pressure estimated. The flux of radicals increases as the rf power increases, and indicates to the frequency change having the the value of about $2{\times}10^{14}{\;}atoms{\;}{\cdots}{\;}cm^{-2}{\;}{\cdots}{\;}S^{-I}$ when the oxygen flow rate of 2.0 seem and rf power 150 W.150 W.