• Journal of the Korean Vacuum Society
• /
• v.16 no.2
• /
• pp.153-159
• /
• 2007

In-cryostat helium lines are under installation to transfer a cryogenic helium into cold components in KSTAR device. In KSTAR, three kinds of helium should be supplied into the cold components, which are supercritical helium Into superconduction(SC) magnet system, liquid helium into current lead system, and gas helium into thermal shields. Cryogenic helium lines consist of transfer lines outside the cryostat, in-cryostat helium lines, and electrical breaks. In-cryostat helium lines should be guaranteed of leak tightness for tong time operation at high internal helium pressure of 20 bar. We wrapped the helium line with multi-layer insulator(MLI) to reduce radiation heat and insulated the surface of the high potential part with prepreg tape. The electrical break was fabricated by brazing ceramic tube with stainless steel tube. To ensure the operation reliability at operation temperature, all the electrical break have been examined by the thermal cycle test at liquid nitrogen and by the hydraulic test at 30 bar. And additional surface insulation was prepared with prepreg tape to give structural safety. At present most of the in-cryostat helium lines have been installed and the final inspection test is progressing.

• 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 for Railway
• /
• v.20 no.2
• /
• pp.203-209
• /
• 2017

Research on condition monitoring and diagnosis of power facilities has been conducted to improve the safety and reliability of electric power supply. Although insulation diagnostic techniques for unit equipment such as gas-insulated switchgears and transformers have been developed rapidly, studies on monitoring of cables have only included aspects such as whether defects exist and partial discharge (PD) detection; other characteristics and features have not been discussed. Therefore, this paper dealt with PD characteristics against void sizes and positions, and with defect localization in XLPE cable. Four types of defects with different sizes and positions were simulated and PD pulses were detected using a high frequency current transformer (HFCT) with a frequency range of 150kHz~30MHz. The results showed that the apparent charge increased when the defect was adjacent to the conductor; the pulse count in the negative half of the applied voltage was about 20% higher than that in the positive half. In addition, the defect location was calculated by time-domain reflectometry (TDR) method, it was revealed that the defect could be localized with an error of less than1m in a 50m cable.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.31 no.5
• /
• pp.307-312
• /
• 2018

Liquid phases in ZnO varistors cause more complex phase development and microstructure, which makes the control of electrical properties and reliability more difficult. Therefore, we have investigated 2 mol% $CaCO_3$ doped $ZnO-Co_3O_4-Cr_2O_3-La_2O_3$ (ZCCLCa) bulk ceramics as one of the compositions without liquid phase sintering additive. The results were as follows: when $CaCO_3$ is added to ZCCLCa ($644{\Omega}cm$) acting as a simple ohmic resistor, CaO does not form a secondary phase with ZnO but is mostly distributed in the grain boundary and has excellent varistor characteristics (high nonlinear coefficient ${\alpha}=78$, low leakage current of $0.06{\mu}A/cm^2$, and high insulation resistance of $1{\times}10^{11}{\Omega}cm$). The main defects $Zn_i^{{\cdot}{\cdot}}$ (AS: 0.16 eV, IS & MS: 0.20 eV) and $V_o^{\bullet}$ (AS: 0.29 eV, IS & MS: 0.37 eV) were found, and the grain boundaries had 1.1 eV with electrically single grain boundary. The resistance of each defect and grain boundary decreases exponentially with increasing the measurement temperature. However, the capacitance (0.2 nF) of the grain boundary was ~1/10 lower than that of the two defects (~3.8 nF, ~2.2 nF) and showed a tendency to decrease as the measurement temperature increased. Therefore, ZCCLCa varistors have high sintering temperature of $1,200^{\circ}C$ due to lack of liquid phase additives, but excellent varistor characteristics are exhibited, which means ZCCLCa is a good candidate for realizing chip type or disc type commercial varistor products with excellent performance.

• Transactions of the Korean hydrogen and new energy society
• /
• v.10 no.1
• /
• pp.59-69
• /
• 1999

AlN thin films were fabricated by reactive sputtering for the application of MIS devices with Al/AlN/Si structure. It has investigated the surface morphology change, I-V characteristics, C-V characteristics, and chemical composition of AlN films with the intriducing time of hydrogen on the fixed deposition condition(RF power: 150W, sputtering pressure: 5mTorr, flow rate ratio of $Ar/N_2=1$, hydrogen concentration: 5%). By addition of the hydrogen the deposition rate decreased drastically whereas the surface morphology changed little. It has been found from the analysis of I-V and C-V characteristics curves that the films deposited with hydrogen addition in initial stage had lower leakage current density, lower flat band voltage and hystersis profile when compared with those with hydrogen addition in last stage. The oxygen concentration in AlN films decreased with addition of hydrogen gas, which suggesting a profitable role in the insulation and C-V characteristics of AlN films.

• Journal of the Korean Vacuum Society
• /
• v.22 no.3
• /
• pp.131-137
• /
• 2013

The thin insulator films for organic memory device were made by the plasma polymerization method using the styrene monomer which was not the wet process but the dry process. For the formation of stable plasma, we make an effort for controlling the monomer with bubbler and circulator system. The thickness of plasma polymerized styrene insulator layer was 430 nm, the thickness of the Au memory layer was 7 nm thickness of plasma polymerized styrene tunneling layer was 30, 60 nm, the thickness of pentacene active layer was 40 nm, the thickness of source and drain electrodes were 50 nm. The I-V characteristics of fabricated memory device got the hysteresis voltage of 45 V at 40/-40 V double sweep measuring conditions. If it compared with the results of previous paper which was the organic memory with the plasma polymerized MMA insulation thin film, this result was greater than 18 V, the improving ratio is 60%. From the paper, styrene indicated a good charge trapping characteristics better than MMA. In the future, we expect to make the organic memory device with plasma polymerized styrene as the memory thin film.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.662-668
• /
• 2018

Most electric rooms are located in the underground spaces of buildings. When a fire occurs in electrical equipment, the fire expands to cable insulation material, resulting in toxic smoke and combustion products. If the smoke and combustion products quickly move vertically and horizontally, the evacuation of occupants and firefighting activities will be hindered. Therefore, it is necessary to design optimal equipment for smoke control in cases of fires in electric rooms. This study analyzes the characteristics of smoke and combustion products in fires in a cubicle-type switchboard in an electric room using PyroSim, which is based on the program Fire Dynamics Simulator (FDS). The fire modeling consists of four scenarios according to the operation mode of the mechanical ventilation equipment, the amount of air supply and exhaust, and the location of the air supply slot. The analysis shows that the mechanical ventilation equipment improves the smoke density, visibility, carbon monoxide concentration, and temperature characteristics. The visibility and temperature characteristics were improved when the air flow rate and the location of the air supply slot from fire defense regulations were applied.

• The Korean Journal of Physiology
• /
• v.23 no.1
• /
• pp.139-149
• /
• 1989

Although tail flick reflex (TFR) in rats has been used as a classic model of the nociceptive test to evaluate the action of analgesics, there have been few studies on the origin of the latent period of TFR. Present study was performed to elucidate the mechanism of increase in latency of TFR by morphine in anesthetized rats. Tail skin and dorsolateral tail nerve were stimulated electrically and EMG activities were recorded from abductor caudae dorsalis muscle participating in tail flick reflex. In the case of noxious radiant heat stimulation to tail, the tail flick tension was recorded before and after administration of morphine. Then changes in latency and conduction velocity of peripheral nerve were evaluated. The results obtained were as follows: 1) The latencies of TFR evoked by the electrical stimulation of tail skin and dorsolateral tail nerve were all within 40 ms and were elongated by several milliseconds from control after the administration of morphine. Peripheral conduction velocities of tail flick afferent nerve were within the range of 10-25 m/s. 2) The conduction velocity of peripheral nerve was significantly reduced after morphine administration, therefore the afferent time (utilization time+conduction time to spinal cord) was significantly increased. But the time for central delay and efferent time was not affected by morphine. 3) The conduction velocity under room temperature $(20-25^{\circ}C)$ was significantly reduced after morphine while that under vasodilation state $(40{\sim}42^{\circ}C)$ increased, 30 min and 45 min after morphine. The conduction velocity under vasodilation state without treatment of morphine increased continuously 4) The latency in tension response of TFR evoked by electrical stimulation was elongated by several milliseconds from control while the latency evoked by noxious radiant heat was elongated by several seconds compared with that of control. From the above results, it could be concluded that: 1) the increased latency of TFR evoked by electrical stimulation of the tail after morphine administration was due to the reducton in conduction velocity of peripheral nerve, which was the secondry effect of morphine on the peripheral vasomotion and 2) increased latency of TFR evoked by noxious radiant heat was also due to the same effect of morphine and the increase in cutaneous insulation to the noxious heat.

• Fire Science and Engineering
• /
• v.24 no.6
• /
• pp.7-12
• /
• 2010

This paper is aimed the measurement and estimation the thermal interrelation based on leakage current. The leakage models are analyzed by parameters in open and closed (Panel) states and with wood dust and rainwater, charcoal powder in leakage current occurred point. The thermal characteristic of open state that leakage current is inputted in 90 mA is higher in charcoal powder ($105^{\circ}C$) than in wood dust and rainwater ($88.8^{\circ}C$). Charcoal powder that is inputted in 110 mA shows high thermal characteristic ($238.6^{\circ}C$) and the thermal ($238.6^{\circ}C$) is caused by material strain of electrical wire insulation and Panel. The closed state in 90 mA is higher in charcoal powder ($250.6^{\circ}C$) than in wood dust and rainwater ($90.8^{\circ}C$). The open and closed state appear thermal characteristics of two times in charcoal powder than in wood dust and rainwater. The thermal data that are open and close states are used electrical fire investigation and thermal characteristic for Leakage current.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.11a
• /
• pp.429-429
• /
• 2007

For various applications of liquid crystal displays (LCDs), the uniform alignment of liquid crystal (LC) molecules on treated surfaces is significantly important. Generally, a rubbing method has been widely used to align the LC molecules on polyimide (PI) surfaces. Rubbed PI surfaces have suitable characteristics, such as uniform alignment. However, the rubbing method has some drawbacks, such as the generation of electrostatic charges and the creation of contaminating particles. Thus, we strongly recommend a non contact alignment technique for future generations of large high-resolution LCDs. Most recently, the LC aligning capabilities achieved by ultraviolet and ion-beam exposures which are non contact methods, on diamond-like carbon (DLC) inorganic thin film layers have been successfully studied because DLC thin films have a high mechanical hardness, a high electrical resistivity, optical transparency, and chemical inertness. In addition, nitrogen-doped DLC (NDLC) thin films exhibit properties similar to those of the DLC thin films and a higher thermal stability than the DLC thin films because C:N bonding in the NDLC thin filmsis stronger against thermal stress than C:H bonding in the DLC thin films. Our research group has already studied the NDLC thin films by an ion-beam alignment method. The $SiN_x$ thin films deposited by plasma-enhanced chemical vapor deposition are widely used as an insulation layer for a thin film transistor, which has characteristics similar to those of DLC inorganic thin films. Therefore, in this paper, we report on LC alignment effects and pretilt angle generation on a $SiN_x$, thin film treated by ion-beam irradiation for various N ratios