• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2000.11a
• /
• pp.399-401
• /
• 2000

Recently polymer insulators are being used for outdoor high voltage applications. Polymer insulators for transmission and distribution line have significant advantages over porcelain and glass insulators, especially for ultra-high voltage(UHV) transmission lines. Their advantages are light weight, vandalism resistance and hydrophobicity. In this paper, polymer line post insulator has been designed and investigated electric field distribution by FEM. Designed LP insulators have been tested as insulator performance test, such as power frequency voltage test, lightning impulse voltage test, artificial pollution test and flexural load test.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2005.07a
• /
• pp.556-557
• /
• 2005

The distribution facilities, such as suspension insulator, line post insulator, lightning arrester, COS, used for long periods in the industrial pollution area were investigated. The electrical test and the material analyses were performed on the polluted and non-polluted facilities. The low frequency dry flashover voltage of polluted suspension was decreased about 8% in comparison with non-polluted one. The polluted materials turned out with the iron which came from the foundries. The polluted materials turned out with the iron which came from the foundries. This conductive materials decreased the leakage distance, resulting in reducing of electrical properties.

• KEPCO Journal on Electric Power and Energy
• /
• v.8 no.2
• /
• pp.137-141
• /
• 2022

To inspect the power facility faults using artificial intelligence, it need that improve the accuracy of the diagnostic model are required. Data augmentation skill using generative adversarial network (GAN) is one of the best ways to improve deep learning performance. GAN model can create realistic-looking fake images using two competitive learning networks such as discriminator and generator. In this study, we intend to verify the effectiveness of virtual data generation technology by including the fake image of power facility generated through GAN in the deep learning training set. The GAN-based fake image was created for damage of LP insulator, and ResNet based normal and defect classification model was developed to verify the effect. Through this, we analyzed the model accuracy according to the ratio of normal and defective training data.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.30A no.10
• /
• pp.33-40
• /
• 1993

A reflection type 16x8 S-SEED array from LP(Low Pressure)-MODVD-grown GaAs/AlGaAs extremely shallow quantum well(ESQW) structures, with 4% Al fraction, has been fabricated. Its intrinsic region consists of 50 pairs of alternating 100.angs. GaAs and 100.angs. $Al_{0.04}$Ga$_{0.96}$As layers. A multilayer reflector stack of $Al_{0.04}$/Ga$_{0.96}$ As(599$\AA$)/AlAs(723$\AA$) was incorporated for the reflection plane below the p-i-n structures. The device processing after the MOCVD growth includes the mesa etching, isolation etching, insulator deposition, p & n metallization, and AR(Anti-Reflection) coating. For switching characteristics of the S-SEED in the form of p-i-n ESQW diode, the maximum optical negative resistance was observed at 856nm. Reflectance measurements showed a change from 15.6% to 43.3% for +0.9V to -6V bias. The maximum contrast ration of the S-SEED array was 2.0 and all the 128 devices showed optical bistability with contrast ratios over 2.4 at 5V reverse bias.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2004.07b
• /
• pp.734-737
• /
• 2004

The metal-ferroelectric-semiconductor (MFS) structure is widely studied for nondestructive readout (NDRO) memory devices, but conventional MFS structure has a critical problem. It is difficult to obtain ferroelectric films like PZT on Si substrate without interdiffusion of impurities such as Pb, Ti and other elements. In order to solve these problems, the metal-ferroelectric-insulator-semiconductor (MFIS) structure has been proposed with a buffer layer of high dielectric constant such as MgO, $Y_2O_3$, and $CeO_2$. In this study, the etching characteristics (etch rate, selectivity) of MgO thin films were etched using $Cl_2/Ar$ plasma. The maximum etch rate of 85 nm/min for MgO thin films was obtained at $Cl_2$(30%)/Ar(70%) gas mixing ratio. Also, the etch rate was measured by varying the etching parameters such as ICP rf power, dc-bias voltage, and chamber pressure. Plasma diagnostics was performed by Langmuir probe (LP) and optical emission spectroscopy (OES).

• Proceedings of the KIEE Conference
• /
• 1994.07b
• /
• pp.1216-1219
• /
• 1994

A reflection type SEED from LP-MOCVD grown InGaAs/GaAs ESQW structures, with 5% In fraction, has been fabricated and its basic characteristics were investigated. Its intrinsic region consists of 50 pairs of alternating $100{\AA}$ $In_{0.05}Ga_{0.95}As$ barrier and $100{\AA}$ GaAs layers. And a multilayer reflector stack of $Al_{0.12}Ga_{0.88}As(641{\AA})-/AlAs(774{\AA})$ was vertically integrated below the p-i-n structures. The device processing includes the mesa etching, insulator deposition, indium metallization, and thermal alloy for Ohmic contact. Photocurrent spectrum measurement showed the exciton absorption peak at 905nm and availability as a optical switching device. This device showed a contrast ratio of 2:1 by the reflectance spectrum measurement.