• Korean Journal of Optics and Photonics
• /
• v.17 no.3
• /
• pp.262-267
• /
• 2006

We proposed a PPLN fabrication setup that measures the voltage and current applied to $LiNbO_3$ in real time during application of a DC electric field. Because the duration for transferring a sufficient electron charge to $LiNbO_3$ increases, we are able to control the electron charge flow transferred to $LiNbO_3$ efficiently. We divided the domain inversion process of PPLN into 5 states: Nucleation (state 1), Spread of the domain inversion region under the electrode(state 2), Accumulation of the electron charge at the insulator/$LiNbO_3$ interface(state 3), Domain inversion under the insulator layer after breakdown(state 4), and Lowering the electric field applied to $LiNbO_3$ (state 5). We have found that the Threshold Point is essential for the domain inversion and that the domain inversion process must be stopped within state 3 for the optimum PPLN. Using these results, we could fabricate a stable and reproducible PPLN efficiently.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.20 no.6
• /
• pp.55-62
• /
• 2006

For the fault diagnosis of three-phase induction motors, we construct a experimental unit and then develop a diagnosis algorithm based on pattern recognition. The experimental unit consists of machinery module for induction motor drive and data acquisition module to obtain the fault signal. As the first step for diagnosis procedure, preprocessing is performed to make the acquired current simplified and normalized. To simplify the data, three-phase current is transformed into the magnitude of Concordia vector. As the next step, feature extraction is performed by kernel principal component analysis(KPCA) and linear discriminant analysis(LDA). Finally, we used the classifier based on radial basis function(RBF) network. To show the effectiveness, the proposed diagnostic system has been intensively tested with the various data acquired under different electrical and mechanical faults with varying load.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.16 no.5
• /
• pp.477-483
• /
• 2011

Existing switching system that is one of the ways which are used for DC/DC power converter is classified to hard-switching system and resonant-soft-switching system, generally. Hard-switching system is inefficient because the power loss of the switching element is large when it is been to trun on or turn off. And resonant-soft-switching system have the defect that need to add the another reactor and capacitor that make it expensive and huge. This paper suggest the ZVS Full-Bridge power converter contrcution of novel switching system for the overcoming these shortcomings. In Suggested soft-switching system, the front of buck converter at diode current, switch is changing on and off at the part of full-bridge converter's zero voltage part. as the result that is possible to be ZVS excepting the reactor and capacitor. also to verify the reasonability of the isolated ZVS full-bridge DC/DC converter as previously suggested, we produced the 500[W] level DC/DC converter and enforced the simulation for Psim, and then it able to conform the superiority of the DC/DC converter's efficient.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.18 no.6
• /
• pp.464-476
• /
• 1998

Determination of location, shape and size of a flaw from its eddy current testing signal is one of the fundamental issues in eddy current nondestructive evaluation of steam generator tubes. Here, we propose an approach to this problem; an inversion of eddy current flaw signal using neural networks trained by finite element model-based synthetic signatures. Total 216 eddy current signals from four different types of axisymmetric flaws in tubes are generated by finite element models of which the accuracy is experimentally validated. From each simulated signature, total 24 eddy current features are extracted and among them 13 features are finally selected for flaw characterization. Based on these features, probabilistic neural networks discriminate flaws into four different types according to the location and the shape, and successively back propagation neural networks determine the size parameters of the discriminated flaw.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.885-896
• /
• 2022

Motor failure in manufacturing plays an important role in future A/S and reliability. Motor failure is detected by measuring sound, current, and vibration. For the data used in this paper, the sound of the car's side mirror motor gear box was used. Motor sound consists of three classes. Sound data is input to the network model through a conversion process through MelSpectrogram. In this paper, various methods were applied, such as data augmentation to improve the performance of classifying fault motors and various methods according to class imbalance were applied resampling, reweighting adjustment, change of loss function and representation learning and classification into two stages. In addition, the curriculum learning method and self-space learning method were compared through a total of five network models such as Bidirectional LSTM Attention, Convolutional Recurrent Neural Network, Multi-Head Attention, Bidirectional Temporal Convolution Network, and Convolution Neural Network, and the optimal configuration was found for motor sound classification.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.130-139
• /
• 2024

Artificial intelligence models are being used to detect facility anomalies using physics data such as vibration, current, and temperature for predictive maintenance in the manufacturing industry. Since the types of facility anomalies, such as facility defects and failures, anomaly detection methods using autoencoder-based unsupervised learning models have been mainly applied. Normal or abnormal facility conditions can be effectively classified using the reconstruction error of the autoencoder, but there is a limit to identifying facility anomalies specifically. When facility anomalies such as unbalance, misalignment, and looseness occur, the facility vibration frequency shows a pattern different from the normal state in a specific frequency range. This paper presents an N-segmentation anomaly detection method that performs anomaly detection by dividing the entire vibration frequency range into N regions. Experiments on nine kinds of anomaly data with different frequencies and amplitudes using vibration data from a compressor showed better performance when N-segmentation was applied. The proposed method helps materialize them after detecting facility anomalies.

• Korean Journal of Soil Science and Fertilizer
• /
• v.7 no.2
• /
• pp.99-105
• /
• 1974

According to the content of sugar and starch of each positional leaf sheath and internode at heading and 4 weeks after it using IR667-Suwon 214 (high yielding var. having tropical Indica parantage) and Jinheung (local leading var. temperate Japonica) rice grown in various cultivation seasons the suitability of grouping into the high sugar type (sugar>starch), sugar, tendency (increasing tendency in sugar content), high starh type (starch>sugar) and starch tendency (increasing tendency in starch) in carbohydrate metabolism was reexamined as follows. 1. Sugar tendency appeared strongly in IR667 than Jinheung, internode than leaf sheath, late season cultivation than early season, 4 weeks after than heading and high temperature than low temperature. Thus at heading, leaf sheath and internode of Jinheung in early and late season cultivation were high starch type, and lower internode in early season cultiattion and leaf sheath and internode in late season for IR667 were high sugar type. In very late season all internodes of both varieties except 1st internode of Jinheung at heading were high starch type. At four week after heading all leaf sheaths except 1st and 4th one of Jinheung and all internodes were high sugar type. High sugar type was intensified 4 weeks after heading in leaf sheaths than in internodes of IR667 in early season and of both varieties in late season. 2. The upper three leaf sheaths and internodes seem to work in the same way for carbohydrate translocation. Among them upper ones showed sugar tendency at heading and starch tendency 4 weeks after heading and it was clear in Jinheung. 3. The later the cultivation season, the higher the carbohydrate content (sugar+starch), and such tendency was clear 4 weeks after heading and in IR667, suggesting teanslocation inhibition by low temperature. 4. Grain filling rate (weight increase per day) was more rapid in early season cultivation and IR667 took shorter days to reach maximum rate. 5. The later the cultivation season, the greater the percent contribution of carbohydrate before heading to yield and it was always greater in IR667, a leaf sheath type. 6. Sugar and starch ratio appears to be determined principally by metabolic characteristics of variety according to growth process and secondly but considerably by environmental factors.

• Restorative Dentistry and Endodontics
• /
• v.31 no.6
• /
• pp.452-459
• /
• 2006

Competition will usually develop between the opposing walls as the restorative resin shrinks during polymerization. Magnitude of this phenomenon may be depended upon cavity configuration and volume. The purpose of this sturdy was to evaluate the effect of cavity configuration and volume on microleakage of composite resin restoration that has margins on the enamel site only. The labial enamel of forty bovine teeth was ground using a model trimmer to expose a flat enamel surface. Four groups with cylindrical cavities were defined, according to volume and configuration factor(Depth x Diameter / C-factor) - Group I : 1.5 mm ${\times}$ 2.0 mm / 4.0, Group II : 1.5 mm ${\times}$ 6.0 mm / 2.0, Group III : 2.Omm ${\times}$ 1.72 mm / 5.62, Group IV : 2.0 mm ${\times}$ 5.23 mm / 2.54. After treating with fifth-generation one-bottle adhesive - BC Plus$^{TM}$ (Vericom, AnYang, Korea), cavities were bulk flted with microhybrid composite resin - Denfill$^{TM}$ (Vericom). Teeth were stored in distilled water for one day at room temperature and were finished and polished with Sof-Lex system. Specimens were thermocycled 500 times between 5$^{\circ}$C and 55$^{\circ}$C for 30 second at each temperature. Teeth were isolated with two layers of nail varnish except the restoration surface and 1 mm surrounding margins. Electrical conductivity (${\mu}$A) was recorded in distilled water by electrochemical method. Microleakage scores were compared and analyzed using two-way ANOVA at 95% level. The results were as follows: 1. Small cavity volume showed lower microleakage score than large one, however, there was no statistically significant difference. 2. There was no relationship between cavity configuration and microleakage. Factors of cavity configuration and volume did not affect on microleakage of resin restorations with enamel margins only.

• Economic and Environmental Geology
• /
• v.37 no.4
• /
• pp.401-411
• /
• 2004

The Earth's total magnetic field was calculated from on board TAM(Three-Axis Magnetometer) observations of KOMPSAT-1 satellite between June 19th and 21st, 2000. The TAM's telemetry data were transformed from ECI(Earth-Centered Inertial Frame) to ECEF(Earth-Centered Earth-Fixed Frame) and then to spherical coordination. Self-induced field from the satellite bus were removed by the symmetric nature of the magnetic field. The 2-D wavenumber correlation filtering and quadrant-swapping method were applied to eliminate the dynamic components and track-line noise. To test the validity of the TAM's geomagnetic field, ${\phi}$rsted satellite's magnetic model and IGRF2000 model were used for statistical comparison. The correlation coefficients between KOMPSAT-1/${\phi}$rsted and KOMPSAT-1/IGRF2000 models are 0.97 and 0.96, respectively. The global spherical harmonic coeffi-cient was then calculated from the KOMPSAT-1 data degree and order of up to 19 and compared with those from IGRF2000, $\phi$rsted, and CHAMP models. The KOMPSAT-1 model was found to be stable to degree & order of up to 5 and it can give new information for the low frequency components of the global geomagtic field.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.9
• /
• pp.652-661
• /
• 2020

Recently, the Korean government developed the RE3020 (renewable energy) policy to overcome environmental problems, such as fine dust, climate change, and large-scale PV systems interconnected with a distribution system. When a large-scale PV system is interconnected in the distribution system, however, a malfunction can occur, and the protection devices may not be operated because of the dividing effect depending on the magnitude and direction of fault current as well as connection types and location of the PV system. Therefore, this paper proposes a search algorithm for the malfunction pattern of protection devices based on various scenarios, when large-scale PV systems are operated and interconnected in a distribution system. This paper presents a malfunction mechanism of protection devices according to the installation locations of recloser (R/C). Furthermore, the modeling of a distribution system with large-scale PV systems was performed using Off-DAS S/W, and the malfunction patterns of protection devices were analyzed based on a range of scenarios. From the simulation results with the proposed model and algorithm for searching for protection devices, it was confirmed that they are useful and effective in identifying a malfunction phenomenon depending on the installation location of the R/C and connection type of PV system.