• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.6
• /
• pp.853-861
• /
• 2018

This paper proposes the application of harmonic constraints to address the problems caused by abnormal voltage increases when electric railway vehicles are running. The AC line that supplies the train with power during operation is used to provide electricity of 25kV/60 Hz, but gradually the size and frequency of harmonics involved in the line are varied with the technological evolution of the railroad vehicle electrical equipment. An increase in heat losses due to the failure of the instrument transformer (PT), the main circuit device, which is a serious problem with the recent train safety operation, or to the main displacement voltage. When high frequency components are introduced through low frequency Transformers of the main circuit device, the high intensity of the components is caused by the high intensity of the core and the current flow of the parasitic core is increased, thus generating heat. To solve this problem, the recent adjustment of the sequence has applied artificial NOTCH OFF of the power converter. However, the method of receiving and controlling the OFF signal operates by interaction between the ground and the vehicle's devices, thus it is invalid in the event of failure, and an actual accident is occurring. Therefore, the harmonic currents were required to prevent possible flow of harmonics, and conducted a study to prevent accidental occurrence of train accidents and to verify feasibility of the device through the simulations of the train's experimental analysis and the simulations of the train for safe operation.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.3 no.3
• /
• pp.231-239
• /
• 1998

This paper presents the dynamic analysis and control loop design of a zero voltage switching full bridge (ZVS-FB) PWM DC/DC converter. The small-signal model is derived incorporating the effects of phase shift control and the utilization of transformer leakage inductance and power FET junction capacitance to achieve zero voltage resonant switching. These effects are modeled by introducing additional feedforward and feedback terms for duty cycle modulation. Based on the results of the small-signal analysis, the control loop is designed using a simple two-pole one-zero compensation circuit. To show the validity of the design procedures, the small signal analysis of the closed loop system is carried out and the potential of the zero voltage switching and the superiority of the dynamic characteristics are verified through the experiment with a 2 kW prototype converter.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.185-187
• /
• 2000

Fault current is flowed into 154/23kV M. Tr when line-to-ground fault occurs in power system. NGR(Neutral Grounded Reactor) is set up in order to prevent M.Tr fault by limiting magnitude of fault currents. Here, disconnection of NGR causes voltage increase by L-C resonance and line-to-ground fault in an unearthed system results in voltage increase at healthy phases. So Over Voltage Ground Relay(OVGR) is used for tripping M.Tr. Also, buses at second phases of M.Trs are all connected with section circuit breakers closed for the purpose of parallel operation and load shedding. In case of speciality buses are comprised of power cable in part for GIS connection. When no-load charged cable or bus is open by a section CB, unbalanced voltage charged on the bus is induced. Also discrepant opening time for circuit breakers on different phases gives rise to unbalanced zero sequence voltage. It was observed that this zero sequence voltage detected in the 22.9kV P.T (Potential Transformer for bus) mal-operated 59GT and tripped M.Tr. The zero sequence voltage of which vanishing time is longer than relay operating time came out by EMTDC simulation. Also, it was shown that the voltage waves of actual test are similar to those of simulation. On the basis of above results, R-C circuit complement on the relay without any effect on a power system made operating time of the relay longer than vanishing time of distorted waves. Consequently, operating time of the relay was delayed and magnitude of distorted waves was decreased by increasing time constant of the relay.

• Fire Science and Engineering
• /
• v.29 no.6
• /
• pp.91-97
• /
• 2015

Skyscrapers, large business buildings, and IT consumers use many appliances, and the electrical power stems can cause fires by overheating. This can result in damaged capacitors, lost data, rising ground potential, and communication obstacles from linear or nonlinear high frequency. To make sure of that we investigated 7 spots of a building, among which 6-spots were fair but the other one needed high frequency control. Spots 3, 6, and 7 needed diagnostic workup, and spots 2, 3, and 5 considered 5 high frequency currents. A phase is all of good but the high frequency current is greater than the standard level except for spot 1. As a result, a zigzag transformer or active filter needs to be installed, and the efficiency needs to be upgraded by investigating load unbalance factors and power factors.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.37 no.2
• /
• pp.175-181
• /
• 2024

In electrical power substations, bulky iron-core potential transformers (PTs) are installed in a tank of gas-insulated switchgear (GIS) to measure system voltages. This paper proposed a low-power voltage transformer (LPVT) that can replace the conventional iron-core PTs in response to the demand for the digitalization of substations. The prototype LPVT consists of a capacitive voltage divider (CVD) which is embedded in a spacer and an impedance matching circuit using passive components. The CVD was fabricated with a flexible PCB to acquire enough insulation performance and withstand vibration and shock during operation. The performance of the LPVT was evaluated at 80%, 100%, and 120% of the rated voltage (38.1 kV) according to IEC 61869-11. An accuracy correction algorithm based on LabVIEW was applied to correct the voltage ratio and phase error. The corrected voltage ratio and phase error were +0.134% and +0.079 min., respectively, which satisfies the accuracy CL 0.2. In addition, the voltage ratio of LPVT was analyzed in ranges of -40~+40℃, and a temperature correction coefficient was applied to maintain the accuracy CL 0.2. By applying the LPVT proposed in this paper to the same rating GIS, it can be reduced the length per GIS bay by 11%, and the amount of SF6 by 5~7%.

• Structural Engineering and Mechanics
• /
• v.89 no.6
• /
• pp.589-599
• /
• 2024

This study presents the usability of the high-rate single-frequency Precise Point Positioning (SF-PPP) technique based on 20 Hz Global Positioning Systems (GPS)-only observations in detecting dynamic motions. SF-PPP solutions were obtained from post-mission and real-time GNSS corrections. These include the International GNSS Service (IGS)-Final, IGS real-time (RT), real-time MADOCA (Multi-GNSS Advanced Demonstration tool for Orbit and Clock Analysis), and real-time products from the Australian/New Zealand satellite-based augmentation systems (SBAS, known as SouthPAN). SF-PPP results were compared with LVDT (Linear Variable Differential Transformer) sensor and single-frequency relative positioning (SF-RP) solutions. The findings show that the SF-PPP technique successfully detects the harmonic motions, and the real-time products-based PPP solutions were as accurate as the final post-mission products. In the frequency domain, all GNSS-based methods evaluated in this contribution correctly detect the dominant frequency of short-term harmonic oscillations, while the differences in the amplitude values corresponding to the peak frequency do not exceed 1.1 mm. However, evaluations in the time domain show that SF-PPP needs high-pass filtering to detect accurate displacement since SF-PPP solutions include trends and low-frequency fluctuations, mainly due to atmospheric effects. Findings obtained in the time domain indicate that final, real-time, and MADOCA-based PPP results capture short-term dynamic behaviors with an accuracy ranging from 3.4 mm to 8.5 mm, and SBAS-based PPP solutions have several times higher RMSE values compared to other methods. However, after high-pass filtering, the accuracies obtained from PPP methods decreased to a few mm. The outcomes demonstrate the potential of the high-rate SF-PPP method to reliably monitor structural and earthquake-induced ground motions and vibration frequencies of structures.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.5
• /
• pp.61-67
• /
• 2024

In recent years, artificial intelligence and deep learning technologies have made significant advances, and the BERT model has been recognized for its excellent contextual understanding in natural language processing based on the transformer architecture. This performance has the potential to take traditional recommendation systems to the next level. In this study, we adopt an approach that combines a collaborative filtering approach with a deep learning model to improve the performance of recommendation systems. Specifically, we implemented a system that uses BERT to analyze the sentiment of user reviews and embed users based on these review sentiments to find and recommend users with similar tastes. In the process, we also utilized Elasticsearch, an open-source search engine, for quick search and retrieval of recommended results. The approach of analyzing users' textual data to increase the accuracy and personalization of recommendations will play an important role in improving the user experience on various online services in the future.

• Journal of The Korean Association For Science Education
• /
• v.43 no.3
• /
• pp.307-319
• /
• 2023

This study aims to explain the key concepts and principles of text-based generative artificial intelligence (AI) that has been receiving increasing interest and utilization, focusing on its application in science education. It also highlights the potential and limitations of utilizing generative AI in science education, providing insights for its implementation and research aspects. Recent advancements in generative AI, predominantly based on transformer models consisting of encoders and decoders, have shown remarkable progress through optimization of reinforcement learning and reward models using human feedback, as well as understanding context. Particularly, it can perform various functions such as writing, summarizing, keyword extraction, evaluation, and feedback based on the ability to understand various user questions and intents. It also offers practical utility in diagnosing learners and structuring educational content based on provided examples by educators. However, it is necessary to examine the concerns regarding the limitations of generative AI, including the potential for conveying inaccurate facts or knowledge, bias resulting from overconfidence, and uncertainties regarding its impact on user attitudes or emotions. Moreover, the responses provided by generative AI are probabilistic based on response data from many individuals, which raises concerns about limiting insightful and innovative thinking that may offer different perspectives or ideas. In light of these considerations, this study provides practical suggestions for the positive utilization of AI in science education.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.10
• /
• pp.2619-2627
• /
• 2009

The ground fault is occupying 70% among the total number of faults in ungrounded distribution power system. When the ground fault occurs in ungrounded system, the fault current is so small that it is hard to detect. But fault handling is very important because to continue power supply during fault conditions may cause the fault spreading and the distribution device in trouble. This paper presents the fault line detection method by using GPT signal detecting zero sequence voltage, and the fault section detection method by detecting whether GPT signal is disappeared or not during shifting normally open switch, which is connecting switch between distribution lines with open state in order to restore the outage area under emergency situation, and during isolating each section one by one which belongs to the fault line. This method is efficient because there is no whole power interruption during the fault section detection, and it is possible to perform both the fault section detection and the service restoration for the outage area at the same time, and it can apply to various distribution system configuration. Program for the fault restoration was developed applying proposed method, and it has been validated by applying to the pilot project of distribution automation system in Vietnam which has the ungrounded distribution system.