• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.118.1-118
• /
• 2001

In this paper, we suggest a new distribution model for a single phase transformer which is different from the existing model which was modeled for only primary parts, but new distribution model is modeled for primary and secondary parts. Using this model, we simulate various faults of the transformer to know how the transfer function vary from the normal one, i.e., the trend of the variation of transfer function. As an another approach, we measure the voltage and current of a three phase transformer while various faults are made at the transformer. From the simulation of the model and experiment, we fine some trends of the variation of transfer function.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.14 no.12
• /
• pp.1032-1040
• /
• 2001

In order to apply the partial discharge measuring technique utilizing electrical pulse to the transformer at sites, this paper describes the measuring technique obtaining only the signals due to internal partial discharge in the transformer, but the noises due to external corona which has been a major problem so far. At first, partial discharge and corona noise were simultaneously generated in the model transformer by using needle-plane electrodes and rod-sphere electrodes out of it in a high voltage laboratory, respectively. It was verified that only the partial discharge signals in the transformer could be measured by removing the noise signals from the superposed signals of partial discharges and noises on the grounding wire of the model transformer. By application to a 345kV transformer in service, it was also confirmed that the partial discharge could be on-line monitored by removing the noise signals measured by the inductance sensor on the grounding wire of a 154kV lightning arrester from the superposed signals of internal partial discharge and external corona noise measured by bushing tap coupler of the transformer.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.65 no.2
• /
• pp.122-129
• /
• 2016

Current differential protection relaying with second harmonic restraint is the main protection for large capacity power transformer. PSCAD simulation program is widely used for modeling of dynamic varying transients phenomena. This paper deals with a power transformer model and transients analysis using PSCAD software to develop IED for power transformer. Simulation was carried out using a three phase 40MVA, 154/22.9kV, 60Hz, two-winding transformer with Y-Y connection used in actual fields. The paper analyzed transformer magnetizing inrush, external fault, and internal fault conditions with this model in the time domain. In addition, we performed an analysis in the frequency domain using FFT during several conditions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.112-114
• /
• 2021

In order to detect urban conditions, the number of means of transportation and traffic flow are essential factors to be identified. This paper improved the detection system capabilities shown in previous studies using the SwinTransformer model, which showed higher performance than existing convolutional neural networks, by learning various vehicle types using existing Mask R-CNN and introducing today's widely used transformer model to detect certain types of vehicles in urban aerial images.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.313-318
• /
• 2023

This paper proposes a neural network applying fine-tuning as a way to improve the performance of Video Classification based on Vision Transformer. Recently, the need for real-time video image analysis based on deep learning has emerged. Due to the characteristics of the existing CNN model used in Image Classification, it is difficult to analyze the association of consecutive frames. We want to find and solve the optimal model by comparing and analyzing the Vision Transformer and Non-local neural network models with the Attention mechanism. In addition, we propose an optimal fine-tuning neural network model by applying various methods of fine-tuning as a transfer learning method. The experiment trained the model with the UCF101 dataset and then verified the performance of the model by applying a transfer learning method to the UTA-RLDD dataset.

• KIEE International Transactions on Power Engineering
• /
• v.5A no.4
• /
• pp.385-389
• /
• 2005

Power transformers would appear to be simple. However, due to their nonlinear and frequency-dependent behaviors, they can be one of the most complex system components to model. It is imperative that the applied models be appropriate for the range of frequencies and excitation levels that the system experiences. Transformer modeling is not a mature field and newer improved models must be made available in ATP packages. Further, there is a lack of published guidance on recommended modeling approaches. And there is typically not enough detailed design or test information available to determine the parameters for a given model. The purpose of this paper is to develop improved transformer core models for ATP and parameter estimation methods that can efficiently utilize the limited available information such as factory test reports.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.22 no.2
• /
• pp.70-77
• /
• 2010

In order to improve the efficiency of a main transformer in a train, the optimal operation of a cooling system is necessary. For the development of optimal control algorithms of a cooling system, mathematical models of a main transformer cooling system were developed. These include static and dynamic models of a main transformer, an oil pump, an oil cooler, and a blower. Static models were used to find optimal oil temperatures of the inlet and the outlet of a transformer. Dynamic models were used to predict transient performances of control algorithms of a blower and an oil pump. Simulation results showed good predictions of the static and the dynamic behavior of a main transformer cooling system. Therefore, mathematical models developed in this study may be effectively used for the development of control algorithms of a main transformer cooling system.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.335-341
• /
• 2022

In this paper, we adopt a transformer that shows remarkable performance in natural language processing as an acoustic model of hybrid speech recognition. The transformer acoustic model uses attention structures to process sequential data and shows high performance with low computational cost. This paper proposes a method to improve the performance of transformer AM by applying each of the four algorithms of sequence discriminative training, a weighted finite-state transducer (wFST)-based learning used in the existing DNN-HMM model. In addition, compared to the Cross Entropy (CE) learning method, sequence discriminative method shows 5 % of the relative Word Error Rate (WER).

• Progress in Superconductivity
• /
• v.7 no.2
• /
• pp.162-166
• /
• 2006

We have proposed a 3 phase, 100 MVA, 154 kV class HTS transformer substituting for a 60 MVA conventional transformer. The power transformer of 154 kV class has a tertiary winding besides primary and secondary windings. So the HTS transformer should have the 3rd superconducting winding. In this paper, we designed conceptually the structure of the superconducting windings of a single phase 33 MVA transformer. The electrical characteristics of the HTS transformer such as % impedance and AC loss vary with the arrangement of the windings and gaps between windings. We analyzed the effects of the winding parameters, evaluated the cost of each design, and proposed a suitable HTS transformer model for future power distribution system.