• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.54 no.12
• /
• pp.553-557
• /
• 2005

This paper describes the application results of an ultrasonic detection system for the power transformer. The ultrasonic detection system with 6 sensors was applied to detect partial discharge in a 154kV transformer with a dangerous levels of $C_{2}H_{2},\;C_{2}H_4$ and $CH_{4}$ gases. The ultrasonic detection tests were carried out 2 times, respectively, to confirm the existence and location of the partial discharge in the transformer. As a result of internal inspection, the arc trace between the pressure ring and core due to the partial discharge was found at the estimated position based on the amplitude and arriving time of the ultrasonic signals. Therefore, it was verified that the ultrasonic detection system is effective as a preventive diagnosis method for the power transformer. Also, the reliability of the ultrasonic detection system in detecting partial discharges in the transformer was also confirmed. It is expected, therefore, that the ultrasonic detection system will have beneficial effects on applications and verifications in detecting partial discharges for the power transformer.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.10a
• /
• pp.533-534
• /
• 2010

Fiber optic sensor has bee widely used in the industrial applications. For the application of acoustic detection of the high voltage electric transformer Sagnac interferometer can be used. In this paper several different materials of mandrel were used in the fiber sensor array. Based on the transformer oil fiber optic sensor is more sensitive than in the air.

• Proceedings of the KIEE Conference
• /
• 1991.11a
• /
• pp.415-418
• /
• 1991

This paper described an instrument for the detection and geometric location of partial discharge(PD) sources in oil transformer. This instrument measures electric current pulse and ultrasonic pulse simultaneously in order to determine the geometric location of PD in transformer. Laboratory experiment systems are made for detection and location of PD in oil transformer. It was observed that there are effects of the barrier, such as insulation papers, silicon steel plate and actual transformer with location and detection of PD in model transformer. Through the laboratory actual test, it was clarified that this measurement device could be used satisfactorily for location of pd in oil transformer.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2005.07a
• /
• pp.574-574
• /
• 2005

To found out the degradation characteristic of transformer insulation, insulation material was depisited into transformer oil and heated. Due to the thermal stress which added to insulation, the density of carbon dioxide which included in transformer oil was mesured by using the gas density detection equipment of gas sensor and air circulation method. As a result, it didn't match with the transformer supervision standard. But it was found that as thermal stress increased, the density of carbon dioxide propertionally increased.

• 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.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.6
• /
• pp.319-327
• /
• 2022

Transformers are the most famous deep learning models that has achieved great success in natural language processing and also showed good performance on computer vision. In this survey, we categorized transformer-based models for computer vision, particularly object detection tasks and perform comprehensive comparative experiments to understand the characteristics of each model. Next, we evaluated the models subdivided into standard transformer, with key point attention, and adding attention with coordinates by performance comparison in terms of object detection accuracy and real-time performance. For performance comparison, we used two metrics: frame per second (FPS) and mean average precision (mAP). Finally, we confirmed the trends and relationships related to the detection and real-time performance of objects in several transformer models using various experiments.

• The Journal of Korea Robotics Society
• /
• v.19 no.3
• /
• pp.266-273
• /
• 2024

With the rising interaction between robots and humans, detecting collisions has become increasingly vital for ensuring safety. In this paper, we propose a novel approach for detecting collisions without using force torque sensors or tactile sensors, utilizing a Transformer-based neural network architecture. The proposed collision detection approach comprises a torque estimator network that predicts the joint torque in a free-motion state using Synchronous time-step encoding, and a collision discriminator network that predicts collisions by leveraging the difference between estimated and actual torques. The collision discriminator finally creates a binary tensor that predicts collisions frame by frame. In simulations, the proposed network exhibited enhanced collision detection performance relative to the other kinds of networks both in terms of prediction speed and accuracy. This underscores the benefits of using Transformer networks for collision detection tasks, where rapid decision-making is essential.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.195-204
• /
• 2024

In this paper, we investigate the comparative performance of two leading object detection architectures, YOLOv7 and Detection Transformer (DETR), across varying levels of data augmentation using CycleGAN. Our experiments focus on chest scan images within the context of biomedical informatics, specifically targeting the detection of abnormalities. The study reveals that YOLOv7 consistently outperforms DETR across all levels of augmented data, maintaining better performance even with 75% augmented data. Additionally, YOLOv7 demonstrates significantly faster convergence, requiring approximately 30 epochs compared to DETR's 300 epochs. These findings underscore the superiority of YOLOv7 for object detection tasks, especially in scenarios with limited data and when rapid convergence is essential. Our results provide valuable insights for researchers and practitioners in the field of computer vision, highlighting the effectiveness of YOLOv7 and the importance of data augmentation in improving model performance and efficiency.

• Structural Engineering and Mechanics
• /
• v.92 no.2
• /
• pp.121-131
• /
• 2024

This paper proposes a fault detection method for blade pitch systems of floating wind turbines using transformer-based deep-learning models. Transformers leverage self-attention mechanisms, efficiently process time-series data, and capture long-term dependencies more effectively than traditional recurrent neural networks (RNNs). The model was trained using normal operational data to detect anomalies through high reconstruction losses when encountering abnormal data. In this study, various fault conditions in a blade pitch system, including environmental load cases, were simulated using a detailed model of a spar-type floating wind turbine, the data collected from these simulations were used to train and test the transformer models. The model demonstrated superior fault-detection capabilities with high accuracy, precision, recall, and F1 scores. The results show that the proposed method successfully identifies faults and achieves high-performance metrics, outperforming existing traditional multi-layer perceptron (MLP) models and long short-term memory-autoencoder (LSTM-AE) models. This study highlights the potential of transformer models for real-time fault detection in wind turbines, contributing to more advanced condition-monitoring systems with minimal human intervention.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.2C no.6
• /
• pp.297-303
• /
• 2002

We have developed an on-line ultrasonic detector to monitor partial discharge in an operating transformer. The ultrasonic sensor has 150[KHz] resonance frequency and contains a pre-amplifier with 60[㏈] gain. The on-line ultrasonic detector has 50~300[KHz] frequency band-pass filter to remove electrical and mechanical noises from the transformer. This detector has an ultrasonic signal discrimination algorithm which discriminates ultrasonic signals due to partial discharge in a transformer. A moving average method of ultrasonic signal number was employed to effectively monitor the increasing trend of the partial discharge. This paper describes an experience of partial discharge detection in a 154[㎸] operating transformer using an ultrasonic detector. With regards to gas analysis in oil, C2H2 gas was produced with a warning level in this transformer We detected ultrasonic signals on the transformer steel wall, and estimated the position of partial discharge. With further inspection, we found carbonized marks due to partial discharge on the supporting bolt which fastens the windings.