http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
도메인 적대적 신경망 기반 Overhead Hoist Transport의 호기별 적응 기법을 활용한 고장 진단
서채현(Chaehyun Suh),박찬희(Chan Hee Park),김형민(Hyeongmin Kim),윤병동(Byeng D. Youn) 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.12
Overhead Hoist Transports (OHT) transport wafers to ensure continuous manufacturing process in semiconductor factories. They are subjected to high load, making them vulnerable to faults. When any of individual unit breaks down, whole manufacturing process is affected consequently. Thus, fault detection of OHT is essential for high productivity. Torque signal has been used for fault detection of automated material handling system including OHT. However, discrepancies among of torque signals from different OHT units exist. Furthermore, there are unlabeled data in some OHT units. These facts make generalization of fault detection algorithm across OHT units difficult. Thus in this study, we utilize domain adversarial neural network in semi supervised manner to transform feature space for unit-wise adaptation of fault detection algorithm. Pre-processing techniques for torque signals are also proposed that result in 3-channel input for deep neural network. The proposed method’s performance was validated using data from real industrial production line.
MPARN: multi-scale path attention residual network for fault diagnosis of rotating machines
김형민,Park Chan Hee,Suh Chaehyun,채민석,Yoon Heonjun,Youn Byeng D. 한국CDE학회 2023 Journal of computational design and engineering Vol.10 No.2
Multi-scale convolutional neural network structures consisting of parallel convolution paths with different kernel sizes have been developed to extract features from multiple temporal scales and applied for fault diagnosis of rotating machines. However, when the extracted features are used to the same extent regardless of the temporal scale inside the network, good diagnostic performance may not be guaranteed due to the influence of the features of certain temporal scale less related to faults. Considering this issue, this paper presents a novel architecture called a multi-scale path attention residual network to further enhance the feature representational ability of a multi-scale structure. Multi-scale path attention residual network adopts a path attention module after a multi-scale dilated convolution layer, assigning different weights to features from different convolution paths. In addition, the network is composed of a stacked multi-scale attention residual block structure to continuously extract meaningful multi-scale characteristics and relationships between scales. The effectiveness of the proposed method is verified by examining its application to a helical gearbox vibration dataset and a permanent magnet synchronous motor current dataset. The results show that the proposed multi-scale path attention residual network can improve the feature learning ability of the multi-scale structure and achieve better fault diagnosis performance.
가상세계 메타버스 이용 의도에 영향을 미치는 요인에 관한 연구
박천호(Cheon-Ho Park),이채현(Chaehyun Lee),정성미(Sung Mi Jung),최정일(Jeong-Il Choi) 한국IT서비스학회 2023 한국IT서비스학회지 Vol.22 No.2
Metaverse is a three-dimensional virtual space where virtual and reality interact and co-evolutionize, and social, cultural, and economic activities are carried out in it to create value. Among the types of metaverse, the virtual world metaverse is expected to bring innovation beyond time and space in all areas of industry and society following the Internet. This study empirically analyzed factors affecting consumer’s intention to use the virtual world metaverse by an innovation diffusion perspective. It was analyzed as an expanded technology acceptance model by setting relative advantages, ease of use, and visibility among the attributes of the innovation diffusion, and social presence, telepresence, and interactivity among the characteristics of the virtual worlds as factors. The proposed research model and hypothesis were verified through a PLS structural equation analysis based on a survey of 216 people. Studies have shown that relative advantage, telepresence, and interactivity have a significant effect on perceived usefulness and perceived enjoyment, but visibility does not have a significant effect on perceived usefulness and perceived enjoyment. It was found that ease of use had a positive effect on perceived enjoyment, social presence had a positive effect on perceived usefulness, and perceived usefulness and perceived enjoyment had a positive effect on the intention to use, respectively. This study is meaningful in that it empirically analyzed the factors affecting the intention to use the virtual world metaverse by dividing the psychological characteristics that induce consumer behavior into perceived usefulness which is an extrinsic motivation and perceived enjoyment which is an intrinsic motivation.
가변운행조건 하 영구자석 동기전동기의 딥러닝 기반 고장진단을 위한 전류 신호 전처리 기법
김형민(Hyeongmin Kim),박찬희(Chan Hee Park),서채현(Chaehyun Suh),윤병동(Byeng D. youn) 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.12
This paper proposes a preprocessing method for faults diagnosis method of a permanent magnet synchronous motor (PMSM) under various operating conditions. Mechanical & electrical faults of PMSM leads to amplitude and phase modulation of the current signal. However, variations from operating conditions also change the amplitude and phase of the current signal, which make it hard to find fault modulated component of the signal. To solve this problem, current signal is resampled to its angular domain based on its estimated angle, and spectral components of the resampled signal are used as an input of deep learning algorithm to test the relationships between spectral components. To validate the performance of the proposed method, normal and two different fault motors with different operating conditions were tested, and performance between preprocessed data and without preprocessed data were compared. The result shows the generalization capability of the proposed method.
Yukgunja-tang for Irritable Bowel Syndrome: A Protocol for a Systematic Review and Meta-Analysis
Kangwook Lee,Seok-Jae Ko,Minjeong Kim,Chaehyun Park,Min-Seok Cho,Jae-Woo Park The Society of Internal Korean Medicine 2023 大韓韓方內科學會誌 Vol.44 No.3
Background: Irritable bowel syndrome (IBS) is a digestive disorder characterized by abdominal discomfort or pain accompanied by a change in stool condition. Owing to its complicated mechanisms, a standard treatment for IBS has not yet been established. Yukgunja-tang (YGT) is a Korean herbal medicine known in Asia to be effective in the treatment of gastrointestinal symptoms. In this study, we will conduct a systematic review of randomized controlled trials (RCTs) to assess the efficacy and safety of YGT in IBS treatment. Methods and analysis: English databases, such as Embase, Medline (via PubMed), Allied and Complementary Medicine Database, and Cochrane Central Register of Controlled Trials, will be searched for articles published up to April 2023. Additional databases, such as five Korean, one Chinese, and one Japanese database, will be included. RCTs and quasi-RCTs will also be included in the assessment of the efficacy of YGT. The overall efficacy rate will be the primary outcome, and data such as IBS quality-of-life measurements, global symptom scores, and adverse events will be the secondary outcomes. Review Manager Version 5.3 will be used for evaluation, and the risk of bias (RoB) will be evaluated using Cochrane Collaboration's RoB tool. The Grading of Recommendations Assessment, Development, and Evaluation approach will be used to score the quality of evidence. Conclusion: This study will demonstrate the efficacy and safety of YGT for treating patients with IBS.