• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.459-462
• /
• 2021

본 연구는 반도체 제조 공정중 발생하는 센서 데이터를 활용하여 딥러닝기반으로 불량을 예측하는 모델을 제안한다. 반도체 공장에서는 FDC((Fault Detection and Classification)라는 불량을 예측하는 시스템이 있지만, 공정의 복잡도가 높고 센서의 종류가 많아 공정 관리자가 모든 센서의 기준을 설정 및 관리하는데 한계가 있다. 이를 해결하기 위해 공정 설비의 센서 데이터를 딥러닝을 활용하여 학습시켜 센서 기준정보로 임계치를 제공하고, 가공중 발생하는 센서 데이터가 입력되면 정상 여부를 판정하는 모델을 제안한다.

• Convergence Security Journal
• /
• v.4 no.3
• /
• pp.1-8
• /
• 2004

Large distributed systems such as computational and data grids require that a substantial amount of monitoring data be collected for various tasks such as fault detection, performance analysis, performance tuning, performance prediction, security analysis and scheduling. to cope with this problem, they are needed network monitoring architecture which can collect various network characteristic and analyze network security state. In this paper, we suggest network performance and security analysis system based on network monitoring. The System suggest that users can see distance network state with tuning network parameters.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2015.08a
• /
• pp.132-132
• /
• 2015

Virtual metrology (VM) model based on plasma information (PI) parameter for C4F8 plasma-assisted oxide etching processes is developed to predict and monitor the process results such as an etching rate with improved performance. To apply fault detection and classification (FDC) or advanced process control (APC) models on to the real mass production lines efficiently, high performance VM model is certainly required and principal component regression (PCR) is preferred technique for VM modeling despite this method requires many number of data set to obtain statistically guaranteed accuracy. In this study, as an effective method to include the 'good information' representing parameter into the VM model, PI parameters are introduced and applied for the etch rate prediction. By the adoption of PI parameters of b-, q-factors and surface passivation parameters as PCs into the PCR based VM model, information about the reactions in the plasma volume, surface, and sheath regions can be efficiently included into the VM model; thus, the performance of VM is secured even for insufficient data set provided cases. For mass production data of 350 wafers, developed PI based VM (PI-VM) model was satisfied required prediction accuracy of industry in C4F8 plasma-assisted oxide etching process.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.2
• /
• pp.146-153
• /
• 2011

Model based machine condition diagnosis methods are generally using a normal and many failure models which need sufficient data to train the models. However, data, especially for failure modes of interest, is very hard to get in real applications. So their industrial applications are either severely limited or impossible when the failure models cannot be trained. In this paper, continuous hidden Markov model(CHMM) with only a normal model has been suggested as a very promising machine condition diagnosis method which can be easily used for industrial applications. Generally hidden Markov model also uses many pattern models to recognize specific patterns and the recognition results of CHMM show the likelihood trend of models. By observing this likelihood trend of a normal model, it is possible to detect failures. This method has been successively applied to arc weld defect diagnosis. The result shows CHMM's big potential as a machine condition monitoring method.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.11
• /
• pp.61-68
• /
• 2004

Recently, there has been an increasing demand for computer-vision based inspection and/or measurement system as a part of factory automation equipment. Existing manual inspection method can inspect only specific samples and has low measuring accuracy as well as it increases working time. Thus, in order to improve the objectivity and reproducibility, computer-aided analysis method is needed. In this paper, front and side profile inspection and/or data transfer system are developed using computer-vision during the inspection process on three kinds of pipes coming from a forming line. Straight line and circle are extracted from profiles obtained from vision using Laplace operator. To reduce inspection time, Hough Transform is used with clustering method for straight line detection and the center points and diameters of inner and outer circle are found to determine eccentricity and whether good or bad. Also, an inspection system has been built that each pipe's data and images of good/bad test are stored as files and transferred to the server so that the center can manage them.

• Journal of the Korea Safety Management & Science
• /
• v.23 no.3
• /
• pp.27-33
• /
• 2021

The purpose of this study is to present a novel indicator for analyzing machine failure based on its idle time and productivity. Existing machine repair plan was limited to machine experts from its manufacturing industries. This study evaluates the repair status of machines and extracts machines that need improvement. In this study, F-RPN was calculated using the etching process data provided by the 2018 PHM Data Challenge. Each S(S: Severity), O(O: Occurence), D(D: Detection) is divided into the idle time of the machine, the number of fault data, and the failure rate, respectively. The repair status of machine is quantified through the F-RPN calculated by multiplying S, O, and D. This study conducts a case study of machine in a semiconductor etching process. The process capability index has the disadvantage of not being able to divide the values outside the range. The performance of this index declines when the manufacturing process is under control, hereby introducing F-RPN to evaluate machine status that are difficult to distinguish by process capability index.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.506-515
• /
• 2023

This study is to find a method to learn engine sound after the start-up of a diesel generator installed in nuclear power plant with an unsupervised deep learning algorithm (CNN autoencoder) and a new method to predict the failure of a diesel generator using it. In order to learn the sound of a diesel generator with a deep learning algorithm, sound data recorded before and after the start-up of two diesel generators was used. The sound data of 20 min and 2 h were cut into 7 s, and the split sound was converted into a spectrogram image. 1200 and 7200 spectrogram images were created from sound data of 20 min and 2 h, respectively. Using two different deep learning algorithms (CNN autoencoder and binary classification), it was investigated whether the diesel generator post-start sounds were learned as normal. It was possible to accurately determine the post-start sounds as normal and the pre-start sounds as abnormal. It was also confirmed that the deep learning algorithm could detect the virtual abnormal sounds created by mixing the unusual sounds with the post-start sounds. This study showed that the unsupervised anomaly detection algorithm has a good accuracy increased about 3% with comparing to the binary classification algorithm.

• Structural Monitoring and Maintenance
• /
• v.1 no.3
• /
• pp.249-271
• /
• 2014

This paper presents a unique study of Structural Health Monitoring (SHM) for the maintenance decision making about a real life movable bridge. The mechanical components of movable bridges are maintained on a scheduled basis. However, it is desired to have a condition-based maintenance by taking advantage of SHM. The main objective is to track the operation of a gearbox and a rack-pinion/open gear assembly, which are critical parts of bascule type movable bridges. Maintenance needs that may lead to major damage to these components needs to be identified and diagnosed timely since an early detection of faults may help avoid unexpected bridge closures or costly repairs. The fault prediction of the gearbox and rack-pinion/open gear is carried out using two types of Artificial Neural Networks (ANNs): 1) Multi-Layer Perceptron Neural Networks (MLP-NNs) and 2) Fuzzy Neural Networks (FNNs). Monitoring data is collected during regular opening and closing of the bridge as well as during artificially induced reversible damage conditions. Several statistical parameters are extracted from the time-domain vibration signals as characteristic features to be fed to the ANNs for constructing the MLP-NNs and FNNs independently. The required training and testing sets are obtained by processing the acceleration data for both damaged and undamaged condition of the aforementioned mechanical components. The performances of the developed ANNs are first evaluated using unseen test sets. Second, the selected networks are used for long-term condition evaluation of the rack-pinion/open gear of the movable bridge. It is shown that the vibration monitoring data with selected statistical parameters and particular network architectures give successful results to predict the undamaged and damaged condition of the bridge. It is also observed that the MLP-NNs performed better than the FNNs in the presented case. The successful results indicate that ANNs are promising tools for maintenance monitoring of movable bridge components and it is also shown that the ANN results can be employed in simple approach for day-to-day operation and maintenance of movable bridges.

• Journal of the Korean Geophysical Society
• /
• v.2 no.3
• /
• pp.217-224
• /
• 1999

Although faults can be found by geological surveys, the surface traces of faults are not easily discovered by traditional geological surveys due to alluvia. In and around faults and fracture zones, the electrical resistivity appears to be lower than that of the surroundings due to the content of groundwater and clay minerals. Therefore, electrical resistivity surveys are effective to search buried faults and fracture zones. The dipole-dipole array electrical resistivity surveys, which could show the two dimensional subsurface electrical resistivity structure, were carried out in two areas, Yongdang-ri, Woongsang-eup, Yangsan-si, Kyungsangnam-do and Malbang-ri, Woedong-eup, Kyungju-si, Kyungsangpook-do. The one was next to the Dongrae Fault and the other near the Ulsan Fault was close to the region in which debatable quaternary fault traces had been found recently. From each measured data set, the electrical resistivity cross-section was obtained using the inversion program the reliability of which was analyzed using analytic solutions. A low resistivity zone was found in the inverted cross-section from the Yongdang-ri area survey data, and two low resistivity zones were found in that from the Malbang-ri area survey data. They were almost vertical and were 15∼20 m wide. Accounting the shape and the very low resistivity values of those zones (<100 Ωm)in the inverted section, they were interpreted as fracture zones although they should be proven by trenching. The reliability of the interpretation might be improved by adding some more parallel resistivity survey lines and interpreting the results in 3 and/or adding other geophysical survey.

• Economic and Environmental Geology
• /
• v.51 no.2
• /
• pp.185-201
• /
• 2018

Nuclear power plants(NPP) are constructed and operated to ensure safety against natural disasters and man-made disasters in all processes including site selection, site survey, design, construction, and operation. This paper will introduce a series of efforts conducted in Korea Hydro and Nuclear Power Co. Ltd., to assure the safety of nuclear power plant against earthquakes and other natural hazards. In particular, the present status of the earthquake, fault, and slope safety monitoring system for nuclear power plants is introduced. A earthquake observatory network for the NPP sites has been built up for nuclear safety and providing adequate seismic design standards for NPP sites by monitoring seismicity in and around NPPs since 1999. The Eupcheon Fault Monitoring System, composed of a strainmeter, seismometer, creepmeter, Global Positioning System, and groundwater meter, was installed to assess the safety of the Wolsung Nuclear Power Plant against earthquakes by monitoring the short- and long-term behavioral characteristics of the Eupcheon fault. Through the analysis of measured data, it was verified that the Eupcheon fault is a relatively stable fault that is not affected by earthquakes occurring around the southeastern part of the Korean peninsula. In addition, it was confirmed that the fault monitoring system could be very useful for seismic safety analysis and earthquake prediction study on the fault. K-SLOPE System for systematic slope monitoring was successfully developed for monitoring of the slope at nuclear power plants. Several kinds of monitoring devices including an inclinometer, tiltmeter, tension-wire, and precipitation gauge were installed on the NPP slope. A macro deformation analysis using terrestrial LiDAR (Light Detection And Ranging) was performed for overall slope deformation evaluation.