• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.8
• /
• pp.216-223
• /
• 2019

Fault detection during production processes is one of the required operational tasks to run production processes both safely and consistently. Unexpected operational events or undetected process faults can have a serious impact on the production systems and subsequently on the final products' quality. In addition, such situations may lead to malfunctions or breakdowns of production processes. To reliably detect such abnormalities, a new one-class classification-based detection scheme has recently been developed The proposed method consists of four steps:1) noise filtering, 2) feature selection, 3) nonlinear representation and 4) outlier detection. The performance of the proposed scheme was demonstrated using the multivariate data obtained from a simulation process. The results have shown that the proposed method produced reliable monitoring results and outperforms any existing methods with an average improvement of 25.4%. The use of proper feature selection in the proposed framework yielded better detection performance.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.4
• /
• pp.160-167
• /
• 2021

Seabed deformation due to the fault failure have both a spatial variation and temporal history. When the faulting process initiates at a certain point beneath seabed, the failure spreads out to neighboring points, resulting in temporal changes of deformation. In particular, such a process induces tsunami waves from the vertical motion of seabed. The uprising speed of seabed affects the formation of initial surface profile, eventually altering the arrival time and runup of tsunamis at the coast. In this work, we developed a numerical model that can simulate the generation and propagation of tsunami waves by considering the horizontal and vertical changes of seabed in an active and dynamic manner. For the verification of the model, it was applied to the 2011 Tohoku-oki earthquake in Japan and the results confirmed that the accuracy was improved compared to the existing passive and static model.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.111-116
• /
• 2022

The causes of motor burnout include overload, phase loss, restraint, interlayer short circuit, winding ground fault, instantaneous overvoltage, and the rotor contacting the stator, leading to insulation breakdown, leading to breakdown or electrical accidents. Therefore, equipment failure causes not only loss due to cost required for equipment maintenance/repair, but also huge economic loss due to productivity decrease due to process stop because the process itself including the motor is stopped. The current level of technology for diagnosing motor failures uses vibration, heat, and power analysis methods, but there is a limit to analyzing the problems only after a considerable amount of time has passed according to the failure. Therefore, in this paper, a device and algorithm for measuring insulation resistance using DC AMP signal was applied to an industrial motor to solve this problem. And by following the insulation resistance state value, we propose a diagnosis of deterioration and failure of the motor that cannot be solved by the existing method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.36-42
• /
• 2021

With recent developments in the Fourth Industrial Revolution, the manufacturing industry has changed rapidly. Through key aspects of Fourth Industrial Revolution super-connections and super-intelligence, machine learning will be able to make fault predictions during the foam-making process. Polyol and isocyanate are components in polyurethane foam. There has been a lot of research that could affect the characteristics of the products, depending on the specific mixture ratio and temperature. Based on these characteristics, this study collects data from each factor during the foam-making process and applies them to machine learning in order to predict faults. The algorithms used in machine learning are the decision tree, kNN, and an ensemble algorithm, and these algorithms learn from 5,147 cases. Based on 1,000 pieces of data for validation, the learning results show up to 98.5% accuracy using the ensemble algorithm. Therefore, the results confirm the faults of currently produced parts by collecting real-time data from each factor during the foam-making process. Furthermore, control of each of the factors may improve the fault rate.

• 한국신재생에너지학회:학술대회논문집
• /
• 2010.06a
• /
• pp.84.1-84.1
• /
• 2010

Currently, new and renewable energy come into the spotlight, such as solar energy, wind power, fuel cell, hybrid car etc., due to the energy resource is being depleted. Especially, in order to solve like this problem, the study of solar cell manufacturing systems are being extensively researched such as vacuum process. But the major fault of the vacuum process are its expensive production price. On the order hand, Roll-to-roll printing system, the new technology of solar cell manufacturing, has low production price compare with the vacuum process. Also roll-to-roll printing system can decrease the 95% of waste water and 99.9% of harmful gasses than the vacuum process. So we addressed the roll to roll printing system for the flexible solar cell by using printing technology. This roll-to-roll printing system is comprised of various modules, such as web handling module, fine pattern printing module, dry/curing module, uniform coating module and laminating module etc.

• 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.

• Journal of the Korean Society of Safety
• /
• v.33 no.6
• /
• pp.37-41
• /
• 2018

Process safety technology has developed from qualitative methods such as HAZOP (hazard and operability study) to semi-quantitative methods such as LOPA (layer of protection analysis), and quantitative methods are actively studied these days. Quantitative risk assessment (QRA) is often based on fault tree analysis (FTA). FTA is efficient, but difficult to apply when failure events are not independent of each other. This problem can be avoided using a Markov process (MP). MP requires definition of all possible states, and thus, generally, is more complicated than FTA. A method is proposed in this work that uses an MP model and a Weibull distribution model in order to construct a reliability model for multiple dependent failures. As a case study, a pressure safety valve (PSV) is considered, for which there are three kinds of failure, i.e. open failure, close failure, and gas tight failure. According to recently reported inspection results, open failure and close failure are dependent on each other. A reliability model for a PSV group is proposed in this work that is to reproduce these results. It is expected that the application of the proposed method can be expanded to QRA of various systems that have partially dependent multiple failure states.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2009.10a
• /
• pp.342-345
• /
• 2009

Electro galvanized steel is electroplated cold roller steel for improving corrosion resistance and paintability, and is widely used in automobiles and home appliances. In the electroplating line for manufacturing electro galvanized steel if plating process is carried out with impurity on conductor roll surface, the defects in manufacturing process occurs because of steel fault. For quality, polishing is always required to separate impurity on surface of conductor roll. In this study, finite element analysis of wear for polisher brush is carried out for replaced time of it.

• Journal of Digital Convergence
• /
• v.11 no.12
• /
• pp.431-438
• /
• 2013

In this paper we proposed a intelligent recommendation processor that the type of auto parts failures that may occur in the checkout process is represented by association relationship and the relationship was implemented with ontology. For this purpose, we defined 10 kinds of failure types and their associated parts, and we designed to simulate the recommendation process of five views. For components to be checked with the type of fault, it was possible to be expansion recommendation to the intelligent by controlling the weight value according to the relationship on the components.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.9
• /
• pp.389-401
• /
• 2002

To complement the conventional fusion methodologies of state fusion and measurement fusion, a time-propagated measurement fusion methodology is proposed. Various aspects of common process noise are investigated regarding information preservation. Based on time-propagated measurement fusion methodology, four compression filters are derived. The derived compression filters are efficient in asynchronous sensor fusion and fault detection since they maintain correct statistical information. A new batch Kalman recursion is proposed to show the optimality under the time-propagated measurement fusion methodology. A simple simulation result evaluates estimation efficiency and characteristic.