• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1360-1363
• /
• 1996

EPD(End Point Detection) is used to decide etching degree of layer which must be removed at wafer etching process in plasma etching process which is one of the most important process in semiconductor manufacturing. In this thesis, the method which detects malfunction of etching process in real-time will be discussed. Several EPD signal traces are collected in normal plasma etching condition and used as reference EPD signal traces. Critical points can be detected by applying differentiation and zero-crossing techniques to reference EPD signal. Mean and standard deviation of critical parameters which is memorized from reference EPD signal are calculated and these determine the lower and higher limit of control chart. And by applying statical control chart to EPD signals which are collected in real etching process malfunctions of process are detected in real-time. By means of applying this method to the real etching process we prove our method can accurately detect the malfunction of etching process and can compensate disadvantage of current industrial method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.3
• /
• pp.965-983
• /
• 2014

Concurrent behaviors of multithreaded programs cannot be described effectively by automata-based models. Thus, concurrent program intrusion attempts cannot be detected. To address this problem, we proposed the process algebra-based detection model for multithreaded programs (PADMP). We generate process expressions by static binary code analysis. We then add concurrency operators to process expressions and propose a model construction algorithm based on process algebra. We also present a definition of process equivalence and behavior detection rules. Experiments demonstrate that the proposed method can accurately detect errors in multithreaded programs and has linear space-time complexity. The proposed method provides effective support for concurrent behavior modeling and detection.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.38-43
• /
• 1994

In milling process, monitoring and diagosis system is very importent to accomplish factory automation. In this study, to drvelope on-line tool breakage detection system in face milling operation, analysis and experiment were performed. The tool breakage detection experiment was performed in machining center and the effectiveness of the detection tool breakage detection alorithm and the usage of feed drive current as a detection signal were verified.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.7
• /
• pp.1618-1631
• /
• 1994

In this paper, in-process chatter detection methodology which utilizes nondimensional characteristic variables is introduced. To obtain nondimensional chatter detection indexes which are constant regardless of the cutting conditions during machining with the same tool and workpiece material, both the cutting forces and accelerations are measured and processed in time and frequency domain. The indexes are calculated from the present and past value of the acceleration and cutting force signals in time and frequency domain. The chatter is identified when these chatter detection indexes are bigger than the threshold which is decided by preliminary experiments. The experiment shows that these indexes works very well in-process chatter detection.

• Transactions on Electrical and Electronic Materials
• /
• v.14 no.2
• /
• pp.71-77
• /
• 2013

Advanced semiconductor manufacturing technology requires methods to maximize tool efficiency and improve product quality by reducing process variability. Real-time plasma process monitoring and diagnosis have become crucial for fault detection and classification (FDC) and advanced process control (APC). Additional sensors may increase the accuracy of detection of process anomalies, and optical monitoring methods are non-invasive. In this paper, we propose the use of a chromatic data acquisition system for real-time in-situ plasma process monitoring called the Plasma Eyes Chromatic System (PECS). The proposed system was initially tested in a six-inch research tool, and it was then further evaluated for its potential to detect process anomalies in an eight-inch production tool for etching blanket oxide films. Chromatic representation of the PECS output shows a clear correlation with small changes in process parameters, such as RF power, pressure, and gas flow. We also present how the PECS may be adapted as an in-situ plasma arc detector. The proposed system can provide useful indications of a faulty process in a timely and non-invasive manner for successful run-to-run (R2R) control and FDC.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.36 no.6
• /
• pp.430-436
• /
• 1987

A Detection filter that can be used for the Detection and Isolation of process faults is proposed by the use of fault vector modelling, and is applied to DC Motor fault detection. The proposed detection filter is a new one in a view point that its outputs are the estimates of fault variables(or linear combination of them) while all the existing filters estimate the state of process. By this properties, the process fault detection systems with this filter can be constructed in very simple structure. Besides the simplicity of structure and design procedure, the filter has an useful feature that various types of fault can be estimated via the filter by choosing appropriate fault models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4285-4299
• /
• 2019

Adversarial attack is a technique that causes a malfunction of classification models by adding noise that cannot be distinguished by humans, which poses a threat to a deep learning model. In this paper, we propose an efficient method to detect adversarial images using Gaussian process regression. Existing deep learning-based adversarial detection methods require numerous adversarial images for their training. The proposed method overcomes this problem by performing classification based on the statistical features of adversarial images and clean images that are extracted by Gaussian process regression with a small number of images. This technique can determine whether the input image is an adversarial image by applying Gaussian process regression based on the intermediate output value of the classification model. Experimental results show that the proposed method achieves higher detection performance than the other deep learning-based adversarial detection methods for powerful attacks. In particular, the Gaussian process regression-based detector shows better detection performance than the baseline models for most attacks in the case with fewer adversarial examples.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.2
• /
• pp.246-255
• /
• 1998

This paper presents a novel method for real-time malfunction detection of plasma etching process using EPD signal traces. First, many reference EPD signal traces are collected using monochromator and data acquisition system in normal etching processes. Critical points are defined by applying differentiation and zero-crossing method to the collected reference signal traces. Critical parameters such as intensity, slope, time, peak, overshoot, etc., determined by critical points, and frame attributes transformed signal-to symbol of reference signal traces are saved. Also, UCL(Upper Control Limit) and LCL(Lower Control Limit) are obtained by mean and standard deviation of critical parameters. Then, test EPD signal traces are collected in the actual processes, and frame attributes and critical parameters are obtained using the above mentioned method. Process malfunctions are detected in real-time by applying SPC(Statistical Process Control) method to critical parameters. the Real-time malfunction detection method presented in this paper was applied to actual processes and the results indicated that it was proved to be able to supplement disadvantages of existing quality control check inspecting or testing random-selected devices and detect process malfunctions correctly in real-time.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.2
• /
• pp.115-123
• /
• 2021

A lot of sensor and control signals is generated by an industrial controller and related internet-of-things in discrete manufacturing system. The acquired signals are such records indicating whether several process operations have been correctly conducted or not in the system, therefore they are usually composed of binary numbers. For example, once a certain sensor turns on, the corresponding value is changed from 0 to 1, and it means the process is finished the previous operation and ready to conduct next operation. If an actuator starts to move, the corresponding value is changed from 0 to 1 and it indicates the corresponding operation is been conducting. Because traditional fault detection approaches are generally conducted with analog sensor signals and the signals show stationary during normal operation states, it is not simple to identify whether the manufacturing process works properly via conventional fault detection methods. However, digital control signals collected from a programmable logic controller continuously vary during normal process operation in order to show inherent sequence information which indicates the conducting operation tasks. Therefore, in this research, it is proposed to a recurrent neural network-based fault detection approach for considering sequential patterns in normal states of the manufacturing process. Using the constructed long short-term memory based fault detection, it is possible to predict the next control signals and detect faulty states by compared the predicted and real control signals in real-time. We validated and verified the proposed fault detection methods using digital control signals which are collected from a laser marking process, and the method provide good detection performance only using binary values.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06b
• /
• pp.16.2-21
• /
• 1998

Old movies are often corrupted by randomly located blotches and scratches. In this paper were present an efficient method for detection and removal of these distortions. The presented method is composed of two separate steps: the detection process and the restoration process. In the detection process, blotch locations are detected through global motion segmentation, the sequential approach to motion segmentation, a robust model-fit criterion and so on, we form the algorithm for the algorithm for the global motion segmentation tuned to the blotch detection problem. In the restoration process, the missing data of the detected blotch areas are temporally extrapolated from the corresponding image areas at the preceding or the succeeding image frame with considering the global motion segmentation results. We apply the presented method to moving image sequences distorted by artificial blotches. The method works very well and provides a subjective improvement of picture quality.