• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.878-882
• /
• 2004

In this paper, we propose a hierarchical hybrid neural network for detecting faults of induction motor. Implementing the classifier based on the input and output data, we apply appropriate transform and classification method at each step. In the proposed method, after obtaining the current of state of motor for each period, we transform it by Principle Component Analysis(PCA) to reduce its dimension. Before the training process, we use the conditional Fuzzy C-means(FCM) for obtaining the initial parameters of neural network for more effective learning procedure. From the various simulations, we find that the proposed method shows better performance to detect and diagnosis of induction motor and compare than other methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.11
• /
• pp.2486-2491
• /
• 2013

There are invisible defects by naked eyes in ceramic material images such as internal stomata, cracks and foreign substances. In this paper we propose a method to detect and extract such defects from ceramic pipe weld zone by applying ART2 learning. In pre-processing, we apply Ends-in Search Stretching to enhance the intensity and then perform fuzzy binarization with triangle type membership function followed by enhanced ART2 that interacts with random input patterns to extract such invisible defects. The experiment verifies that this proposed method is sufficiently effective.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.10
• /
• pp.2123-2135
• /
• 1997

In this paepr, we consider resistive shorts on drain-source, drain-gate, and gate-source as well as opens in MOSFETs included in typical memory cell of VLSI SRAM. Behavior of memeory is observed by analyzing voltage at storage nodes of memeory and IDDQ(quiescent power supply current) through PSPICE simulation. Using this behavioral analysis, an effective testing algorithm of complexity O(N) which can be applied to both functional testing and IDDQ testing simultaeously is proposed. Built-In Self Test(BIST) circuit which detects faults in memories and Built-In Current Sensor(BICS) which monitors the power supply bus for abnormalities in quescent current are developed and imprlemented to improve the quality and efficiency of testing. Implemented BIST and BICS circuits can detect locations of faults and defects in order to repair faulty memories.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2004.10a
• /
• pp.254-257
• /
• 2004

An electromagnetic acoustic transducers (EMAT) can usually generate or detect an ultrasonic wave into specimens across a small gap. Especially stiffness of composites depends on layup sequence of CFRP(carbon fiber reinforced plastics) laminates because the layup of composite laminates influences there properties. It is very important to evaluate the layup errors in prepreg laminates. A nondestructive technique can therefore serve as a useful measurement for detecting layup errors. It was shown experimentally that this shear waves for detecting the presence of the errors is very sensitive. It is found that high probability shows between tests and the model developed in characterizing cured layups of the laminates. Also a C-scan method was used for detecting layup of the laminates because of extracting fiber orientation information from the ultrasonic reflection caused by structural imperfections in the laminates. Therefore, it was found that interface C-scan images show the fiber orientation information by using two-dimensional fast Fourier transform(2-D FFT).

• Journal of the Korean Society for Nondestructive Testing
• /
• v.22 no.3
• /
• pp.254-260
• /
• 2002

In this study, two types of mobile robotic systems using NDT (Non-destructive testing) method are developed for automatic diagnosis of the boiler tubes and large size pipelines. The developed mobile robots crawl the outer surface of the tubes or pipelines and detect in-pipe defects such as pinholes, cracks and thickness reduction by corrosion and/or erosion using EMAT (Electro-magnetic Acoustic Transducer). Automation of fault detection by means of mobile robotic systems for these large-scale structures helps to prevent significant troubles without danger of human beings under harmful environment.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.3
• /
• pp.149-155
• /
• 2004

This paper presents how to find proper operating points of FACTS devices to enhance the steady-state security level considering line contigency analysis. Three generic types of FACTS devices such as series controllers, shunt controllers, and series-shunt controllers are introduced and applied to moximize a security margin and to minimize security indices. Security indices related to line flows and bus voltages are utilized and minimized iteratively in this paper. Contingency analysis is performed to detect the most severe single line fault. In various load conditions, FACTS devices are tested to establish appropriate preventive or corrective action without generation re-dispatching or load shedding. The FACTS operation scheme is verified on the IEEE 57-bus system in a line-faulted contingency.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.24 no.3
• /
• pp.294-301
• /
• 2015

Oil filters for vehicular parts are used under high and low temperatures as the engine is cooled, and defects can be generated with repeated changes in the operating environment and with changes in the shape, such as very high internal losses for the bolts. Visually checking for defects inside a bolt is impossible. Nondestructive evaluation methods such as eddy current testing (ECT) are recommended as a more effective way to examine inside a bolt and detect surface defects in a short amount of time. In this study, the fit bobbin coil eddy current probe was applied to checking the bolts. The bolt parameters were calculated by using a COMSOL analysis program to obtain parameters for professional interior design and fault diagnosis.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.9 no.1
• /
• pp.88-95
• /
• 2004

Micro stepping method is adopted in order to eliminate effectively the resonant phenomena and to increase the positional resolution. Exist micro-step method by using Sinusoidal waveform, drive circuit is complex by using micro controller and ROM, it have fault on cost Increase. This paper proposed trapezoidal current wave form for simple control circuit and micro stepping method by using a low cost controller. This paper proposed method verify by using CPLD(EPM9320RC208-15) of low cost. This paper make experiment that comparison of exist method and proposed method. This paper obstruct a escape of motor by using high speed detect.

• Proceedings of the KIEE Conference
• /
• 2002.11b
• /
• pp.241-243
• /
• 2002

Recently, great efforts are concentrated on the autonomous, adaptive protection schemes with advanced artificial intelligence and digital technology. It is highly required for a next generation protective system not only to detect and to clear a fault, but also to fit itself to the changing environment. In this paper it is suggested an evaluation method for the protection ability of a protective system in a distributed system. The suggested method is of bottom-up scheme, in other words, protection ability is estimated from the lowest level of parameters in each protective devices to the highest level of the whole protective system. This feature makes it possible to evaluate the protection ability either for the protective device(or a system), or for a protected system. And, in addition, it is enabled that the protectability concept can be applied in the design stage of a protective system for a distribution network. The proposed method is applied to a simple distributed network to show its effectiveness.

• 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.