• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.32 no.2
• /
• pp.151-156
• /
• 2019

The aim of this investigation is to detect specific waveforms in a distribution line prior to the occurrence of a fault. Conditions were introduced such that a feeder remote terminal unit (FRTU) of the distribution automation system selects and stores fault waveforms from the different waveforms detected in the distribution line. In addition, an algorithm was developed to detect specific waveforms from the fault waveforms stored using the FRTU. This algorithm exploits the duration and periodicity of harmonic changes in voltage and current. The efficacy of the algorithm was confirmed based on the measurements of fault waveforms in an actual distribution line. The results indicated that faults in a distribution line can be predicted via experimental measurements.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.10
• /
• pp.35-42
• /
• 2020

In this study, we propose a crack detection method using limited data with a U-Net based image inpainting technique that is a modified unsupervised anomaly detection method. Concrete cracking occurs due to a variety of causes and is a factor that can cause serious damage to the structure in the long term. In general, crack investigation uses an inspector's visual inspection on the concrete surfaces, which is less objective in judgment and has a high possibility of human error. Therefore, a method with objective and accurate image analysis processing is required. In recent years, the methods using deep learning have been studied to detect cracks quickly and accurately. However, when the amount of crack data on the building or infrastructure to be inspected is small, existing crack detection models using it often show a limited performance. Therefore, in this study, an unsupervised anomaly detection method was used to augment the data on the object to be inspected, and as a result of learning using the data, we confirmed the performance of 98.78% of accuracy and 82.67% of harmonic average (F1_Score).

• Journal of Broadcast Engineering
• /
• v.22 no.5
• /
• pp.632-642
• /
• 2017

In this paper, we propose a DNN based speech detection system using acoustic characteristics and context information of media audio. The speech detection for discriminating between speech and non-speech included in the media audio is a necessary preprocessing technique for effective speech processing. However, since the media audio signal includes various types of sound sources, it has been difficult to achieve high performance with the conventional signal processing techniques. The proposed method improves the speech detection performance by separating the harmonic and percussive components of the media audio and constructing the DNN input vector reflecting the acoustic characteristics and context information of the media audio. In order to verify the performance of the proposed system, a data set for speech detection was made using more than 20 hours of drama, and an 8-hour Hollywood movie data set, which was publicly available, was further acquired and used for experiments. In the experiment, it is shown that the proposed system provides better performance than the conventional method through the cross validation for two data sets.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.831-835
• /
• 2006

NOAA/AVHRR data were used to analyze sea surface temperatures (SSTs) and thermal fronts (TFs) in the Korean seas. Temporal and spatial analyses were based on data from 1993 to 2000. Harmonic analysis revealed mean SST distributions of $10{\sim}25^{\circ}C$. Annual amplitudes and phases were $4{\sim}11^{\circ}C$ and $210{\sim}240^{\circ}$, respectively. Inverse distributions of annual amplitudes and phases were found for the study seas, with the exception of the East China Sea, which is affected by the Kuroshio Current. Areas with high amplitudes (large variations in SSTs) showed 'low phases' (early maximum SST); areas with low amplitudes (small variations in SSTs) had 'high phases' (late maximum SST). Empirical orthogonal function (EOF) analyses of SSTs revealed a first-mode variance of 97.6%. Annually, greater SST variations occurred closer to the continent. Temporal components of the second mode showed higher values in 1993, 1994, and 1995. These phenomena seemed to the effect of El $Ni{\tilde{n}}o$. The Sobel edge detection method (SEDM) delineated four fronts: the Subpolar Front (SPF) separating the northern and southern parts of the East Sea; the Kuroshio Front (KF) in the East China Sea, the South Sea Coastal Front (SSCF) in the South Sea, and a tidal front (TDF) in the West Sea. Thermal fronts generally occurred over steep bathymetric slopes. Annual amplitudes and phases were bounded within these frontal areas. EOF analysis of SST gradient values revealed the temporal and spatial variations in the TFs. The SPF and SSCF were most intense in March and October; the KF was most significant in March and May.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.28 no.4
• /
• pp.352-357
• /
• 2008

The detection of micro cracks in materials at the early stage of fracture is important in many structural safety assurance problems. The nonlinear ultrasonic technique (NUT) has been considered as a positive method for this, since it is more sensitive to micro crack than conventional linear ultrasonic methods. The basic principle is that the waveform is distorted by nonlinear stress-displacement relationship on the crack interface when the ultrasonic wave transmits through, and resultantly higher order harmonics are generated. This phenomenon is called the contact acoustic nonlinearity (CAN). The purpose of this paper is to prove the applicability of CAN experimentally by detection of micro fatigue crack artificailly initiated in Aluminum specimen. For this, we prepared fatigue specimens of Al6061 material with V-notch to initiate the crack, and the amplitude of second order harmonic was measured by scanning along the crack direction. From the results, we could see that the harmonic amplitude had good correlation with COD and it can be used to detect the crack depth in more accurately than the common 6 dB drop echo method.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.4
• /
• pp.12-20
• /
• 1995

This paper describes principles and experimental results in conjunction with examining the possibility of the detection of effects from a cylindrical objects. Cylindrical objects have its own resonance frequency. The frequency varies according to the diameter, length etc. And acoustic sound, whose frequency and the harmonic components corresponding to resonance frequency, is radiated from the object. When an object have defect in itself, the radiated acoustic sound is different from normal one. So we can detect its defect by analysing frequency components of acoustic sound. We proved that detection of defective objects by acoustic signal analysis is possible automatically. Also the result can be applied to other kinds of objects.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.17 no.4
• /
• pp.315-321
• /
• 2012

The utility-interactive inverter with critical loads should supply continuous and stable voltage to critical loads even during the grid fault. The conventional control method which performs current control for grid-connected mode and voltage control for stand-alone mode undergoes the critical load voltage variation during grid fault. The critical load voltage may have large transient when the inverter performs mode transfer after the islanding detection. Recently, the indirect current control method which does not have the transient state during not only islanding detection but also the mode transfer has been proposed. However, since the voltage control is maintained even during the grid-connected mode it is difficult to detect the islanding. This paper proposes an active anti-islanding method suitable for the indirect current control method which does not have NDZ(Non-Detection Zone).

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.27 no.11
• /
• pp.27-36
• /
• 2013

Islanding in a gird connected inverter of wind power generation system may influence a bad effect on equipments or yield safety hazards on grid so it should be detected rapidly and exactly. A passive method to detect islanding is comparatively simpler than an active method but suffers from non detection zone (NDZ). On the other hand, the active method can significantly reduce NDZ by injecting a disturbance into inverter output. To improve the reliability of islanding detection, this paper proposes a hybrid anti-islanding detection method combining the conventional passive method as well as the active method based on novel harmonic injection method using fourier transform. The proposed scheme is fast to detect islanding when NDZ does not exist because it has the nature of passive method. Under NDZ, the active method can detect occurrence of islanding reliably. The effectiveness and validity of the proposed scheme is proved through comparative simulations.

• Journal of the Korean Solar Energy Society
• /
• v.29 no.5
• /
• pp.65-72
• /
• 2009

The new method is proposed to improve high speed detection of grid voltage phase and magnitude during a voltage dip due to a grid faults. Usually, A LPF(Low Pass Filter) is used in the feedback loop of PLL (Phase Locked Loop) system because the measured grid voltage contains harmonic distortions and sensor noises. so, a new design method of the loop gain of the PI -type controller in the PLL system is proposed with the consideration of the dynamics of the LPF. As a result, a better transient response can be obtained with the proposed design method. The LPF frequency and PI controller gain are designed in coordination according to the steady state and dynamic performance requirement. This paper shows the feasibility and the usefulness of the proposed methods through the computer simulation and the lab-scale experiments.

• Journal of the Korean Solar Energy Society
• /
• v.35 no.2
• /
• pp.11-19
• /
• 2015

Islanding is a phenomenon that EPS(Electric Power System) is continuously energized by PV PCS(Photovoltaic Power Conditioning System) even when EPS is isolated from the grid. Unintentional islanding will result in safety hazard, power quality degradation and many other issues. So, islanding protection of grid-connected PV PCS is a key function for standards compliance. Nowadays, many anti-islanding schemes are researched. But existing anti-islanding schemes used in PV PCS have power quality degradation and non-detection zone issues. This paper analyses not only detection performance of existed anti-islanding schemes using active frequency drift but also THD of PCS output current according to each value disturbance for anti-islanding. In addition, the lowest value of disturbance in each scheme was tabulated under guarantee of anti-islanding condition.