• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권7호
• /
• pp.317-322
• /
• 2002

This paper presents a linear temperature/voltage relation over a temperature range(0~100$^{\circ}C$) with high-resolution, reasonably god response linearity, reliability, and overall improved performance. A notable feature of the proposed method is that the whole temperature span is divided by 16 subbands thus the variation span of measured voltage is narrowed. Therefore noise characteristic has been improved. The output voltage is amplified after the uniformed DC voltage component is eliminated by offset. We proposed circuit using median value and mean value of digital signal in order to reduce the noise effect. The proposed circuit offers linear temperature/voltage conversion over a wide dynamic range using DSP(TMS320C31) and steinhart equation.

• 대한전기학회논문지:전력기술부문A
• /
• 제53권10호
• /
• pp.542-549
• /
• 2004

The conventional fault location algorithms based on the travelling waves have an inherent problem. In cases of the close-up faults occurring near the relaying point and of the faults having zero degree inception angle of voltage signals, the conventional algorithms can not estimate an accurate fault distance. It is because the shapes of travelling waves are near sinusoidal in those cases. A new method solving this problem is presented in this paper. An FIR(Finite Impulse response) filter which makes high frequency components prominent and makes the power frequency component and dc-offset attenuated is used. With this method, the cross-correlation peak is to be very clear when a close-up fault or a fault having near zero-degree inception angle occurs. The cross-correlation peaks can be clearly distinguished and accurate fault location is practically possible consequently. A series of simulation studies using EMTP(Electromagnetic Transients Program) show that the proposed algorithm can calculate an accurate fault distance having maximum 2% or less error.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권10호
• /
• pp.1215-1221
• /
• 1999

A high speed digital distance relaying algorithm based on a Wavelet Transformation is proposed. To obtain stable phasor values very quickly, first, a lowpass filter which has low cutoff frequency is used. Secondly, db2(Daubechies 2) Wavelet which has the data window of 4 samples is used. A FIR filter which removes the DC-offset component in current relaying signals is applied. In accordance with a series of tests, the operation time of the relaying algorithm is less than 3/4 cycles after faults in a 80 [km], 154[kV], 60[Hz] over-head transmission line system.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권7호
• /
• pp.855-862
• /
• 1999

A high speed digital distance relaying algorithm based on a least square error method is proposed. To obtain stable phasor values very quickly, first, a lowpass filter which has very short transient period and no overshoot is presented. Secondly, the least square error method having the data window of 3 samples is used by applying a FIR filter which removes the DC-offset component in current relaying signals. Test results show that the proposed distance relaying algorithm detects most of internal faults within a half cycle after faults in a 154[kV] overhead transmission line system.

• 전기학회논문지
• /
• 제57권6호
• /
• pp.938-943
• /
• 2008

Coupling capacitor voltage transformers(CCVTs) have been used in extra or ultra high voltage systems to obtain the standard low voltage signal for protection and measurement. For fast suppression of the phenomenon of ferroresonance, three winding CCVTs are used instead of two winding CCVTs. A tuning reactor is connected between a capacitor voltage divider and a voltage transformer to reduce the phase angle difference between the primary and secondary voltages in the steady state. Slight distortion of the secondary voltage is generated when no fault occurs. However, when a fault occurs, the secondary voltage of the CCVT has significant errors due to the transient components such as dc offset component and/or high frequency components resulting from the fault. This paper proposes an algorithm for compensating the secondary voltage of a three winding CCVT in the time domain. With the values of the measured secondary voltage of the three winding CCVT, the secondary, tertiary and primary currents and voltages are estimated; then the voltages across the capacitor and the tuning reactor are calculated and then added to the measured voltage. Test results indicate that the algorithm can successfully compensate the distorted secondary voltage of the three winding CCVT irrespective of the fault distance, the fault impedance and the fault inception angle as well as in the steady state.

• 전기학회논문지
• /
• 제57권6호
• /
• pp.909-914
• /
• 2008

A coupling capacitor voltage transformer(CCVT) is used in an extra or ultra high voltage system to obtain the standard low voltage signal for protection. To avoid the phase angle error between the primary and secondary voltages, a tuning reactor is connected between a capacitor and a voltage transformer. The inductance of the reactor is designed based on the power system frequency. If a fault occurs on the power system, the secondary voltage of the CCVT contains some errors due to a dc offset component and harmonic components resulting from the fault. The errors become severe in the case of a close-in fault. This paper proposes an algorithm for compensating the secondary voltage of a CCVT in the time-domain. From the measured secondary voltage of the CCVT, the secondary and primary currents are obtained; then the voltage across the capacitor and the inductor is calculated and then added to the measured secondary voltage to obtain the correct primary voltage. Test results indicate that the proposed algorithm can compensate the distorted secondary voltage of the CCVT irrespective of the fault distance, the fault inception angle, and the burden of the CCVT.

• Journal of Electrical Engineering and information Science
• /
• 제2권6호
• /
• pp.202-207
• /
• 1997

Design and fabrication issues for an L-band GaAs Monolithic Microwave Integrated Circuit(MMIC) Voltage Controlled Oscillator(VCO) as a component of Personal Communications Systems(PCS) Radio Frequency(RF) transceiver are discussed. An ion-implanted GaAs MESFET tailored toward low current and low noise with 0.5mm gate length and 300mm gate width has been used as an active device, while an FET with the drain shorted to the source has been used as the voltage variable capacitor. The principal design was based on a self-biased FET with capacitive feedback. A tuning range of 140MHz and 58MHz has been obtained by 3V change for a 600mm and a 300mm devices, respectively. The oscillator output power was 6.5dBm wth 14mA DC current supply at 3.6V. The phase noise without any buffer or PLL was 93dB/1Hz at 100KHz offset. Harmonic balance analysis was used for the non-linear simulation after a linear simulation. All layout induced parasitics were incorporated into the simulation with EEFET2 non-linear FET model. The fabricated circuits were measured using a coplanar-type probe for bare chips and test jigs with ceramic packages.

• 자원환경지질
• /
• 제31권1호
• /
• pp.53-57
• /
• 1998

Transient electromagnetic (TEM) method is also affected by cultural and natural electromagnetic (EM) noises, since it uses part of the broadband ($10^{-2}$ to $10^5Hz$) spectrum. Especially, predominant EM noise which affects a moving transmitter-receiver TEM system is sensor motion-induced noise. This noise is caused by the sensor motion in the earth magnetic field. The technique for reducing the sensor motion-induced EM noise presented in this paper is based on Halverson stacking. This Halverson stacking is generally used in a time-domain induced polarisation (IP) system to reject DC offset and linear drift. According to spectrum analysis of the vertical component of sensor motion-induced noise, the frequency range affected by the motion of an EM sensor is less than about 700 Hz in this study. With the decrease of the frequency, the spectral power caused by the motion of a sensor increases. For example, at the frequency of 200 Hz, the spectral power of the sensor motion-induced noise is $-90dBVrms^2$ while the spectral power of the EM noise measured with a fixed sensor on the ground is $-105dBVrms^2$, and at the frequency of 100 Hz, the spectral power of the sensor motion-induced noise is $-70dBVrms^2$ while the spectral power of the EM noise measured with a fixed sensor on the ground is $-105dBVrms^2$. With applying Halverson stacking to an artificial noise transient generated by adding a noise-free transient to sensor motion-induced noise measured without pulsing, it is shown that the filtered transient is nearly consistent with the noise-free transient within a delay time of $0.5{{\mu}sec}$. The inversion obtained from this filtered transient is in accord with the true model with an error of 5%.