| Title |
A Study on the Detection of Single line-to-ground fault in High Resistance Grounding System using Convolutional Neural Network |
| Authors |
황동준(Dong-Jun Hwang) ; 김철환(Chul-Hwan Kim) |
| DOI |
https://doi.org/10.5370/KIEE.2023.72.9.987 |
| Keywords |
CNN; FFT; Harmonics; High resistance grounding system; Single line-to-ground fault |
| Abstract |
The voltage source is distorted or the voltage distorted by the Switching Modulation Power Supply and cable impedance components generates harmonics in the leakage current, which causes erroneous detection of single line-to-ground fault(SLGF). In the past, to prevent erroneous detection of SLGF due to leakage current, Fast Fourier Transform(FFT) was used to determine SLGF only with the fundamental wave component, but FFT can generate errors depending on the sampling frequency. This paper proposed a new type of zero-phase current detection method using CNN in High Resistance Grounding System. The simulation was performed in the proposed High resistance grounding system(HRGS), and a CNN model generated with a distorted voltage source (reflecting harmonics frequently generated in the proposed system) and a harmonics generating load (rectifier) was verified. As a result, it was confirmed that the zero-sequence current fundamental wave was accurately detected and that an accurate SLGF determination was possible. |