3D Wavelet Network and Wavelet Transform Used for Transmission Lines Fault Detection and Their Classification

摘要

Accurate detection and classification of transmission line faults for permanent protection in avoiding costly maintenance remain challenging to power system engineers. To resolve this issue, we used Wavelet Transform (WT) and 3D-Wavelet Network (3DWN) to detect and classify various types of faults in transmission lines depending on the emanating waves from the power system. First, the WT was used to extract the vector features for each type of faults. Next, these features were analyzed using three level decompositions. The wavelet toolbox in MATLAB/Simulink was utilized to calculate the maximum norm values, maximum detail coefficients and energy of the current signals. Furthermore, 3DWN was employed to classify the single line to ground faults, line-to-line faults, double line to ground faults and three lines faults. Result obtained using WT and 3DWN confirmed the possibility of developing an accurate fault classification scheme useful for reliable transient-based protection approaches where this applicable for each case of faults.