Fault classification of a transmission line using wavelet transform & fuzzy logic

This paper implemented wavelet transforms integrated with a fuzzy logic system to classify all 11 types of faults in transmission line without compensation. The proposed scheme uses only two successive cycles (one cycle pre and one cycle post fault samples) of the three phase current signals to estimate ground current signal at one end. Energy is obtained by decomposition of discrete wavelet transforms using Haar mother wavelet. Then, a novel feature selection technique based on differential wavelet energy is implemented in wavelet components in the sort of 1 KHz- 2 KHz. Finally, optimal features are normalized and fed to fuzzy logic for classification tasks. The possibility of the proposed scheme is tested in a 400-kV, 300 km long transmission line under different simulation conditions using MATLAB software. The results show that the planned scheme can acceptably classify all fault types with wide distinctions in system settings and in noisy environment.