Wavelet analysis and neural network based adaptive single-pole autoreclosure scheme for EHV transmission systems

This paper proposes a wavelet analysis and neural network based adaptive single-pole autoreclosure scheme for Extra High Voltage (EHV) transmission systems. First, the fault transients generated by the secondary arc and permanent faults are analysed using discrete wavelet transform with particular reference to the development of the adaptive autoreclosure scheme. Daubechies D4 wavelet transform is adopted and the numerical analyses reveal that certain wavelet components can be effectively used as the features to detect and identify the fault relevant characteristics in transmission systems. Several results of wavelet analysis are used as the feature vectors of artificial neural network which is designed to distinguish between transient and permanent faults, and to determine the secondary arc extinction point. The outcome of the study clearly indicates that the wavelet analysis combined with neural network approach can be used as an attractive and effective means of realising an adaptive autoreclosing scheme.