Wavelet Customization for Improved Fault-Location Quality in Power Networks

A procedure for computing customized wavelets suitable for identifying the location of a fault along the length of a transmission line is proposed. Measuring the time difference of arrival (TDOA) between two consecutive transient reflections or the TDOA between the two initial peaks of two synchronized voltage/current bus measurements provides an accurate fault-location estimate. Reliable TDOA estimates can be obtained only when the transformed fault signals consist of short/peaked pulses that arrive on a bus with a small group delay. Our customization method improves the "peakedness" and shortens the group delay of the transient signal by maximizing a suitable peakedness objective function. It is shown that this approach is superior in terms of accuracy compared to the "classic" (Daubechies) wavelet-based technique found in the literature. The proposed method is independent of the fault type and can be used to reduce the required data sampling rate while maintaining reliable TDOA estimates.