Evaluating the Use of Shapelets in Traveling Wave Based Fault Detection and Classification in Distribution Systems

The application of traveling wave (TW) principles for fault detection in
distribution systems
is challenging because of multiple reflections from the laterals and other lumped elements, particularly when we consider communication-free applications. In Biswal et al. (2022), we proposed the use of Shapelets to characterize fault signatures as we move towards a more realistic data-driven solution. The Shapelet can be correlated with the physical characteristics of TW generation process, in contrast to choosing a predefined basis such as Wavelets. This paper is an extension of this work to comprehensively evaluate Shapelets for identifying the fault types, in addition to their locations in single-phase, three-phase, and IEEE 13-bus distribution systems. We also propose using the voltage waveforms instead of the forward waves, which eliminates the need for a CT. The proposed method can identify fault locations within a local region with
$\approx 99\%$
accuracy and classify the fault types with an accuracy of over 90%.