A new wave-based fault detection scheme during power swing

This paper presents a new fault detection scheme during a power swing by using a time-domain signal processing tool, i.e., Intrinsic Time-scale Decomposition (ITD). The ITD decomposes the input signal into proper rotation and monotonic signals, which are further processed to compute the fault index (FI). The decomposition of the input signal via ITD does not require a priori assumption about the signal being analyzed, therefore it is considered as a basis free. In order to decompose the input signal, the ITD uses extrema information of the input signal which makes it as a wave-based tool. The main distinctive attributes of the proposed scheme are its independency from the fault parameters and dependency on the morphological features of the input waveform. We take advantage of several scenarios to verify the performance of the proposed scheme on two sample systems. The high-speed wave-based FI detector identifies the fault in the time-domain considering crucial scenarios, such as close-in faults, load angle close to 180, high resistance fault, current transformer (CT) saturation, compensated system, non-fault events, and fast swing conditions. Altogether, comparative results demonstrate the merit of the proposed scheme over other methods.

Automated summary

This paper introduces a new fault detection scheme for power swing using the Intrinsic Time-scale Decomposition (ITD) method. The ITD decomposes the input signal into rotation and monotonic signals to compute the fault index (FI). The proposed scheme is independent of fault parameters and relies on the morphological features of the input waveform. Experimental results show the superiority of this scheme in detecting faults in various scenarios.