Unreliant effect of mother wavelets in localization of power system faults for non-homogeneous AC microgrid

Identification of fault localization in microgrid (MG) network using travelling wave (TW) mainly depends on the propagation velocity (\(\vartheta\)) of the reflected wave and type of mother wavelet chosen for extraction TW using discrete wavelet transform (DWT). Estimation of \(\vartheta\) in non-homogeneous MG network and selection of the appropriate mother wavelet are a challenging task. This paper proposes a method to overcome the defined challenges in a non-homogeneous MG network. Proposed method majorly works in two parts: i) fault detection, faulty section identification and ii) fault location scheme in a medium voltage MG network. The first part utilizes added decoupled RMS currents (ADRMSC) values to detect and identify faulty section. Second part utilizes the consecutive sample difference approach (CSDA), and first arrival time instants of travelling wave (TW) at section terminals, computed using discrete wavelet transform (DWT) to locate the fault. Proposed scheme depends on the capture of time instants, which makes it least affected by the change in mother wavelets. The proposed scheme also performs well for a non-homogeneous system as it is independent of TW propagation velocity to locate the fault. The performance of the proposed method is validated on the non-homogeneous IEEE-33 bus system in EMTDC/PSCAD for various case studies which includes changes in fault parameters such as fault resistance, fault inception angle and noise level and found effective even with the change in mother wavelets.