Fault line selection method for small-current grounding system based on injected signal and improved VMD energy entropy

To address the challenge of signal extraction in resonance-grounded fault line selection (FLS) using the injected signal method, this paper proposes a novel FLS approach for small-current grounding systems. We examine how the injected signal is distributed and determine the parameters for the square wave signal being injected. The power frequency notch filtering method and Variational Mode Decomposition (VMD) optimized by the Sparrow Search Algorithm (SSA) are employed to extract the response components of the injected signal from the faulty zero-sequence current (ZSC). FLS is achieved by calculating and comparing the energy entropy of the VMD for the response signals across various lines. Simulation and field experiment results show that the proposed method is robust against transition resistance, fault distance, and neutral grounding methods, showing excellent resistance to high transition resistance and noise.