High-impedance faults in distribution networks are characterized by steady-state features, making them difficult to identify. A Hilbert-Huang Transform (HHT) analysis technique is utilized in this paper to identify fault lines. The Intrinsic Mode Function (IMF) components, derived from the decomposition of zero-sequence currents in the aftermath of a fault, are then used to generate characteristic values through a time-frequency energy matrix and similarity measures. In this paper, a single-phase high-impedance ground fault line selection algorithm is proposed for distribution networks. Additionally, the first-order IMF is subjected to a Hilbert transform to obtain HHT spectral peak values, which are then combined with the peak values of zero-sequence currents to form a multi-feature fusion criterion for fault line selection. Verification of the efficacy of the suggested fault line selection technique is accomplished using simulation data from a single-phase ground fault model in a distribution network.