A Voltage-Drop-Based Method for Fault Classification, Branch Identification and Distance Estimation in Power Grids and AC/DC Microgrids
Year: 2026
Type: preprint
Abstract: Accurate fault location is a prerequisite for fast service restoration, yet most established techniques depend on fault current measurements, synchronised phasor hardware, or large offline databases. These assumptions are difficult to satisfy in networks dominated by inverter-based distributed energy resources, where fault currents are limited and bidirectional. This study proposes a fault diagnosis method that uses steady-state RMS bus voltage magnitudes alone. The bus with the largest three-phase-average percentage drop is taken as the initially identified faulted bus; the fault type is classified from the relative ordering of the three phase-wise drops at that bus using a tolerance band; a reference fault of the classified type is then injected there, and the neighbour with the largest positive residual completes the faulted branch; and a second injection at the far-end bus provides the two anchors of a secant interpolation for the fault distance. Because no current signal is used, the method is unaffected by current transformer saturation and by the limited short-circuit contribution of converter-interfaced sources. Validation in MATLAB/Simulink on a modified IEEE 14-bus system and a hybrid AC/DC microgrid, over 22 scenarios, gave correct branch and type identification in every case, with a mean distance error of 2.61%.
Institution: Khulna University of Engineering and Technology
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7213059656