Drone-assisted time-varying magnetic field analysis for fault diagnosis in grounding grids

Abstract

Grounding grids are essential for ensuring the safety of power substations, but their performance can degrade due to corrosion, fractures, or other faults. Traditional fault diagnosis methods are time-consuming, labor-intensive, and require physical access to substations, posing safety risks. This paper introduces a drone-based approach for magnetic field sensing to diagnose grounding grid faults, significantly reducing operational risks and improving efficiency. However, the movement of the drone introduces time-varying electromagnetic interference (EMI) from substation equipment and the drone itself, complicating the isolation of grounding grid signals. To address this problem, we propose a time-varying un-mixing technique combined with the Fast Independent Component Analysis (FastICA) algorithm to effectively suppress the EMI and extract the grounding grid signals. Simulation results demonstrate the efficacy of the proposed technique in separating grounding grid signals under time-varying conditions, outperforming the FastICA algorithm by 96.36% and the Independent Vector Analysis (IVA) by 41.17% at a block length of 4000 and ΓΔk=0.05. These results highlight the robustness and applicability of the proposed approach for real-world grounding grid fault diagnosis, ensuring accuracy and safety in EMI-rich environments. However, the performance of the proposed technique degrades at higher values of ΓΔk, which represents the speed of the flying drone.