To address the technical challenge that traditional methods face in adapting to inclined conductor environments, this study proposes an innovative online condition assessment method. The approach leverages the linear correlation between magnetic fields (MFs) and current to enable the non-contact condition assessment of current sensors in power systems. A precise three-dimensional MF model, adaptable to any inclination angle, is developed as the foundation. Additionally, a multi-dimensional sensor array is designed, comprising 8 horizontal ring sensors and 4 vertical sensors, alongside the implementation of a two-stage tilt angle estimation algorithm and a dual-criterion fault detection mechanism. Simulation on the Comsol-Matlab platform validates the effectiveness of the proposed method on conductors with tilt angles exceeding \(30^\circ \) , where the conductor position reconstruction and angle estimation errors are within 5 mm and < 6%, respectively. Furthermore, the detection rates for four common fault types—gain drift, offset error, complete failure, and random error—range from 91% to 100%, with a false alarm rate of less than 3%. Compared to traditional methods, the proposed approach reduces errors by over 80%.