The high-frequency attenuation of the fault transient traveling wave signal caused by line loss significantly reduces the resolution and reliability of the traveling wave time difference method and the fault location method based on the linear correlation of transient signals. To address this issue

this paper proposes a fault location method based on nonlinear correlation estimation using a Gaussian kernel function. The method maps transient fault signals into a high-dimensional Hilbert space

where the kernel correlation principle is employed to amplify the subtle differences in resonant waveform features. By tuning a controllable scale parameter

the proposed method achieves enhanced spatial resolution. Simulation results demonstrate that

in lossy transmission media

such as underground cables and long-distance overhead lines

and under high-resistance fault conditions

the method maintains high resolution and strong robustness

effectively mitigating the resolution degradation and sidelobe interference observed in linear criteria. Field tests on a 10 kV distribution network further validate the approach

achieving a relative location error of only 0.85% and a spatial resolution of 60 m which is approximately five times higher than that of the conventional linear correlation method (310 m).