To enhance the accuracy and timeliness of field testing for grounding impulse impedance in complex soil environments, this paper addresses the limitations of traditional peak-ratio methods—such as susceptibility to noise interference and the inability to reflect dynamic impedance variations—by proposing an identification method that combines an improved Hanning window with recursive least squares (RLS). During signal preprocessing, an improved Hanning window with adjustable parameters and energy normalization is employed to enhance the main-lobe energy concentration of impulse voltage and current signals while effectively suppressing high-frequency sidelobe leakage. In the parameter estimation stage, a low-order discrete linear model is established and an RLS algorithm with a forgetting factor is introduced to achieve full-time adaptive estimation of impulse impedance. Using a simulated surge test circuit, 18 sets of typical operating conditions with varying inductance and resistance parameters are designed. The same voltage and current data are processed using three processing methods: no windowing, standard Hanning windowing, and improved Hanning windowing. Results show that the average relative error of surge impedance is 9.16% without windowing, the standard Hanning window reduced the error to 3.78%, and the modified Hanning window further decreased the error to approximately 1.51%. Comparative analysis of different forgetting factor settings indicates that a value of approximately λ = 0.98 achieves an optimal trade-off between dynamic tracking capability and steady-state smoothness. The research results demonstrate that the proposed method achieves high identification accuracy for impact impedance and exhibits satisfactory parameter robustness under strong noise and multiple operating conditions, providing a reference for grounding impact characteristic testing and lightning protection design.
Wan et al. (Sun,) studied this question.