In order to solve the problem that the internal discharge（PD）signal of gas insulated switchgear（GIS）contains noise

this paper builds a simulated partial discharge environment

and using ultra-high frequency（UHF）to acquire the defective PD signal. In view of the characteristics of periodic narrow-band noise and white noise forUHF PD signals

a denoising study of UHF PD signals based on improved experience wavelet（EWT）and wavelettransform is proposed. Firstly

the EWT transform can be used to decompose the signal into multi-frequency modalfunctions

then the modal function is wavelet denoised

divide the denoised modal function according to the kurtosisvalue

and select UHF PD according to the appropriate threshold. The active component of the signal and recon-structs the signal. Finally

the UHF PD simulation signal is constructed and the measured data is used to verify theeffectiveness of the proposed algorithm. The simulation experiment and the measured denoising results show that theimproved denoising algorithm proposed in this paper has good noise suppression ability. This paper provides a refer-ence for the denoising of UHF PD signals inside GIS equipment.