Partial discharge（PD） detection is one of the main means of monitoring the insulation state of power cables. However

on-site PD testing is susceptible to white noise interference. An adaptive white noise suppression method for partial discharge based on wave motions（WM） is proposed to save this problem. Firstly

in this method

the fluctuation characteristics of signal in the time domain are utilized to suppress white noise white noise

and the time threshold window is utilized to further eliminate redundant noise. In order to reduce the interference of artificial selection

the Otsu algorithm（OSTU） is used to adaptively select the threshold. Secondly

the feasibility of the method is verified by two typical noisy partial discharge signals simulated on the MATLAB. The proposed method has been examined on simulated noisy PD signals

and its results have been compared with discrete wavelet transform（DWT） de-noised results and adaptive singular value decomposition（ASVD） de-noised results. Finally

the validity of the method is verified by PD tests on typical cable defects under AC voltage in the laboratory. The de-noised results of this method have been compared with DWT de-noised results

ASVD de-noised results

and the method de-noised results. The results show that this method has better denoising effect and shorter algorithm time compared with DWT and ASVD methods.