Partial discharge (PD) test is an important means to detect the insulation performance of power cable. Because the electromagnetic environment of the test site is complicated, the extracted PD signal is submerged by noise. In order to get a more real PD signal, the authors proposes a denoising method based on wavelet transform and high order partial differential equation (PDE). It uses wavelet transform to provide better pre-processing and post-processing platform of partial discharge signal, and utilizes the fourth-order PDE to iterate the low-frequency sub-signal. At the same time, the signal to noise ratio (SNR) is used as the iteration termination condition. The denoising effect of this method is compared with the traditional wavelet threshold denoising effect. Finally, the experimental results show that this method can suppress noise effectively and keep the property of edge and detail. It has better denoising performance than traditional wavelet threshold denoising method.