This paper aims to use only one sensor to achieve better PD fault location of power cables in a low signal-to-noise ratio (SNR) noise environment. Therefore, an improved phase difference… Click to show full abstract
This paper aims to use only one sensor to achieve better PD fault location of power cables in a low signal-to-noise ratio (SNR) noise environment. Therefore, an improved phase difference (IPD) method based on one sensor is used to estimate the PD fault location of power cables. This method has a good advantage of using only the first two PD peaks, which is suitable for location estimation in the low SNR noise environment. The white Gaussian noise is regarded as background noise to simulate the noise environment. In order to enhance the accuracy of the IPD method in the noisy environment, a new denoising algorithm combining high-order statistics (HOS) criterion and generalized cross-validation (GCV) thresholding is proposed to process PD signals. The results show the good performance of the proposed denoising algorithm and higher location accuracy of the IPD method in the low SNR noise environment.
               
Click one of the above tabs to view related content.