Research on anti-interferences methods in the PD detection on power transformer

• Published in: 2017 1st International Conference on Electrical Materials and Power Equipment (ICEMPE) pp. 374 - 377
• Main Authors: Jing Liu, Jingfeng Wu, Yanhua Han, Shuangzan Ren, Yang Liu, Xuefeng Zhao, Ting Ren, Yuanjing Lu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2017
• Subjects: Acoustic measurements; detection; noise; Noise reduction; partial discharge; Partial discharge measurement; Partial discharges; Pollution measurement; transformer; Wavelet analysis; Wavelet transforms
• DOI: 10.1109/ICEMPE.2017.7982106

The paper concerns improvement of the acoustic emission (AE) methods applied in diagnosing of insulations systems of electric power transformer. During detection of AE signals emitted by partial discharges (PD) occurring in insulation systems of power transformer, there exist different forms of interference sources, which may have a direct impact on the correct evaluation of the measured signals obtained using the AE method. Thus, analysis of the impact of interfering signals on the PD detection and classification is performed. In this study, a correlation-based wavelet base selection method is introduced. The proposed wavelet analysis approach capable of conserving a significant portion of the original signal, while the threshold value determined based on the relative difference between the original and noisy signal. The proposed method is applied to PD signals that obtained in the field. With the comparison of other de-noising methods, the de-noised PD signals indicate that the proposed method yields higher reduction in noise levels.