Search

The acoustic emission signals produced by discharges of a point-plane electrode on the surface (surface discharge) of different oil-impregnated aramid pressboards are detected by a 150 kHz acoustic emission sensor mounted on the outside of the experimental tank sample. Numbers, maximum amplitudes and average amplitudes of acoustic bursts are measured via a digital signal processor device and evaluated. The effect of transition mechanism from spark to pseudoglow discharge on acoustic emission signals at different voltages is examined. The tracking patterns on the surface of four different aramid pressboard insulations are compared together by using Matlab image processing program. By experimental results, tracking patterns on the surface of the bulk impregnated pressboard insulation have a significant effect on the acoustic emission signals. Acoustic emission signals amplitudes decrease dramatically on damaged pressboards and cause so called silent PDs. [ABSTRACT FROM AUTHOR]

Partial discharge (PD) detection in the ultra-high frequency (UHF) range is state of the art. One of the most critical points in this detection method is adopting proper and suitable antenna design. In this paper, a new design of UHF antennas is introduced which is suitable to be used on the transformers in the HV laboratories of the manufacturers. To ensure the applicability of the new antenna, experimental UHF PD measurements are carried out on a transformer tank model as well as on the real power transformers. Installing several UHF antennas on the transformer tank to perform defect localization is also an important issue. There are rigorous limitations and lack of the places to install the antennas. This problem is almost solved by using the mentioned new antenna design. The proper positioning of the antennas is the other critical point in the UHF method. Placing the antennas in improper positions will damage the localization accuracy. In this paper, an automatic algorithm is introduced to determine the best installation positions for the antennas to enhance the localization accuracy. The efficacy of this algorithm is investigated by using the results of experimental measurements on power transformers. [ABSTRACT FROM AUTHOR]