The most expensive and most important equipment of the energy supply chain are transformers. Malfunctions on these equipment could occur big problems in energy safety and also result high costs for system operators. Therefore, it is highly important to monitor the transformers and detect the problems before the malfunctions.
A relatively new online diagnostics method is UHF partial discharge detection. With this method we could detect partial discharges (PD) before it turns into a real short-circuit or strikeover. In UHF (Ultra High Frequency) frequency range we could detect the electromagnetic impulses emitted by partial discharges and with some complex mathematical methods it is possible to determine the source of the discharges. There are some factors which influence the signal propagation in transformer oils. In my work I investigated some of these factors and finally the job could continue to develop a method to localize PD more accurate.
In my thesis, I examined some unique problems regarding to UHF partial discharge investigation methods. My work started with reviewing high number of references about discharges, methods, equipment and I tried to figure out that how many information are they containing to build up an UHF model. One of my goals is measuring the spectrum of a partial discharge. For this measurement I used the GTEM cell of the Department of Electric Power Engineering. This equipment is suitable to measure high frequency signals. I had to solve a problem about putting the high voltage into the housing without degrade the electromagnetic shielding of the GTEM cell.
My other task was to examine the propagation in transformer oil. The permittivity is frequency-dependent, therefore if we would like to develop an accurate localization method, we also have to know the complex permittivity in the relevant frequency range. In the second part of my work I designed a permittivity measurement cell and I measured the complex permittivity of different kind of insulating liquids.