Investigation of factors that affect electrical partial discharge and modelling with finite element method

Partial discharge and harmful effects of partial discharge to the insulation system of high voltage power components begin to appear with the use of high voltage for generation and transmission of electricity. Degradation in the insulation system damages to the high voltage power system and it causes both a loss of time as well as material loss. Thus, the importance of partial discharge detection and measurement methods increased. Partial discharge is the electrical discharge occurring through partial bridges as a result of partial degradation of the dielectric material between the conductors. Partial discharge is an important event that threatens the safety of high voltage systems. That's why modelling by analyzing the factors that affect the partial discharge. In this thesis, electrical partial discharge, the factors that affect the partial discharge and how to do partial discharge measurement were studied. For the detection of the factors that affect the partial discharge have been conducted some experiments. Rod electrode and plane electrode were used in the experiments. Furthermore, finite element method (FEM) was stated, the modellings were done by using this method, the simulation results were interpreted and these results were compared with the results obtained in laboratory. For the detection of partial discharge characterization is needed to specify the partial discharge inception voltage and partial discharge extinction voltage. Partial discharge inception voltage is the smallest voltage starting partial discharge in case applied voltage increases slowly. The partial discharge extinction voltage is the voltage out of the partial discharge when applied voltage decreases slowly, starting from at least partial discharge inception voltage. In experiments performed in ITU High Voltage Laboratory, partial discharge characterization were investigated with three different electrodes and two different planes in five different gap spacings. During the experiments, the voltage is increased by the manuel variac and it continues to increase till a specific level after the partial discharge is also detected. After that the experiment is finished while the voltage is reduced slowly and besides, all experiment process was measured by the partial discharge measurement device and recorded. In experiments, firstly sparkover voltages were stated. Sparkover voltage is the voltage caused to perforation of the insulator and short circuit between conductors, when passing over the withstand voltage of the insulating material between conductors. After all reviews, it was seen that sparkover voltage increases, if the gap spacing or the radius of the electrode increase. We compare the data obtained in the experiments performed in laboratory and then radius of electrode, feature of plane and the gap spacing between electrode and plane is determined how it affects partial discharge characterization. In this study, it is detected to cause to the increasing of partial discharge inception voltage, if electrode radius increases. Similarly, increasing of the gap spacing between electrode and plane causes also to rise of the partial discharge inception voltage. In experiments, disc and Rogowski planes were used. In this study is stated also how the planes affect to the partial discharge characterization. The impact of partial discharge in low pressure was also investigated in laboratory by performing of the experiments with electrode and disc plane in low pressure. The experiments were done under four different low pressure. As a result, it was established that partial discharge inception voltage in low pressure was lower than normal situation. Furthermore, partial discharge inception voltage also decreased, when the pressure was reduced. After the experiments were completed, partial discharge characterization was modelled by using of finite element method (FEM). FEM is a method used for numerical solution of complex problems in different fields of engineering. From other side, the method aims to seperate to the sub regions of solution zones called finite element in many numbers, basic, small and connected each other. For the modelling have been utilized by computer software. Electrode and plane layout was drawn like in the same experimental setup by using of FEMM 4.2 program. The results of the mathematical model have been obtained with the informations like material specifications and boundary values. As a result of mathematical modelling in computer was established some parameters like the geometry of the electrode and plane, mesh of the finite elements, the potential distribution, equipotential lines, the electric field distribution, potential changes on the surface between the electrodes and planes, tangential electric field component on the surface between the electrodes and planes and the magnitude of the electrical field. The results obtained from the modelling showed that the largest electric field is formed by at least gap spacing between the electrodes and planes. Therefore, the results established by modelling show that risk of the sparkover and perforation increases when the electrode and the plane come to each other. According to the modelling results, electrode radius in the same gap spacing decreases and then electrical field increases. Thereby, risk of the sparkover and perforation increases. When results were evaluated in terms of partial discharge, it was seen that the results of partial discharge measurement in laboratory are parallel with the results established by using of the FEMM program. For example, partial discharge inception voltage increased when the gap spacing or radius of the electrode in the same gap spacing increased. Therefore, if the gap spacing or radius of the electrode in the same gap spacing reduced, the risk of the partial discharge increased under the same voltage and in same physical conditions. Generally, the results established by modelling with the help of FEMM 4.2 program covered with the results in laboratory. One of the important reason is that FEMM program using finite element method allows to obtain the solution as a result of the creation of the problem structure and entering material properties and boundary conditions. Using this program, quite realistic results can be obtained. This work which is extremely helpful in understanding the behavior of the system, can help in the design of the insulation systems or in the construction of high voltage systems. In addition, this study can be developed by creating different situations or conditions. More effective results can be achieved by including neglected or unpredictable rough edges. For example, using different electrode systems, performing experiments under different gas pressures, using different dielectric materials can be improved and more information about the partial discharge characteristic can be collected.