| Global Energy Interconnection was proposed, and the construction and operation of the ultra-high voltage grid became the focal point. The large power transformer was one of the most basic facilities in the ultra-high voltage grid, which of the reliable and security operation had a huge impact on the entire grid. The transformer oil was polluted inevitably by the metal particles in the manufacturing, transportation, site filling, operation, maintenance, etc. The metal particle defect was one of the important factors of accidents in the power transformer. Partial discharge(PD) caused by metal particles in the insulating medium was researched a lot at home and abroad in order to ensure the safe and stable operation of power transformer. But the discharge characteristics of metal particles were studied in stationary state by home and abroad scholars at present. The insulating oil in large power transformer was in a flow state rather than a static state on account of the submersible pump and the temperature difference. The PD properties and the influencing factors of the metal particles in flow oil were different from ones in static oil, which still needed further research. The physical experimental model was designed considering the typical oil tract of large power transformer and the field distribution in the transformer oil in this paper. The mathematical simulation model of movement of the metal particles was built in the flow oil. The motion characteristics and distribution of metal particles were analyzed theoretically in the flow oil. The PD characteristics and influencing factors and mechanisms of the metal particles were studied in the transformer oil in flow state. Details were as follows:(1) The parameters in experiment were selected reasonably according to the characteristics of typical oil tract in large power transformer, and the velocity of the oil, size and number of metal particles, oil temperature, and the electric field of oil etc. in operation of transformer. The experimental system and the experimental analysis of the PD of the metal particles were designed in the insulation oil under the flow state. The electric field of the oil tract was simulated by the finite element method. The simulation and experimental results showed that the experiment system designed of PD of the metal particles in the insulated oil under the flow state could simulate effectively the PD of the metal particles in real power transformer. These were contributed to the study on PD and its influencing factors of the metal particles in the insulating oil under the flow state.(2) After the detailed force analysis of metal particle based on solid-liquid two-phase flow, when metal particles move in oil, and simulation analysis of metal particles trajectories and distribution characteristics with Fluent, a classic software of fluid mechanics, the influence on metal particles trajectories and distribution characteristics were analyzed of particle size, the oil speed, the electric field intensity and the number of particle. All the analysis supported the follow results that the number of colliding between metal particles and the bottom electrode markedly increase, with the increase of particle size; the increase of oil speed reduced the frequency of collisions between metal particles and electrodes by reducing the time in electric field area; with the increase of electric field in oil tract, the number of collisions between metal particles group and electrodes significantly increased, and the frequency of oscillation in the vertical direction was much higher; the increase in the number of metal particles obviously led to more collisions between metal particles and electrodes.(3) Based on the experimental system, the discharge experiments of a large number of metal particles were accomplished. At first, a single PD pulse current wave arising from metal particles in flowing insulating oil was compared to that in static insulating oil. Then, single PD UHF signals in two conditions were analyzed and transformed by FFT, and ?-U-N 3-D diagrams were established. Statistics characteristic parameters, including the number of discharge, the magnitude of discharge, the time interval of discharge, the equivalent cumulative discharge and signal entropy, were drew. At last, four types of discharge in flowing insulating oil were summarized. The study showed that: for the dispersity of metal particle discharge, it is not available to distinguish discharges in two conditions by one single discharge wave. However, the differences of their statistics characteristic parameters in ?-U-N 3-D diagrams are very clear. Compared to discharge in flowing condition, static discharge has the greater number of discharge, the bigger magnitude of discharge, the smaller time interval of discharge, the larger equivalent cumulative discharge and also the bigger signal entropy. In another word, PD in static station is easier than that in flowing condition, so the movement of oil reduces the intensity of metal particles discharge. Discharge in flowing oil can be divided into four types, including charge transfer of ball-electrode, charge transfer between metal particles, micro-discharge and corona discharge.(4) Based on a significant amount of experiment data, inception PD voltage, PRPD model and other statistics characteristic parameters were drew, and the influence on PD characteristic of the oil speed, the particle size and the number of metal particles was studied. According to the above four types of discharge, it drew some explanations of the influence of various parameters on metal particles discharge. The study shows that with the increase of oil speed, the intensity of metal particles PD reduces, while with the bigger metal particle size and the more number of metal particles, the intensity of metal particles PD increases. The reasons for the above results are the difference influence on trajectories and distribution of metal particles, which will affect four types discharge of metal particle, leading to variation of statistics characteristic parameters. |
PDF Full Text Request