Probability Distribution Model Based on Statistical Properties of DC Basin Insulators Surface Flashover
摘要
Gas insulated metal-enclosed transmission line (GIL) is an important development direction for high-voltage direct current (DC) transmission system, and it is of great significance to study the flashover characteristics of DC GIL insulators and establish the flashover probability distribution model. In this paper, a flashover experimental platform based on scaled down model basin insulators was established. Short-time continuous-rising voltage tests for two kinds of epoxy-formulated insulators and long-time withstand tests under different voltages were performed, and the data were analyzed by two-parameter Weibull distribution. The results show that, the flashover probability increases with increasing voltage, and insulators with lower resistivity have higher average withstand voltage and less dispersibility. The flashover probability increases with increasing voltage duration, and insulators under lower voltage have longer average withstand time and lower dispersibility. The probability distribution model determined by the applied voltage and duration is established by fitting and analyzing the statistical parameter of the withstand time. The predicted steady state flashover voltage U0.1% is 381 kV. It provides a feasible numerical analysis method for flashover voltage prediction with low probability in practice.