Research On The Stochastic Model Of Long Air Gap Discharge Channels

The fractal geometry is a powerful tool for describing the figures with the self-similarity structure. The actual observation results believe that the gas discharge often generates cabined narrow discharge channels, whose strongly branches form a complicated random figures. As the gas discharge channel of random long air gap exhibits the self-similarity, this paper introduces the fractal geometry to the field of the azig-zag and branch phenomena simulations. It is significant and applicable for founding more accurate and proposing more reasonable design method for lightning shielding performance of the HVDC transmission lines .A stochastic model which considering the growth time of gas discharge streamer is found, which is based on NPW model. It introduces physical mechanism of the stochastic model of streamer growth, presents computer simulation steps. First of all, according to the peek corona inception criterion,it judges the corona inception; then, it calculates the area electric potential distribution. To the discharge outside ,its channel electric potential is calculated by Laplace equation, while that of the discharge inside channel electric potential is calculated by Poission equation, continuity eqution and Ohm's law. At last, it calculates the streamer growth time ,and choose the shortest time point as the next discharge point.It selects the rod-plane gap as the simulation electrode and analyses the relationship between the developing probability exponentηand fractal dimension, breakdown time and the space charge of the streamer zone. It also analyses the influence of the rod radius on the corona inception time, fractal dimension of streamer zone, and space charge of the streamer zone, as well as the influence of the conductivity of streamer zone on the fractal dimension of streamer zone and the space charge of the streamer zone; then it selects the rod-conductors gap as simulation electrode object and analyses three streamer growth time criterion ,which are natural logarithm form, exponent form, power function form, streamer growth .In order to compare the model and experiment, it designs the simulation experiment objects, DC charge insulation circuit, lightning shielding performance, and do a computer simulation on the experiment of lightning shielding performance. It also compares with experiment results and simulation results, and does some analyses about why the model simulation results can not meet the experiment results ideally.