Preliminary study on the modelling of negative leader discharges

Abstract

Nowadays, there is considerable interest in understanding the physics underlying positive and negative discharges because of the importance of improving lightning protection systems and of coordinating the insulation for high voltages. Numerical simulations of positive switching impulses made in long spark gaps in a laboratory are achievable because the physics of the process is reasonably well understood and because of the availability of powerful computational methods. However, the existing work on the simulation of negative switching discharges has been held up by a lack of experimental data and the absence of a full understanding of the physics involved. In the scientific community, it is well known that most of the lightning discharges that occur in nature are of negative polarity, and because of their complexity, the only way to understand them is to generate the discharges in laboratories under controlled conditions.The voltage impulse waveshape used in laboratories is a negative switching impulse. With the aim of applying the available information to a self-consistent physical method, an electrostatic approximation of the negative leader discharge process is presented here. The simulation procedure takes into consideration the physics of positive and negative discharges, considering that the negative leader propagates towards a grounded electrode and the positive leader towards a rod electrode. The simulation considers the leader channel to be thermodynamic, and assumes that the conditions required to generate a thermal channel are the same for positive and negative leaders. However, the magnitude of the electrical charge necessary to reproduce their propagation and thermalization is different, and both values are based on experimental data. The positive and negative streamer development is based on the constant electric field characteristics of these discharges, as found during experimental measurements made by different authors. As a computational tool, a finite element method based software is employed. The simulations are compared with experimental data available in the literature.