Computational framework for studying charge transport in high-voltage gas-insulated systems

Doktorsavhandling, 2017

Electrical switchgear industry dealing with transmission and distribution of power is strongly affected by the changing dynamics of climate control. Their main contribution to the global warming is through SF6 gas, having a GWP of 23500 (times of CO2). SF6 plays an important role in providing electrical insulation for these high voltage systems. So, there is a need to find an alternative ‘green’ insulation system. The search is hindered because of lack of design criteria for dielectric withstand in highly inhomogeneous fields, as seen typically in gas- insulated switchgears (GIS). This needs a physical understanding of charge transport and dynamics under such electric fields. Additionally, the replacement gases are characterized by lower dielectric withstand performance and hence needs solid insulation together with gaseous insulation to enhance the electrical properties. This raises a need to develop combined gas-solid physical framework for charge transport and interaction in such systems. This includes initiation and propagation of electrical discharges in gas phase, interactions of produced gas discharge plasma with solid dielectric surfaces, charge transport though solid material, etc. The thesis focuses on developing a numerical simulation framework of discharge initiation and propagation, incorporating various charge generation processes for different insulation systems under highly inhomogeneous electric fields. The mathematical model of non-thermal electrical discharges in the present thesis is composed of a set of time dependent highly non-linear partial differential equations (PDEs) describing charge transport by drift and diffusion under the influence of electric field incorporating various reactions (electron impact ionization, electron attachment, recombination, etc.). The reaction rate coefficients are calculated by two term Boltzmann approximation of electron energy distribution function. Another coupled set of PDEs for radiative transport process (photoionization), provides additional reaction sources. Solid insulation modelling includes additional PDEs for electron and hole transport inside the solid insulation. Charge injection from gaseous medium together with charge accumulation at the solid-gas interface is accounted for. The PDEs are solved by custom finite element logarithmic weak formulation with Petrov- Galerkin stabilization. This method is implemented in commercial code (COMSOL- Mathematical module). Adaptive mesh refinement techniques are used to speed up the calculation time. The developed method is validated against reference experimental data (nanosecond resolved) and published cases. Later, non-axial discharges, 3d branching, hybrid insulation study cases are presented.