Electric field analysis and optimal design of inner electrodes for HVDC wall bushing

A new-type of SF₆ gas-insulated HVDC wall bushing has been manufactured. The double-layer electrodes are introduced to grade electric fields instead of traditional capacitor foils. Therefore, the electric-field distribution of inner insulation is quite different from traditional productions as well as structure design. This paper presents finite element analysis (FEA) of this type of bushing. The features of inner electric-field distribution with main influence factors are discussed. In order to reduce electric-field strength on the surface of inner electrodes, this paper adopts generalized regression neural network (GRNN) and fast non-dominated sorting genetic algorithm (NSGA-II) to optimize the inner electrode design. First, optimization variables and constraint conditions are determined by electrical and geometrical specification. Then, GRNN is employed to rebuild and predict the objective functions. Finally, after NSGA-II optimization, the Pareto solutions are classified using a clustering algorithm in order to acquire the ultimate solution. Comparative checking for initial and optimal inner electrode design is performed by FEA. The results show that the electric-field strength on the surface of electrode is reduced. The electric field analysis and optimization results in this paper can give some directions for the design of SF₆ gas-insulated HVDC wall bushing.