@phdthesis{Bao2014, author = {Bao, Han}, title = {Optimal design of 245kV SF₆ bushing by using genetic algorithm}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:co1-opus4-31751}, school = {BTU Cottbus - Senftenberg}, year = {2014}, abstract = {Currently more and more researches of high voltage bushings are focused on the requirements for a simple structure, compactness and light-weight. The general operation of SF6 gas-filled bushings (SF6 bushings) is satisfying the requirements, but in the operation and design of SF6 bushings still many improvements may be possible. The minimizing of dimension might enhance the electric field strength (E) on the crucial points of bushings, which may lead to partial discharge, flashover and even break down. Electric field distribution along the surface mainly depends on contour design, besides the effect of contamination. This dissertation mainly describes the optimization of the bushings design by genetic algorithm. First, a model of SF6 bushing was developed and simulated for the theory purpose. Then, the potential breakdown problems were defined and the mechanisms of the potential breakdown were explained. Afterwards, the dissertation proposes an approach, i.e. genetic algorithm to optimize the contour design of SF6 bushings. The approach improved the execution efficiency by accessing the fitness values of searched solutions during the optimization process. To verify the effectiveness of the genetic algorithm, it has been applied to minimize the electric field strength at the critical positions. Furthermore, the critical points of SF6 bushings were optimized. Several new structures of the ground electrode were proposed and optimized separately by genetic algorithm. A new curve, i.e. cubic spline was applied to the contour of the top flange to avoid the influence of the triple points. By optimization E on the surface of top flange was minimized. The potential distribution on the surface of insulator was optimized by a new structure. By the genetic algorithm the contour of composite weather sheds (WS) was optimized as water-drop form. In summary, a more uniform potential distribution along the surface of weather sheds and minimal values of electric field strength at critical points can be derived effectively by genetic algorithm. In addition, a smaller dimension of SF6 bushing was obtained in comparison with presently available ones.}, subject = {Optimized contour design; SF6 bushing; Streamer theory; Ansoft Maxwell; B{\´e}zier curve; Optimale Gestaltung; SF6 Durchf{\"u}hrung; Streamer Theorie; Ansoft Maxwell; B{\´e}zier Kurve; Hochspannungsanlage; Schwefelhexafluorid; Genetischer Algorithmus; Optimierung}, language = {en} }