Nozzle optimization of SF6 circuit breaker based on artificial neural network and genetic algorithm

Nozzle plays very important role to control the gas flow during the interruption for SF₆ circuit breaker (CB). Due to the higher non-linear global mapping relationship between interruption performance of SF₆ CB and its nozzle structural parameters, artificial neural network (ANN) and genetic algorithm (GA) were applied to the nozzle parameter optimization of SF₆ CB on the basis of the non-linear mapping properties of ANN and parallel processing, stochastic, and self-adapting search abilities of GA. And the parameter optimization system was established to study the influence of the nozzle structural parameter to the dielectric recovery. The application program is compiled in engineering computing language, which is used in calculating the parameter value predicted by neural network and the result of genetic algorithm optimization. The comparison and error analysis have been carried out between the results predicted by network and CAE simulated results, which shows that the BP network is stable and reliable. The optimized outcome, after verified by computer aided engineering (CAE) simulation, has been proved to be correct. It has been indicated that the nozzle structural parameter optimization method based on the artificial neural network and genetic algorithm approach is feasible. And this optimization strategy provides a feasible scheme for the complex structural optimization.