Design of Microstrip Line Adaptor for W-band Microstrip Meander Line Slow Wave Structure Based on BP-GA Neural Network

In order to speed up the optimization of the transition structure from waveguide to microstrip meander line at W-band, a microstrip line adaptor from metal waveguide to microstrip line is designed based on neural network and genetic algorithm. The algorithms are using MATLAB-Python-CST, where MATLAB is applied to organize the swept parameter data retrieved from CST, which is optimized using BP-GA neural network programmed with python. The results show that the method can shorten design time and improve design accuracy. The feasibility of the method in transition structure optimization is verified with CST simulation.