Knowledge-Driven Automatic Synthesis of Microwave Circulator with Machine-Learning -Assisted Optimization

Circulator is an significant component in microwave and radio-frequency systems. In this paper, a machine-learning-assisted automatic synthesis algorithm based on prior knowledge is proposed. A circulator design example is used to verify the feasibility and effectiveness of the algorithm. Compared with traditional design methods, the proposed algorithm in this paper can achieve better performance and a shorter design time.