Network Based Control System Design Optimization for Buck Circuits Using Genetic Algorithm

Optimization of control systems for power electronic systems using evolutionary algorithms has gained significant attention in current research. However, the effectiveness of such methods heavily relies on the accuracy of initial data collection, often leading to prolonged time consumption with higher precision. This paper presents an approach for optimizing control systems of Buck circuits by integrating BP neural networks to address both data accuracy and time efficiency using genetic algorithm. Initially, data corresponding to controller parameters and system indicators are collected. Subsequently, BP neural networks are utilized to model the relationship between multiple controller parameters and control indicators. After that, the genetic algorithm is employed to optimize system controller parameters continuously. Finally, the proposed method is implemented in a Matlab/Simulink control system, and its efficacy is validated through simulation results.