An Adaptive Two-Layer Control System Optimization for Buck Converters Using Evolutionary Algorithm

Traditional evolutionary algorithms typically rely on fixed weight coefficients in the objective function, which are determined by system topology and hardware configuration. When these parameters change, the coefficients must be readjusted, a process that is often complex and time-consuming. To address this issue, this paper proposes an adaptive two-layer control system optimization method for buck converters. The method integrates Particle Swarm Optimization (PSO) and Differential Evolution (DE). Specifically, PSO dynamically adjusts the weight coefficients in the objective function to accommodate load variations and voltage fluctuations without relying on fixed system configurations, while DE optimizes the proportional gain (Kp) and integral gain (Ki) of the PI controller to improve control sensitivity. Through the collaboration of these two algorithms, the control strategy can be adaptively adjusted, thereby enhancing optimization efficiency and dynamic response. Compared with traditional methods that rely on expert experience for weight assignment, the proposed approach is more flexible and achieves faster response, lower overshoot, and smaller steady-state error in buck converters.