Design of filtered proportional-integral-derivative controlled non-ideal buck converter system using a novel snake optimizer algorithm with a constrained objective function

This paper proposes a new metaheuristic algorithm called OSONM, an enhanced version of the snake optimizer (SO) algorithm. OSONM combines opposition-based learning (OBL) and the Nelder-Mead simplex search (NMSS) method to optimally control a filtered proportional-integral-derivative (FPID) controller for a non-ideal buck converter system. The proposed algorithm has been designed with the crossover frequency constraint to prevent signal noise in high frequencies while performing well under various disturbances. The effectiveness of the proposed algorithm is evaluated using statistical box plot analysis and convergence response comparisons with the SO algorithm. The performance of the OSONM-based FPID-controlled non-ideal buck converter system is also compared with the original SO algorithm-based and the pole placement (PP) method-based systems in terms of transient and frequency responses, disturbance rejection and robustness analysis. The results demonstrate that the OSONM-based system is more stable, efficient, and robust than the SO- and PP-based systems.