Designing a Filtered Proportional–Integral–Derivative Controller With Disturbance Rejection for a Nonideal Buck Converter Utilizing an Upgraded Genetic Algorithm and Pattern Search

This research introduces an enhanced metaheuristic algorithm named GAPS, a combination of the genetic algorithm (GA) with tournament selection (TS) and the pattern search (PS) algorithm. The primary objective is improving GA's capacity for exploring and exploiting potential solutions. The proposed algorithm optimizes a Nonideal buck converter's output voltage controlled by a proportional–integral–derivative (PID) controller with an added low-pass filter (PID-N-F). The algorithm is carefully designed, incorporating a strategically imposed crossover frequency constraint to counteract signal noise at higher frequencies. This approach ensures robust performance in the presence of various disturbances. The algorithm's effectiveness is evaluated using statistical box plots and by comparing convergence rates with the standard GA method. It is also compared how the GAPS-optimized PID-N-F controller performs in the buck converter relative to the standard GA approach and classical pole placement (PP) method. The comprehensive evaluation covers robustness analysis, frequency and transient responses, load and input voltage variation as disturbance rejection. The results indicate that the GAPS-based system performs better than the GA- and PP-based systems in various aspects. These findings affirm the GAPS-based system's superior stability, efficiency, and robustness relative to the GA- and PP-based alternatives.