Advanced PID Parameter Tuning for Boost Converters using Metaheuristic Optimization Techniques

Hybrid HAGFB is a hybrid metaheuristic optimization algorithm developed for tuning the gains of a PID controller. Its core idea relies on simultaneously simulating four natural behaviors-ant pheromone tracking for memorizing good paths, grey wolf pack hunting for broad exploration of the search space, firefly light attraction for intensifying the search around better solutions, and bald eagle spiral motion for accurately approaching the optimum. The algorithm begins by generating a random population of candidate PID gain sets and evaluating their quality using an objective function such as ITAE. Based on the achieved performance, each candidate is assigned a pheromone value so that higher-quality solutions exert stronger influence in later iterations. Throughout the optimization, the method automatically alternates between an early exploration phase and a later exploitation phase, enabling an intelligent balance between global and local search to avoid premature convergence. By maintaining solution diversity through the combined inspiration sources and enhancing information exchange through social mechanisms like fighting and mating, HAGFB improves its ability to escape local optima and yields more stable and accurate PID tuning results, particularly for complex, multimodal error landscapes, compared with single-source approaches such as pure Ant Colony Optimization (ACO) or pure Grey Wolf Optimizer (GWO), and pure Bald Eagle Search (BES).

کلیدواژه ها:

Grey Wolf Optimizer (GWO) ، Bald Eagle Search (BES) ، PID controller ، fuel cell