Abstract
ABSTRACT The paper proposes an Improved Whale Optimization Algorithm (IWOA). Its performance is validated by solving 23 benchmark functions. Comparing the results of IWOA with well-known meta-heuristic algorithms shows its efficiency. Three non-parametric statistical tests, namely, Friedman, Friedman aligned and Quade tests are used to confirm the proposed algorithm’s superiority. IWOA is employed to design the parameters of a controller, namely, PID plus second-order derivative (PIDD2) for an automatic voltage regulator system (AVR). In fact, the proposed technique benefits from an evolutionary operator crossover to promote the diversity of solutions while maintaining a reasonable local search behavior. The results are compared with the results of similar algorithms including Particle Swarm Optimization, Genetic Algorithm, Teaching Learning-Based Optimization, Differential Evolution, Cuckoo Search algorithm and Artificial Bee Colony, demonstrating the advantages and the efficiency of the IWOA-PIDD2 controller. Robustness analysis of the optimal design obtained is conducted by varying the time constants of the AVR system component. The results proved that the proposed technique reliably outperforms most of the current techniques.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call