Sigmoid-Function-Based Adaptive Pelican Optimization Algorithm for Global Optimization

Abstract

This paper introduces the Sigmoid-function-based Adaptive Pelican Optimization Algorithm (MPOA), an enhanced version of the traditional Pelican Optimization Algorithm (POA) aimed at improving the POA's performance. Inspired by the hunting behavior of pelicans, the POA features two main strategies: the Exploration phase and the Exploitation phase. The Exploration phase involves searching new areas within the solution space, while the Exploitation phase focuses on refining the optimal solution space to achieve convergence. However, the Exploitation phase is inefficient, leading to slower convergence rates when striving for a global optimum. The MPOA incorporates an adaptive inertia weight mechanism that leverages the sigmoid function to balance exploration and exploitation throughout the optimization process. This adaptive approach ensures an efficient transition between searching for new solution areas and refining existing ones, thereby enhancing the overall optimization process. The algorithm was tested using a set of widely recognized standard benchmark functions to assess its performance. The results demonstrated that the MPOA significantly improved both convergence speed and solution quality compared to the original POA. Additionally, the MPOA outperformed other traditional optimization algorithms, such as Particle Swarm Optimization (PSO) and Genetic Algorithms (GA), in terms of achieving better optimization results. These findings suggest that the proposed MPOA provides an efficient optimization approach, leading to faster convergence and higher-quality solutions.