Enhancing PSO: Enabling Dynamic Solution Space Adjustment for Effective Optimization

This paper introduces an innovative adaptive solution methodology that integrates with the Particle Swarm Optimization (PSO) algorithm. The methodology is specifically designed to autonomously adjust the solution space, particularly in dimensions crucial to the design space. In situations where the global optimum is beyond the initially defined solution space, the particles may struggle to attain the desired fitness value despite extensive searching. To overcome this challenge, the proposed methodology enables the optimization algorithm to operate efficiently, even when the global optimum lies outside the initial bounds. The performance of the adaptive solution space methodology undergoes evaluation using both mathematical benchmark functions and an antenna optimization application. Additionally, the paper conducts a comparative analysis with a method involving the expansion of the solution space in all dimensions after a specified iteration. The results not only showcase the effectiveness of the proposed method in dynamically expanding the solution space but also demonstrate a significant improvement in convergence speed compared to an unspecified expansion method.