Core Concepts
The proposed enhanced Parameter Optimal State Transition Algorithm (POSTA) utilizes historical information more efficiently by integrating Nelder-Mead simplex search and quadratic interpolation, leading to faster convergence speed and higher solution accuracy compared to the original POSTA.
Abstract
The content describes an enhanced version of the Parameter Optimal State Transition Algorithm (POSTA), a metaheuristic optimization method. The key points are:
The original POSTA suffers from slow convergence speed and low solution accuracy due to insufficient utilization of historical information.
The proposed enhanced POSTA, named NM-POSTA, integrates Nelder-Mead (NM) simplex search and quadratic interpolation (QI) to better utilize the historical information.
NM simplex search is used to store and utilize the historical solutions more comprehensively compared to the linear transformation in the original POSTA.
QI is introduced in the later stage of the search to strengthen the exploitation capacity by approximating the objective function using historical solutions.
Experimental results on benchmark functions demonstrate that the enhanced POSTA outperforms the original POSTA in terms of convergence speed and solution accuracy.
The proposed method successfully combines the merits of POSTA, NM simplex search, and QI, achieving better overall performance.
Stats
The number of function evaluations (FEs) required for the POSTA families to reach the global optimum on the benchmark functions.