Comparative Study of Gorilla Troops Optimizer and Stochastic Fractal Search with fuzzy dynamic parameter adaptation

Abstract

Metaheuristics has a very important role today in solving optimization problems; the vast majority of these methods are based on principles that imitate natural processes to achieve their results. The objective of this research is the analysis of the adaptability and stability of two bio-inspired methods, proposing a comparison between two optimization algorithms to evaluate and compare the performance and effectiveness of the algorithms in different optimization problems, the first, inspired by the social behavior of gorillas, which is called Artificial Gorilla Troops Optimizer (GTO), which is mathematically formulated to achieve exploration and exploitation in a given search space. The second algorithm is the one nature-inspired by imitating fractal behavior, known as Stochastic Fractal Search (SFS), where each of the particles moves stochastically until the objective function is found. By comparing both methods using benchmark functions, in this case CEC'2017 functions and performing the corresponding statistical analysis, we can conclude that with the GTO method, we obtained better results, since they are closer to the global optimum of the functions in comparison. with the SFS algorithm.