Research Area:  Metaheuristic Computing

Abstract:

Metaheuristic optimization methods; It is a well-known global optimization approach for large-scale search and optimization problems, commonly used to find the solution many different optimization problems. Slime mould optimization algorithm (SMA) is a recently presented metaheuristic technique that is inspired by the behavior of slime mould. Slow convergence speed is a fundamental problem in SMA as in other metaheuristic optimization methods. In order to improve the SMA method, 10 different chaotic maps have been applied for the first time in this article to generate chaotic values instead of random values in SMA. Using chaotic maps, it is aimed to increase the speed of SMA’s global convergence and prevent it from getting stuck in its local solutions. The Chaotic SMA (CSMA) proposed for the first time in this study was applied to 62 different benchmark functions. These are unimodal, multimodal, fixed dimension, CEC2019, and CEC2017 test suite. The results of the application have been comparatively analyzed and statistical analysis performed with the well-known metaheuristic optimization methods, particle swarm optimization and differential evolution algorithm, and recently proposed grey wolf optimization (GWO) and whale optimization algorithm (WOA). In addition, in the CEC2017 test suite, the CSMA method has been compared with the SMA, WOA, GWO, harris hawk optimization, archimedes optimization algorithm and COOT algorithms that have been proposed in recent years, and statistical analyzes have been made. In addition, CSMA has been tested in 3 different real-world engineering design problems. According to the experimental results, it was observed that CSMA achieved relatively more successful results in 62 different benchmark functions and real-world engineering design problems compared to other compared methods and standard SMA.

Keywords:  
Chaotic map
Metaheuristic optimization methods
Slime mould optimization

Author(s) Name:  Osman Altay

Publisher name:  Springer

DOI:  10.1007/s10462-021-10100-5

Volume Information:  volume 55, pages: 3979–4040 (2022)

Paper Link:   https://link.springer.com/article/10.1007/s10462-021-10100-5