Journal of Bionic Engineering ›› 2023, Vol. 20 ›› Issue (4): 1791-1827.doi: 10.1007/s42235-022-00330-w
BEESO: Multi-strategy Boosted Snake-Inspired Optimizer for Engineering Applications
Gang Hu1,2; Rui Yang1; Muhammad Abbas3; Guo Wei4
- 1 Department of Applied Mathematics, Xi’an University of Technology, Xi’an 710054, China 2 School of Computer Science and Engineering, Xi’an University of Technology, Xi’an 710048, China 3 Department of Mathematics, University of Sargodha, Sargodha 40100, Pakistan 4 University of North Carolina at Pembroke, Pembroke, NC 28372, USA
BEESO: Multi-strategy Boosted Snake-Inspired Optimizer for Engineering Applications
Gang Hu1,2; Rui Yang1; Muhammad Abbas3; Guo Wei4
- 1 Department of Applied Mathematics, Xi’an University of Technology, Xi’an 710054, China 2 School of Computer Science and Engineering, Xi’an University of Technology, Xi’an 710048, China 3 Department of Mathematics, University of Sargodha, Sargodha 40100, Pakistan 4 University of North Carolina at Pembroke, Pembroke, NC 28372, USA
摘要： This paper presents an efficient enhanced snake optimizer termed BEESO for global optimization and engineering applications. As a newly mooted meta-heuristic algorithm, snake optimizer (SO) mathematically models the mating characteristics of snakes to find the optimal solution. SO has a simple structure and offers a delicate balance between exploitation and exploration. However, it also has some shortcomings to be improved. The proposed BEESO consequently aims to lighten the issues of lack of population diversity, convergence slowness, and the tendency to be stuck in local optima in SO. The presentation of Bi-Directional Search (BDS) is to approach the global optimal value along the direction guided by the best and the worst individuals, which makes the convergence speed faster. The increase in population diversity in BEESO benefits from Modified Evolutionary Population Dynamics (MEPD), and the replacement of poorer quality individuals improves population quality. The Elite Opposition-Based Learning (EOBL) provides improved local exploitation ability of BEESO by utilizing solid solutions with good performance. The performance of BEESO is illustrated by comparing its experimental results with several algorithms on benchmark functions and engineering designs. Additionally, the results of the experiment are analyzed again from a statistical point of view using the Friedman and Wilcoxon rank sum tests. The findings show that these introduced strategies provide some improvements in the performance of SO, and the accuracy and stability of the optimization results provided by the proposed BEESO are competitive among all algorithms. To conclude, the proposed BEESO offers a good alternative to solving optimization issues.