DGS-SCSO: Enhancing Sand Cat Swarm Optimization with Dynamic Pinhole Imaging and Golden Sine Algorithm for improved
Oluwatayomi Rereloluwa Adegboye1, Afi Kekeli Feda2, Oluwaseun Racheal Ojekemi3
1Management Information System Department, University of Mediterranean Karpasia, Mersin-10, Turkey.
Scientific Reports
|January 17, 2024
Summary
This study introduces DGS-SCSO, an enhanced optimizer improving Sand Cat Swarm Optimization (SCSO) by integrating Dynamic Pinhole Imaging and Golden Sine Algorithm. DGS-SCSO effectively addresses convergence issues and demonstrates superior performance in optimization tasks.
Area of Science:
- Computational Intelligence
- Optimization Algorithms
- Metaheuristic Computing
Background:
- Original Sand Cat Swarm Optimization (SCSO) faces limitations like local optima entrapment and premature convergence.
- Need for enhanced optimization techniques to improve global exploration and exploitation capabilities.
Purpose of the Study:
- Introduce DGS-SCSO, a novel optimizer enhancing SCSO.
- Mitigate convergence issues and improve performance on complex optimization problems.
Main Methods:
- Integration of Dynamic Pinhole Imaging for enhanced global exploration.
- Incorporation of Golden Sine Algorithm for improved exploitation and convergence.
- Systematic performance evaluation on benchmark functions and engineering problems.
Main Results:
- DGS-SCSO demonstrates significant superiority over the original SCSO algorithm.
- Achieved high efficiency rates (59.66% in 30D, 76.92% in 50-100D) on optimization functions.
- Competitive results on practical engineering problems, validated by statistical tests.
Conclusions:
- DGS-SCSO effectively overcomes limitations of the original SCSO algorithm.
- The proposed optimizer shows robust performance and significant improvements in efficiency and convergence.
- Validated efficiency and significant improvements through Wilcoxon Rank Sum and Friedman Tests.
