A stable constraint-guided artificial bee colony algorithm (SCG-ABC) for turbulence grid shape optimization
Applied Soft Computing, cilt.203, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 203
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.asoc.2026.116144
- Dergi Adı: Applied Soft Computing
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Applied Science & Technology Source, Compendex, INSPEC
- Anahtar Kelimeler: Artificial Bee Colony (ABC), Constraint-guided optimization, Non-Dominated Sorting Genetic Algorithm II (NSGA-II), Shape optimization, Stability, Turbulence grid
- Erciyes Üniversitesi Adresli: Evet
Özet
This study presents a Stable Constraint-Guided Artificial Bee Colony (SCG-ABC) framework for turbulence grid shape optimization and demonstrates its applicability to constrained engineering design problems. Turbulence grids are widely used in wind tunnel experiments to generate controlled atmospheric turbulence, and their geometric design requires the simultaneous satisfaction of several geometric and flow-development constraints. The SCG-ABC framework was applied to the optimization of turbulence grid geometry defined by grid bar thickness and mesh size parameters. The optimization process considered porosity requirements and downstream flow-development constraints to identify feasible grid configurations. To evaluate its performance, the proposed approach was compared with the widely used Non-Dominated Sorting Genetic Algorithm II (NSGA-II) algorithm under identical computational conditions. Both algorithms were executed in 30 independent runs, and the results were assessed using statistical indicators including mean, standard deviation, minimum, median, and maximum values. The results showed that both algorithms successfully identified feasible turbulence grid configurations within the prescribed design space. However, SCG-ABC produced more concentrated and repeatable solutions with lower variability in design variables, indicating a more stable convergence behavior. Furthermore, the proposed framework achieved approximately 30% lower computational time than NSGA-II while maintaining consistent constraint satisfaction across repeated runs.