Hybrid Taguchi-Levy flight dis-tribution optimization algorithm for solving real-world design optimization problems


Yildiz M., Panagant N., Pholdee N., Bureerat S., Sait S. M., YILDIZ A. R.

MATERIALS TESTING, vol.63, no.6, pp.547-551, 2021 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 63 Issue: 6
  • Publication Date: 2021
  • Doi Number: 10.1515/mt-2020-0091
  • Journal Name: MATERIALS TESTING
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.547-551
  • Keywords: Taguchi method, hybrid algorithm, Levy flight distribution optimization algorithm, optimum design, vehicle design, metaheuristic, STRUCTURAL DESIGN, OPTIMUM DESIGN, DIFFERENTIAL EVOLUTION, ROBUST DESIGN, CRASHWORTHINESS, MODEL
  • Bursa Uludag University Affiliated: Yes

Abstract

The Levy flight distribution optimization algorithm is a recently developed meta-heuristic. In this study, the Levy flight distribution optimization algorithm and the Taguchi method are hybridized to solve the shape optimization problem, which is the final step in developing optimum structural components. The new method is termed the hybrid Levy flight distribution and Taguchi (HLFD-T) algorithm. Geometric dimensions are used as design variables in the optimization, and the problem is aimed at mass minimization. The constraint in the problem is the maximum stress value. The well-known Kriging meta-modeling approach and a specifically developed hybrid approach have been coupled in this paper to find the component's optimal geometry. The results show that the proposed hybrid algorithm (HLFD-T) has more robust features than the ant lion algorithm, the whale algorithm, and the Levy flight distribution optimization algorithm for obtaining an optimal component geometry.