Improved special relativity search algorithm for optimal design of truss structures

Articles in Press

Document Type : Original Article

Authors

• Farnaz Salajegheh ¹
• Vahid Goodarzimehr ²

¹ Department of Civil Engineering, Shahid Bahonar University of Kerman, Kerman, Iran.

² Faculty of Civil Engineering and Architecture, Shahid Chamran University of Ahvaz,Ahvaz, Iran

10.22067/jfcei.2025.88012.1297

Abstract

Optimal design of structural engineering problems, particularly truss structures, is highly challenging due to the presence of design constraints such as stress and displacement. These constraints are often defined in a nonlinear and dynamic manner, which causes optimization algorithms to become trapped in local optima. In contrast, metaheuristic algorithms generate acceptable solutions for such problems using specialized programming techniques and operators. The standard Special Relativity Search (SRS) algorithm fails to solve these types of problems effectively due to the lack of coordination between two crucial parameters: exploitation and exploration, leading to convergence to local optima. In this study, two new techniques are introduced and implemented to enhance the performance of this algorithm. The gradient descent algorithm is employed as a complementary method to the SRS algorithm, with the movement step controlled using specific coefficients. To demonstrate the effectiveness of the proposed method compared to other advanced approaches, their results are compared. The findings indicate that the proposed algorithm has a higher convergence rate than previous methods and successfully establishes a proper balance between exploitation and exploration parameters.

Keywords

• Optimal Design
• Special Relativity Search Algorithm
• Metaheuristic Algorithm
• Soft Computing
• SRS

---