Article: TSALSHADE: improved LSHADE algorithm with tangent search Journal: International Journal of Computational Science and Engineering (IJCSE) 2024 Vol.27 No.6 pp.703 - 717 Abstract: The differential evolution (DE) algorithm is a widely recognised numerical optimisation technique. However, it exhibits certain limitations, including inadequate exploration and a tendency to get stuck in local minima. Even efficient DE variants like LSHADE encounter challenges when confronted with complex composite functions containing global optima that are hard to attain. In contrast, the Tangent search algorithm (TSA) has proven its efficacy in tackling intricate optimisation problems. TSA achieves this by employing the tangent flight operator, which facilitates escaping local minima while retaining exceptional exploration capabilities. This paper introduces a novel hybrid algorithm known as TSALSHADE. TSALSHADE combines elements from TSA and LSHADE to address the previous issues. In this hybrid paradigm, the tangent flight operator is integrated into the LSHADE framework, complemented by the introduction of a novel differential mutation operator. The primary strength of TSALSHADE lies in its capacity to effectively address intricate composite functions. To validate its performance, comprehensive experimental studies were conducted on the latest CEC 2022 benchmark functions. The results unequivocally demonstrate that TSALSHADE consistently produces promising and competitive outcomes across a spectrum of benchmark functions. This success can be primarily attributed to the algorithm's improved equilibrium between search exploration and exploitation. Inderscience Publishers - linking academia, business and industry through research

Title: TSALSHADE: improved LSHADE algorithm with tangent search

Authors: Abdesslem Layeb

Addresses: LISIA Laboratory, NTIC Faculty, Computer Science and its Application Department, University of Constantine 2, Algeria

Abstract: The differential evolution (DE) algorithm is a widely recognised numerical optimisation technique. However, it exhibits certain limitations, including inadequate exploration and a tendency to get stuck in local minima. Even efficient DE variants like LSHADE encounter challenges when confronted with complex composite functions containing global optima that are hard to attain. In contrast, the Tangent search algorithm (TSA) has proven its efficacy in tackling intricate optimisation problems. TSA achieves this by employing the tangent flight operator, which facilitates escaping local minima while retaining exceptional exploration capabilities. This paper introduces a novel hybrid algorithm known as TSALSHADE. TSALSHADE combines elements from TSA and LSHADE to address the previous issues. In this hybrid paradigm, the tangent flight operator is integrated into the LSHADE framework, complemented by the introduction of a novel differential mutation operator. The primary strength of TSALSHADE lies in its capacity to effectively address intricate composite functions. To validate its performance, comprehensive experimental studies were conducted on the latest CEC 2022 benchmark functions. The results unequivocally demonstrate that TSALSHADE consistently produces promising and competitive outcomes across a spectrum of benchmark functions. This success can be primarily attributed to the algorithm's improved equilibrium between search exploration and exploitation.

Keywords: numerical optimisation; differential evolution; LSHADE; tangent search algorithm; TSA; exploration and intensification search.

DOI: 10.1504/IJCSE.2024.142836

International Journal of Computational Science and Engineering, 2024 Vol.27 No.6, pp.703 - 717

Received: 05 Jan 2023
Accepted: 14 Jan 2024
Published online: 28 Nov 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: TSALSHADE: improved LSHADE algorithm with tangent search

Keep up-to-date