A Multi-Objective Optimization Problem Solving Method Based on Improved Golden Jackal Optimization Algorithm and Its...

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Traditional golden jackal optimization (GJO) suffers from slow convergence, low accuracy, and susceptibility to local optima.
• Existing optimization algorithms often struggle with complex problems requiring both global exploration and precise local exploitation.

Purpose of the Study:

• To propose a novel enhanced golden jackal optimization algorithm (SCMGJO) addressing the limitations of the traditional GJO.
• To improve the convergence speed, accuracy, and global search capabilities of the golden jackal optimization algorithm.
• To validate the effectiveness of the SCMGJO algorithm on benchmark functions and real-world engineering problems.

Main Methods:

• Introduced tent mapping reverse learning for population initialization to enhance global exploration.
• Incorporated sine and cosine strategies in prey position updates for improved exploration.
• Integrated Cauchy mutation for perturbation and optimal solution updates to refine local search.
• Evaluated the SCMGJO algorithm using 23 benchmark test functions.

Main Results:

• The SCMGJO algorithm demonstrated significant improvements in convergence speed and solution accuracy compared to traditional methods.
• Experimental results on benchmark functions indicated superior performance of SCMGJO.
• The algorithm successfully solved engineering design problems, including spring design, truss design, and UAV path planning.

Conclusions:

• The proposed SCMGJO algorithm effectively overcomes the limitations of the traditional GJO, offering enhanced optimization capabilities.
• SCMGJO exhibits robust performance in terms of speed, accuracy, and escaping local optima.
• The algorithm shows strong potential for practical application in complex engineering optimization tasks.