Ivy Optimization Algorithm Combining Sine-Cosine Operator and Adaptive T-Distribution and Its Engineering Application

Area of Science:

• Swarm Intelligence
• Nature-Inspired Optimization Algorithms
• Computational Intelligence

Background:

• The Ivy Optimization Algorithm (IVY) simulates plant phototropic growth for optimization tasks.
• Existing algorithms often face challenges with premature convergence and limited exploration-exploitation balance.
• Enhancing swarm intelligence algorithms is crucial for solving complex real-world problems.

Purpose of the Study:

• To propose an enhanced Ivy Optimization Algorithm (LSIVY) with improved performance.
• To integrate chaotic mapping, sine-cosine operators, and adaptive mutation for better optimization.
• To validate LSIVY's effectiveness against mainstream metaheuristics and in engineering applications.

Main Methods:

• Improved Logistics chaotic mapping with double arcsine transformation for diverse population initialization.
• Sine-cosine operator integrated into IVY's update mechanisms for adaptive exploration-exploitation.
• Adaptive t-distribution mutation strategy for enhanced local search and escaping local optima.

Main Results:

• LSIVY achieved 20-30 orders of magnitude higher accuracy and 4-6 orders lower standard deviation than standard IVY on benchmark functions.
• LSIVY ranked first on all CEC 2020 composite functions, reducing iterations by over 30% compared to native IVY.
• Engineering verification showed LSIVY reduced manufacturing cost by 9.2% for a pressure vessel and improved robustness across mechanical design problems.

Conclusions:

• The proposed LSIVY significantly enhances the performance of plant-inspired swarm intelligence algorithms.
• LSIVY demonstrates superior optimization accuracy, convergence speed, and robustness.
• LSIVY holds strong potential for practical applications in lightweight design and cost optimization.