A Multi-Objective Sine Cosine Algorithm Based on a Competitive Mechanism and Its Application in Engineering Design...

Area of Science:

• Computational Intelligence
• Optimization Algorithms

Background:

• Multi-objective optimization problems (MOPs) are prevalent in real-world scenarios.
• Existing multi-objective evolutionary algorithms (MOEAs) face challenges in balancing convergence and diversity of non-dominated solutions.

Purpose of the Study:

• To propose an efficient multi-objective sine cosine algorithm with a competitive mechanism (CMOSCA) to address the convergence-diversity trade-off in MOPs.

Main Methods:

• The CMOSCA utilizes non-dominated sorting and crowding distance for agent ranking and selection.
• A novel position updating operator is developed using a competitive mechanism based on shift-based density estimation.
• Outstanding agents guide the evolutionary process, with competition winners integrated into the position update scheme.

Main Results:

• The CMOSCA demonstrated superior performance on DTLZ, WFG, and ZDT benchmark suites, achieving better convergence and diversity.
• Statistical results confirmed the CMOSCA's efficiency and effectiveness when applied to engineering design problems.

Conclusions:

• The proposed CMOSCA effectively balances convergence and diversity for multi-objective optimization.
• CMOSCA presents a promising approach for solving complex engineering design challenges.