A novel quantum motivated particle swarm method with enhanced search strategy for electromagnetic optimization...

Area of Science:

• Electromagnetics
• Computational Intelligence
• Optimization Algorithms

Background:

• Parameter estimation in electromagnetic design is challenging for conventional methods.
• Existing optimization techniques often suffer from premature convergence and unstable parameters.
• Quantum-inspired optimization offers potential advantages for complex design problems.

Purpose of the Study:

• To propose a novel search-based quantum-inspired particle swarm optimization (SQPSO) for electromagnetic applications.
• To address limitations of existing methods, including premature convergence and parameter instability.
• To enhance solution quality, convergence speed, and robustness in electromagnetic design.

Main Methods:

• Developed a SQPSO algorithm incorporating a probability-based mutation mechanism guided by the global best solution for improved exploration.
• Implemented a tournament-driver selection strategy to enhance particle selection and search diversity.
• Introduced a dynamic parameter strategy to improve global search capability and algorithmic stability.

Main Results:

• The proposed SQPSO method demonstrated faster convergence compared to existing metaheuristic techniques.
• Achieved higher-quality solutions and improved robustness against local optima in benchmark tests and case studies.
• Validated the effectiveness of the probability-based mutation, tournament selection, and dynamic parameter strategies.

Conclusions:

• The SQPSO algorithm provides a superior approach for parameter estimation in electromagnetic design.
• The novel mechanisms enhance exploration, exploitation, and algorithmic stability, leading to better performance.
• SQPSO offers a robust and efficient alternative to traditional and other metaheuristic optimization techniques.