A Multi-Strategy Parrot Optimization Algorithm and Its Application.

Area of Science:

• Computational intelligence
• Engineering optimization
• Metaheuristic algorithms

Background:

• Intelligent optimization algorithms are vital for complex engineering challenges.
• The standard Parrot Optimization (PO) algorithm suffers from local-optimum trapping and slow convergence.
• Enhancements are needed to improve PO's performance and applicability.

Purpose of the Study:

• To introduce a novel enhanced Parrot Optimization algorithm, the Chaotic-Gaussian-Barycenter Parrot Optimization (CGBPO).
• To address the limitations of the standard PO algorithm, specifically premature convergence and local optima.
• To evaluate the performance of CGBPO on benchmark functions and practical engineering problems.

Main Methods:

• Incorporation of chaotic logistic mapping for enhanced initial population diversity.
• Application of Gaussian mutation to prevent premature convergence to local optima.
• Integration of barycenter opposition-based learning to expand the search space during iterations.

Main Results:

• CGBPO demonstrated superior performance over seven other algorithms on CEC2017 and CEC2022 benchmark suites.
• The proposed algorithm achieved faster convergence, higher solution accuracy, and improved stability.
• In an indoor visible light positioning simulation, CGBPO provided more accurate position estimations than the standard PO algorithm.

Conclusions:

• The Chaotic-Gaussian-Barycenter Parrot Optimization (CGBPO) algorithm effectively overcomes the limitations of the standard PO.
• CGBPO shows significant improvements in convergence speed, accuracy, and stability for optimization tasks.
• The enhanced algorithm exhibits superior adaptability and robustness in practical engineering applications, including visible light positioning.