Red-Crowned Crane Optimization: A Novel Biomimetic Metaheuristic Algorithm for Engineering Applications.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Bio-inspired Computing

Background:

• Metaheuristic algorithms are crucial for solving complex optimization problems.
• Existing algorithms often struggle with balancing exploration and exploitation or avoiding local optima.

Purpose of the Study:

• To introduce a novel bio-inspired metaheuristic algorithm, the Red-crowned Crane Optimization (RCO) algorithm.
• To evaluate the performance of the RCO algorithm against established optimization techniques.

Main Methods:

• The RCO algorithm mathematically models four red-crowned crane behaviors: foraging, roosting, dancing, and escaping danger.
• The algorithm's exploration and exploitation capabilities are enhanced through these modeled behaviors.
• Performance is assessed using numerous benchmark functions (CEC-2005, CEC-2022) and practical engineering problems.

Main Results:

• The RCO algorithm achieved superior solutions for 74% of CEC-2005 and 50% of CEC-2022 test functions.
• It demonstrated fast convergence, high search accuracy, and effectiveness on high-dimensional problems.
• Wilcoxon signed-rank tests confirmed the RCO algorithm's significant advantage over competing algorithms.

Conclusions:

• The Red-crowned Crane Optimization algorithm is a highly effective and robust metaheuristic.
• Its bio-inspired design provides a strong balance between exploration and exploitation, leading to near-optimal solutions for complex problems.