Chaos-Integrated Difference-Enhanced Greater Cane Rat Algorithm and Its Application

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristic Computing

Background:

• The Greater Cane Rat Algorithm (GCRA) faces challenges with population distribution, local optima, and balancing exploration/exploitation.
• Existing optimization algorithms often struggle with premature convergence and insufficient accuracy.

Purpose of the Study:

• To propose a Chaos-Integrated Difference-Enhanced Greater Cane Rat Algorithm (CEGCRA) addressing GCRA's limitations.
• To enhance population diversity, optimize exploration-exploitation balance, and improve convergence speed and accuracy.

Main Methods:

• Utilizing a piecewise chaotic map for improved initial population distribution and diversity.
• Implementing an accumulated difference foraging strategy for adaptive search direction and step size adjustment.
• Refining the exploration-exploitation switching mechanism and boundary constraint handling.

Main Results:

• CEGCRA achieved an average 35.3% reduction in optimal fitness value and 22.7% in standard deviation compared to GCRA.
• Convergence speed increased by an average of 28.9% on benchmark test suites.
• CEGCRA outperformed GCRA, PSO, DE, and SSA in optimization accuracy, convergence speed, robustness, and constraint handling for engineering problems.

Conclusions:

• The CEGCRA effectively overcomes the limitations of the original GCRA, offering enhanced performance.
• The algorithm demonstrates significant improvements in accuracy, speed, and robustness for complex, high-dimensional, and constrained optimization problems.
• CEGCRA shows strong potential for solving real-world engineering design challenges.