An Improved Multi-Strategy Crayfish Optimization Algorithm for Solving Numerical Optimization Problems.

Area of Science:

• Computational Intelligence
• Metaheuristic Optimization
• Swarm Intelligence

Background:

• The crayfish optimization algorithm (COA) is a novel metaheuristic inspired by crayfish behaviors.
• COA exhibits good performance but struggles with slow convergence and premature local optima.
• Addressing these limitations is crucial for enhancing its applicability in complex optimization scenarios.

Purpose of the Study:

• To propose an improved multi-strategy crayfish optimization algorithm (IMCOA) for numerical optimization.
• To enhance the global exploration and local exploitation balance of the original COA.
• To improve convergence speed, accuracy, and the ability to escape local optima.

Main Methods:

• Introduced a cave candidacy strategy and a fitness-distance balanced competition strategy to improve summer heat avoidance and competition phases.
• Modified the foraging phase with a food covariance learning strategy to boost population diversity and accuracy.
• Incorporated an optimal non-monopoly search strategy to refine the global best solution.

Main Results:

• IMCOA demonstrated a superior balance between exploration and exploitation compared to COA and other algorithms.
• Experiments on CEC2017 and CEC2022 test suites showed significant improvements in convergence speed and optimization accuracy.
• Statistical analyses confirmed IMCOA's enhanced performance and robustness, outperforming traditional COA.

Conclusions:

• IMCOA effectively addresses the shortcomings of the original COA, particularly in convergence speed and local optima avoidance.
• The proposed strategies enhance population diversity and the ability to find global optima.
• IMCOA shows practical potential for solving real-world engineering design optimization problems.