Improved Black Widow Spider Optimization Algorithm Integrating Multiple Strategies.

Area of Science:

• Computational Intelligence
• Metaheuristic Optimization
• Algorithm Design

Background:

• The standard black widow spider optimization algorithm (BWOA) suffers from slow convergence and premature convergence to local optima.
• Addressing these limitations is crucial for effective application in complex optimization tasks.

Purpose of the Study:

• To propose an enhanced multi-strategy black widow spider optimization algorithm (IBWOA).
• To improve the convergence speed, global search ability, and overall performance of the BWOA.

Main Methods:

• Introduced Gauss chaotic mapping for population initialization to ensure initial diversity.
• Incorporated sine cosine strategy for iterative perturbation to enhance global search.
• Applied elite opposition-based learning to accelerate convergence.
• Integrated differential evolution mutation for reorganizing individuals with poor fitness.

Main Results:

• Verified the effectiveness of each strategy through benchmark functions (13 test functions and CEC2017).
• Demonstrated significant improvements in solution accuracy, performance, and convergence speed compared to existing algorithms.
• Successfully applied IBWOA to six practical constrained engineering problems, showcasing excellent optimization ability and scalability.

Conclusions:

• The proposed IBWOA effectively overcomes the limitations of the original BWOA.
• The multi-strategy enhancements lead to superior optimization performance and robustness.
• IBWOA shows strong potential for solving complex real-world engineering optimization problems.