CMGWO: Grey wolf optimizer for fusion cell-like P systems

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Bio-inspired Computing

Background:

• The Grey Wolf Optimizer (GWO) is a popular metaheuristic algorithm but suffers from premature convergence and local optima.
• Cell-like P systems offer parallel computation and inter-cell communication, providing a framework to enhance swarm intelligence algorithms.

Purpose of the Study:

• To propose an enhanced Grey Wolf Optimizer using fusion cell-like P systems to improve global search capabilities.
• To address the limitations of traditional GWO, specifically its tendency to get trapped in local optima.
• To evaluate the performance of the proposed algorithm on various optimization problems.

Main Methods:

• Integration of fusion cell-like P systems with the Grey Wolf Optimizer.
• Design of novel convergence factors and dynamic weights to balance exploration and exploitation.
• Parallel computation and inter-cell communication mechanisms within the P system framework.
• Testing on 24 benchmark functions, support vector machine parameter optimization, and constrained engineering design problems.

Main Results:

• The proposed Grey Wolf Optimizer for fusion cell-like P systems demonstrates superior performance over traditional algorithms.
• Achieved higher accuracy, faster convergence speed, and improved stability on test functions.
• Successfully optimized Support Vector Machine parameters across six benchmark datasets.
• Showed competitive results on three real-world constrained engineering design problems.

Conclusions:

• The fusion cell-like P systems effectively enhance the Grey Wolf Optimizer's ability to escape local optima.
• The modified algorithm exhibits improved population search capabilities, faster convergence, and greater stability.
• The proposed approach offers a promising direction for developing more robust and efficient optimization techniques.