An Improved Grey Wolf Optimizer Based on Differential Evolution and Elimination Mechanism

Area of Science:

• Computational Intelligence
• Swarm Intelligence
• Optimization Algorithms

Background:

• The grey wolf optimizer (GWO) is a metaheuristic optimization algorithm inspired by wolf pack hierarchy.
• Existing GWO algorithms may face challenges in balancing exploration and exploitation, potentially leading to premature convergence or suboptimal solutions.

Purpose of the Study:

• To propose an improved grey wolf optimizer (IGWO) that enhances convergence speed and optimization accuracy.
• To integrate biological evolution and the "survival of the fittest" (SOF) principle into the GWO framework.
• To address the limitations of the basic GWO by preventing local optima through an evolution and elimination mechanism.

Main Methods:

• The proposed IGWO algorithm incorporates differential evolution (DE) as the evolutionary pattern for wolves.
• A "survival of the fittest" (SOF) mechanism is implemented, where the worst-performing wolves are eliminated and replaced with new ones in each iteration.
• The performance of IGWO was evaluated using 12 typical benchmark functions and compared against GWO variants (DGWO, SGWO), DE, particle swarm optimization (PSO), artificial bee colony (ABC), and cuckoo search (CS) algorithms.

Main Results:

• Simulation experiments demonstrated that IGWO achieved superior convergence velocity compared to the other algorithms tested.
• IGWO exhibited enhanced optimization accuracy, outperforming GWO, DE, PSO, ABC, and CS on the benchmark functions.
• The integration of DE and the SOF principle effectively improved the exploration-exploitation balance and prevented the algorithm from getting trapped in local optima.

Conclusions:

• The IGWO algorithm offers a significant improvement over the standard GWO and other contemporary optimization algorithms.
• The proposed evolution and elimination mechanism is effective in accelerating convergence and increasing optimization accuracy.
• IGWO provides a robust and efficient approach for solving complex optimization problems.