A hybridizing-enhanced differential evolution for optimization.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Differential Evolution (DE) is a widely used optimization algorithm.
• A key limitation of DE is its slow convergence rate.
• Existing DE variants often struggle with premature convergence and local optima.

Purpose of the Study:

• To enhance the convergence rate and solution accuracy of the Differential Evolution algorithm.
• To introduce a novel hybrid optimization algorithm by integrating the Gray Wolf Optimizer (GWO) with DE.
• To evaluate the performance of the proposed Hunting Differential Evolution (HDE) algorithm against established benchmarks.

Main Methods:

• A new hybrid algorithm, Hunting Differential Evolution (HDE), is proposed.
• HDE combines the search capabilities of DE with the convergence acceleration of GWO.
• Parameter tuning of crossover rate and mutation probability allows for balancing DE and GWO strengths.

Main Results:

• The HDE algorithm demonstrated superior performance on CEC-2019 and CEC-2014 benchmark functions.
• Specific variants like HDE/current-to-rand/1 and HDE/current-to-best/1 showed significant improvements over other HDE variants and existing algorithms.
• jHDE variant showed notable improvements over jDE on 17 tested functions.

Conclusions:

• The proposed HDE algorithm effectively accelerates convergence and improves optimal solution finding.
• HDE overcomes the limitations of the standard DE algorithm, including avoiding local minima.
• HDE offers a robust and efficient approach for complex optimization tasks.