A novel particle swarm optimization based on hybrid-learning model.

Area of Science:

• Computational Intelligence
• Optimization Algorithms

Background:

• Particle Swarm Optimization (PSO) performance is significantly influenced by convergence speed and population diversity.
• Balancing exploration and exploitation is crucial for effective optimization.

Purpose of the Study:

• To propose a novel Particle Swarm Optimization based on a Hybrid Learning Model (PSO-HLM).
• To enhance convergence speed and population diversity in PSO.
• To improve the overall performance of optimization algorithms.

Main Methods:

• Implementing a hybrid learning model for particle velocity updates in early iterations.
• Employing a multi-pools fusion strategy for particle mutation in later iterations.
• Conducting experiments on 30 benchmark functions to evaluate performance.

Main Results:

• The proposed PSO-HLM demonstrates improved convergence speed.
• PSO-HLM effectively expands population diversity, preventing local optima.
• Experimental results indicate superior performance compared to state-of-the-art algorithms.

Conclusions:

• PSO-HLM offers a robust approach to balancing exploration and exploitation.
• The hybrid learning model and multi-pools fusion strategy significantly enhance optimization capabilities.
• PSO-HLM represents an advancement in population-based optimization techniques.