Hybrid particle swarm optimization with wavelet mutation and its industrial applications

Area of Science:

• Computational intelligence
• Optimization algorithms
• Applied mathematics

Background:

• Particle Swarm Optimization (PSO) is a widely used metaheuristic algorithm.
• Enhancing PSO's exploration capability remains a key research challenge.
• Wavelet theory offers potential for improving search dynamics.

Purpose of the Study:

• To introduce a hybrid PSO incorporating wavelet theory for improved optimization.
• To evaluate the enhanced PSO's effectiveness in exploring solution spaces.
• To assess the applicability of the proposed method in industrial engineering problems.

Main Methods:

• Developed a hybrid Particle Swarm Optimization (PSO) algorithm.
• Integrated a wavelet-theory-based mutation operator into the PSO framework.
• Tested the method on benchmark functions and industrial applications: load flow, fluid dispensing, and neural network controller design.

Main Results:

• The proposed hybrid PSO significantly improved convergence speed.
• Enhanced solution quality and stability were observed compared to existing methods.
• Demonstrated strong performance in complex industrial problem-solving.

Conclusions:

• The wavelet-theory-based mutation effectively enhances PSO's exploration capabilities.
• The hybrid PSO offers a robust and efficient optimization approach.
• The method shows significant potential for real-world engineering applications.