A Multi-Strategy Adaptive Comprehensive Learning PSO Algorithm and Its Application.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• Particle Swarm Optimization (PSO) is a widely used metaheuristic algorithm.
• Existing PSO variants often struggle with premature convergence and limited search diversity.
• Optimizing photovoltaic (PV) systems requires robust parameter estimation for efficiency.

Purpose of the Study:

• To propose a novel Multi-Strategy Adaptive Comprehensive Learning Particle Swarm Optimization (MSACL-PSO) algorithm.
• To enhance the global and local search abilities of PSO.
• To improve the parameter optimization of photovoltaic systems.

Main Methods:

• Introducing comprehensive learning, multi-population parallel strategies, and parameter adaptation into PSO.
• Implementing population particle exchange and mutation for information sharing.
• Designing new velocity update, learning factor adjustment, and adaptive inertia weight strategies.
• Utilizing benchmark functions and photovoltaic parameter optimization for evaluation.

Main Results:

• The proposed MSACL-PSO algorithm demonstrated superior performance on 6 out of 10 benchmark functions.
• Significant improvements in population diversity, solution accuracy, and overall search ability were observed.
• The algorithm outperformed several PSO variants and other competing algorithms.

Conclusions:

• The MSACL-PSO algorithm effectively addresses limitations of traditional PSO, offering enhanced exploration and exploitation.
• It provides a more effective parameter combination for photovoltaic system optimization, leading to improved energy conversion efficiency.
• This adaptive approach offers a promising direction for complex engineering optimization problems.