A Novel Consensus-Based Particle Swarm Optimization-Assisted Trust-Tech Methodology for Large-Scale Global...

Area of Science:

• Computational mathematics
• Optimization algorithms
• Artificial intelligence

Background:

• Nonlinear optimization problems present significant computational challenges.
• Existing particle swarm optimization (PSO) algorithms have limitations in finding global optima.
• Developing robust and scalable optimization methodologies is crucial for scientific and engineering applications.

Purpose of the Study:

• To introduce a novel three-stage methodology for solving nonlinear optimization problems.
• To integrate Trust-Tech methods with consensus-based PSO for enhanced global optimization.
• To evaluate the performance and scalability of the proposed methodology.

Main Methods:

• A three-stage approach combining Trust-Tech methods, consensus-based PSO, and local optimization.
• Integration of diverse optimization techniques to generate high-quality local optima potentially containing the global optimum.
• Comparative analysis against recent PSO algorithms using benchmark and large-dimension test functions (CEC 2010).

Main Results:

• The proposed methodology rapidly obtains high-quality optimal solutions.
• Experimental results show favorable comparisons with existing PSO algorithms.
• Demonstrated effectiveness on both small-dimension and large-dimension optimization problems.
• The methodology exhibits promising scalability.

Conclusions:

• The consensus-based PSO-assisted Trust-Tech methodology is effective for finding global optimal solutions.
• The integrated approach provides a robust and efficient solution for nonlinear optimization.
• The methodology's performance and scalability suggest broad applicability.