A New Hybrid Particle Swarm Optimization-Teaching-Learning-Based Optimization for Solving Optimization Problems.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristic Computing

Background:

• Optimization problems are ubiquitous in science and engineering.
• Existing metaheuristic algorithms often struggle with balancing exploration and exploitation.
• Particle Swarm Optimization (PSO) excels at local search (exploitation), while Teaching-Learning-Based Optimization (TLBO) is strong in global search (exploration).

Purpose of the Study:

• To develop a novel hybrid metaheuristic algorithm, hPSO-TLBO, by integrating PSO and TLBO.
• To enhance the search capabilities by combining PSO's exploitation with TLBO's exploration.
• To evaluate the effectiveness of hPSO-TLBO on benchmark functions and real-world engineering problems.

Main Methods:

• A hybrid algorithm, hPSO-TLBO, was designed by combining the teacher phase of TLBO with PSO's speed equation.
• The learning phase of TLBO was enhanced by enabling students to learn from superior peers.
• The algorithm was mathematically modeled and rigorously tested on unimodal, multimodal, and high-dimensional benchmark functions, including CEC 2017 problems.

Main Results:

• hPSO-TLBO demonstrated remarkable performance across diverse benchmark functions.
• The algorithm effectively balanced exploration and exploitation of the search space.
• Comparative analysis showed hPSO-TLBO consistently outperformed twelve other metaheuristic algorithms.

Conclusions:

• The proposed hPSO-TLBO algorithm offers a powerful and effective approach to solving complex optimization problems.
• Its superior performance highlights its potential for addressing challenging real-world engineering applications.
• The hybrid strategy successfully leverages the strengths of both PSO and TLBO.