A twinning bare bones particle swarm optimization algorithm.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• Particle Swarm Optimization (PSO) is a widely used metaheuristic for solving complex optimization problems.
• Traditional PSO variants often require careful parameter tuning, limiting their applicability.
• Escaping local minima remains a significant challenge in many optimization tasks.

Purpose of the Study:

• To introduce a novel, parameter-free optimization algorithm: Twinning Bare Bones Particle Swarm Optimization (TBBPSO).
• To enhance the local minimum escaping ability and overall search performance of PSO.
• To demonstrate the algorithm's effectiveness across a range of benchmark optimization problems.

Main Methods:

• The proposed TBBPSO algorithm integrates two novel operators: the Twins Grouping Operator (TGO) and the Merger Operator (MO).
• The TGO reorganizes the particle swarm into 'twins,' with one particle focusing on global search and the other on local search.
• The MO merges these twins to improve the swarm's collective search capability.

Main Results:

• TBBPSO demonstrated a superior ability to escape local minima compared to standard optimization methods.
• The algorithm achieved high precision results on the CEC2014 benchmark functions.
• TBBPSO exhibited robust performance across various types of optimization problems without requiring parameter adjustments.

Conclusions:

• The TBBPSO algorithm offers an effective and robust approach to complex optimization problems.
• Its parameter-free nature and integrated search strategies make it a versatile tool for diverse applications.
• The proposed TGO and MO significantly enhance swarm intelligence optimization capabilities.