An electronic transition-based bare bones particle swarm optimization algorithm for high dimensional optimization...

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• High-dimensional single-objective optimization problems present significant computational challenges.
• Existing particle swarm optimization (PSO) variants often struggle with premature convergence and escaping local optima in complex search spaces.

Purpose of the Study:

• To introduce a novel particle swarm optimization algorithm, the electronic transition-based bare bones particle swarm optimization (ETBBPSO).
• To enhance the precision and global search capabilities of PSO for high-dimensional single-objective optimization problems.

Main Methods:

• The proposed ETBBPSO algorithm divides particles into distinct orbits.
• Introduces a transition operator to improve global search and facilitate escape from local minimums.
• Incorporates an orbit merge operator to consolidate search efforts from less effective orbits into more promising ones.

Main Results:

• ETBBPSO demonstrated high precision in solving high-dimensional single-objective optimization problems.
• Extensive experiments were conducted using benchmark datasets CEC2014 and CEC2020.
• Performance was compared against four well-established population-based algorithms.

Conclusions:

• The ETBBPSO algorithm effectively addresses the challenges of high-dimensional single-objective optimization.
• The novel transition and orbit merge operators significantly enhance the algorithm's ability to find precise solutions.
• ETBBPSO shows superior performance compared to existing algorithms in benchmark tests.