Development of efficient particle swarm optimizers by using concepts from evolutionary algorithms

Particle swarm optimization (PSO) has been in practice for more than 10 years now and has gained wide popularity in various optimization tasks. In the context to single objective optimization, this paper studies two aspects of PSO: (i) First, its ability to approach an "optimal basin", and (ii) To find the optimum with high precision, once it enters the region of interest. We test standard PSO algorithms and discover their inability in handling both aspects efficiently. To address these issues in PSO, we propose an EA which is algorithmically similar to PSO, and then borrow different EA-specific operators to enhance the PSO's performance. Our final proposed PSO contains a parent-centric recombination operator instead of usual particle update rule and has a performance comparable to a well-known GA (and outperforms the GA in some occasions). Thus, this study emphasizes that efforts spend in establishing equivalences between different optimization algorithms, such as vari...

• Type: Conference
• Year: 2010
• Where: GECCO
• Authors: Kalyanmoy Deb, Nikhil Padhye
• Keywords: Final Proposed Pso, GECCO 2010, Optimization, Particle Swarm Optimization, Standard Pso Algorithms