Two Sub-swarms Particle Swarm Optimization Algorithm

• Published in: Advances in Natural Computation pp. 515 - 524
• Main Authors: Chen, Guochu, Yu, Jinshou
• Format: Book Chapter Conference Proceeding
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2005; Springer
• Series: Lecture Notes in Computer Science
• Subjects: Applied sciences; Artificial intelligence; Computer science; control theory; systems; Exact sciences and technology; Swarm intelligence; Neural network; Modeling; Evolutionary algorithm; Mathematical programming
• ISBN: 9783540283201; 354028320X; 3540283234; 9783540283232
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/11539902_63

This paper proposes a two sub-warms particle swarm optimization algorithm (TSPSO) and its iteration equations. The new algorithm assumes that particles are divided into two sub-swarms. The two sub-swarms have different move directions. One sub-swarm moves toward the global best position. Another moves in the opposite direction. Not only its own move experience and the best individual’s position of its own sub-swarm, but also the global best position of the whole swarm can affect each particle’s move in every iteration. If the fitness of the global best position can’t be improved for fifteen successive steps, the particles of the two sub-swarms are exchanged. At the same time, the worst individual of one sub-swarm is replaced with the best individual of another. Then, both TSPSO and PSO are used to resolve ten well-known and widely used test functions’ optimization problems. Results show that TSPSO has greater optimization efficiency, better optimization performance and more advantages in many aspects than PSO.