A New Particle Swarm Optimization Algorithm for Solve Problems in Continuous Scope

A New Particle Swarm Optimization Algorithm for Solve Problems in Continuous Scope

Mohammad Souri Laki1 Hamid Reza Sahebi2

1) Department of Computer Engineering, Islamic Azad University Ashtyan, Iran
2) Department of Computer Engineering, Islamic Azad University Ashtyan, Iran

IOI Code : XYHD-BFZEB
Publication : International Conference On Research Science And Technology(restconf.org)
Abstract :
    Particle Swarm Optimization (PSO) has been shown as an effective methodology for solving optimization problems over continuous scope. This algorithm has been one of the most popular and used by researchers in real engineering design and economic problems. In this paper, the propose is proposing a new version of PSO algorithm using improvement of the best solution in each iteration in addition to generation of multipoint in the neighborhood and selection of the best solution of all for next step. The goal of using this technique is increment of the exploration and exploitation, in other words, creation of a tradeoff between them. For testing this algorithm, we used the important multiple and complex Benchmark function. The results were compared mathematically with the proposed function algorithm via standard algorithm. It was shown that this technique could result in an improved exploration, local optima avoidance, exploitation and convergence.

Keywords : global optimization heuristics particle swarm optimization simulated annealing