Hybrid Particle Swarm Algorithm with Levy Mutation

The standard Particle Swarm Optimization (PSO) studies its own previous best solution and the group’s previous best to optimize problems. One problem exists in PSO is its tendency of trapping into local optimum. To avoid this problem, in this paper a new PSO (LHPSO)is presented by combining a Lévy mutation based on Lévy Distribution on the best particle so that the mutated best particle could lead all the rest of particles to the better positions. The idea to introduce Lévy Mutation operator is to increase the probability of a particle escaping from a local optimum. As it is known the variance of Lévy distribution is infinite, so that Lévy mutation could make a particle to have a long jump. LHPSO has been compared with Standard PSO on a set of benchmark functions. The results show that proposed idea is an effective and efficient idea. KeywordsPSO, Lévy Probability Distribution, Mutation, Exploration, Exploitation.