A Particle Swarm Optimizer with randomized quasi-random and general recognition

In order to improve the convergent speed and raise the accurate level of solutions further, in this study, we present a novel particle swarm optimizer, called Particle Swarm Optimizer with randomized quasi-random initialization and general recognition. The proposed algorithm uses Quasi-random sequence to initialize the population for a more uniform population distribution. Cauchy distribution and general recognition are employed to enrich the diversity of particles in runs. The experimental results show that the accurate level of the optima and the convergent speed both are outperformed than the algorithms initialize with a pseudo-random sequence.