A Novel Evolutionary Strategy for Particle Swarm Optimization

A novel evolutionary strategy for Particle swarm optimization (PSO) to enhance the convergence speed and avoid the local optima is presented. The positive experience and negative lesson from the individual particle's cognition and the swarm's social knowledge are used to accumulate the system's intelligence and guide the swarm's evolution behaviors. The new generation of swarms (named as Child Swarm) and the adjacent former swarms (named as Parent Swarm) are mixed to select the survival of the fittest. The eliminated particles are replaced by the random particles from the outside surroundings. Darwinian evolution method contributes to the convergence and the durative interactions between the swarms and the surroundings who contribute to the global search. This new method can converges faster, gives more robust and precise result and can prevent prematurity more effectively. The corresponding simulation results are presented.