A Stochastic Gradient-Directed Particle Swarm Optimization

Particle swarm optimization algorithms have been widely used for optimization problems. These kinds of algorithms are popular due to their simplicity and good efficiency. In this paper we propose a novel algorithm based on conventional PSO with some modifications which outperforms previous methods and presents more accurate results. Our proposed method uses gradient coefficient for exactness and normal distribution for diversity in exploration. We examined the performance of our method with several well-known mathematical functions