In order to enhance the convergence rate of differential evolution algorithm (DE),an improved differential evolution algorithm (IDE) is proposed which introduced the social learning part of particle swarm optimization algorithm into itself. Firstly, the new algorithm improves the diversity of population and global searching ability by small probability random mutation operator. Then, the variation vector and the result which learns from the best individual in population are crossed. The evolutionary process is guided by best individual so that the convergence rate and the optimization precision of DE are improved.The simulation results show that the improved algorithm has better optimization performance.

swarm intelligence; differential evolution algorithm; particle swarm optimization; random mutation; learning factor; diversity