A New Hybrid Algorithm of Differential Evolution Algorithm and Particle Swarm Algorithm

In this paper, we point out some shortcomings of Differential evolution algorithm (DE) with lower search efficiency and advance to the local optimal value easily. Combining with particle swarm algorithm (PSO)’s advantages of convergence rate, we put forward a new hybrid algorithm (DPM) to overcome these shortcomings. Instead of dividing all individuals into two equal size groups, in DPM algorithm, according to the concentration parameter, all individuals dynamic divided into two different size groups after a certain iterations. And then, merger the two groups, a novel mutation has been developed to some individuals to generate the new individuals. Finally, we will apply six famous benchmark functions to test and evaluate the performance of DPM. Compared with DE and PSO, we obtain that DPM has a faster and better convergence rate to the global optimum from the experimental results. Keywords—Differential evolution algorithm, Particle swarm algorithm, Concentration parameter, Novel mutation, Group dynamic division.