Hybrid Differential Evolution and Particle Swarm Optimization Algorithm Based on Random Inertia Weight

摘要

A new hybrid differential evolution and particle swarm optimization algorithm called RWDEPSO is proposed, which combines the advantages of particle swarm optimization (PSO) with fast convergence speed and differential evolution (DE) with high search accuracy. In the new algorithm, the random inertia weight is introduced to strengthen the global exploration ability and local exploition ability of the PSO optimization process. Then, the optimized individuals of PSO and DE are cross-operated to generate new individuals, which inherit the dominant characteristics of both algorithms. Comparing with the simulations of the other intelligent algorithms in six typical Benchmark functions, the results show that the proposed algorithm RWDEPSO has faster convergence speed and stronger global research ability.