A Hybrid Algorithm Of Adaptive Particle Swarm Optimization Based On Adaptive Moment Estimation Method

Particle swarm optimization (PSO) algorithm is a promising swarm intelligence optimization technology. It has been applied to a variety of complex optimization problems due to its outstanding global search ability. However, it suffers from premature convergence and slow convergence rate. Motivated by adaptive moment estimation (Adam) method, which is computationally efficient, little memory-required and also appropriate for non-stationary objectives, a hybrid algorithm combining adaptive PSO with a modified Adam method (AdamPSO) is proposed in this paper. Adaptive particle swarm optimization (APSO) is first used to perform stochastic and rough search. In the solution space obtained by APSO, Adam method is then used to perform further search, which may establish a new solution space. Depending on the fitness value of particles, the position of each particle switches alternately between APSO and Adam. The experimental results on six well-known benchmark functions show that our proposed algorithm gets better convergence performance compared to other five classical PSOs.