Traditional particleswarmoptimizationalgorithm has some disadvantages, such as slow convergence speed and easy to fall into local extremes. In order to improve the performance, an improvedadaptiveparticleswarm o...
详细信息
Traditional particleswarmoptimizationalgorithm has some disadvantages, such as slow convergence speed and easy to fall into local extremes. In order to improve the performance, an improved adaptive particle swarm optimization algorithm with a two-way learning method is proposed. First, the algorithmadaptively adjusts the algorithm according to the iteration periods of the optimization ***, the inertia weight and the value of the learning factor are changed nonlinearly, so as to better balance the search behavior of the particles in the group;Second,the idea of beetle search is introduced into the particleswarmalgorithm to form a new two-waylearning mechanism, which overcomes the limitations of the traditional particleswarmalgorithm and help to increase the diversity of the *** this way, the search scope is expanded, and the search accuracy of the algorithm is enhanced. Finally, the simulation is carried out on several multi-dimensional functions, and compared with other two related algorithms. The experimental results show that under the same experimental conditions, the improvedalgorithm has obvious advantages in optimization ability and convergence speed.
暂无评论