Research on Global Optimal Cluster Structure Prediction based on Intelligent Algorithm

The prediction of the ground state structure of atomic clusters is one of the core problems in the study of clusters, and the results will directly affect the discussion on the properties of clusters. For the difficult cluster structure prediction problem, the cluster energy prediction model of nonmetallic clusters is established first, and B is taken as an example to study. Secondly, B24 genmer tool is used to generate isomers of clusters. Then, an improved simulated annealing algorithm is used to obtain the local optimal configuration for cluster optimization. Finally, the global optimal structure of Bn(n=24,53) is predicted and the stability of B53 and B24 is analyzed from three aspects of geometry, bonding and dynamics. The experimental results show that the global optimal structure is a sea urchin structure with a football-shaped interior. Its surface is composed of 8 triangles and 6 squares, and there is an atom at each vertex in the direction of reverse extension from the center. The structure of B53 is hollow and irregular. The improved simulated annealing algorithm used in this paper can provide an effective method for the study of complex cluster structure, and the results obtained have certain reference significance for the experimental study of atomic cluster structure.