Social Emotional Optimization Algorithm In The Cluster Structure Optimization

Cluster structural optimization problem is a NP problem, The main difficulty lies in the number of local minima increases with the atomic number of exponential growth, and the local minimum and global minimum is very close to make the algorithm vulnerable local optimum. LJ(Lennard-Jones) issues and structural problems of Ag clusters and clusters are two typical optimization problem, which is widely used by LJ. LJ function can be used not only to optimize the structure of inert gas and description of protein folding, and can be used to study carbon nanotubes in the C atom. The interaction is existed between Cs atoms and carbon-carbon and founded between the tube and spherical fullerene C-C interaction. LJ effectively solve structural problems in molecular biology and material science developments in the field has great significance. Ag clusters in catalysis, electronic materials and alloys have important applications.Social Emotion Optimization Algorithm(SEOA) is proposed a new group intelligent optimization algorithm. In algorithm, each individual who represents a virtual person improve their social status through cooperation, competition and so on. In each evolution generation (time slice), they choose behavior that can improve their social status, according to the corresponding sentiment index, their experience and social experience. And selected results feedback behavioral correctness through the degree of social recognition. If the behavior is accepted, their sentiment index rises, otherwise, down. SEOA which optimizes cluster structure and optimization results are LJ clusters structure of the atomic number less than 17 and Ag clusters structure of the atomic number less than 9.Seed technology of the structural optimization is very effective, and the use of seed technology allows the algorithm to optimize the search when there is guidance to improve search efficiency. The disadvantage is difficulty in seed selection, to select a configuration similar to the cluster structure as the seed, If the selected seed and configurations are optimized clusters or outer structure of different atoms vary widely distributed, it will make the algorithm into a local optimum. In order to improve the optimization efficiency, seed technology is introduced in this paper, reducing the optimization of LJ clusters difficult to optimize the number of 310, which have been optimized the structure of Ag43.Lattice Searching technique is to optimize the atoms on its possible positions, and may select an optimal location to optimize the location through the algorithm. This paper introduces the discrete optimization of social emotions and the introduction of a lattice Searching technique, which reduces the scope of the search and turns the search space into several discrete points, so the optimization effect is significantly better than the previous two optimization methods. To simplify the algorithm and get the optimization algorithm shorten the time depends on the growth characteristics of cluster structure cluster atoms, which grow from the inside out, filled only after the inner atomic arrangement of atoms in the outer layer, and optimized clusters in this article. Only the outermost layer of the structure optimization, the default is the full atomic structure of the inner layer shell. This will not only improve the success rate of the algorithm, but also shorten the optimization time. And optimized LJ582.