Research And Application On The Absolute Quasi-Physical Algorithms Based On The Density Diffusing Strategy

Nowadays, research on stable structures of atomic clusters is one of the frontier topics of Physics and Chemistry. Properties of clusters depend on the ground-state structures which are unavailable to achieve by experiment methods. In order to gain the ground-state structure of a cluster, scholars, according to the experiments, have put forward a series of potential energy models that possesses different applicable scopes to describe the interaction between atoms in cluster, and have been using modern computer technologies to simulate and study structures of atomic clusters by means of theory methods. Nowadays, there are empirical potential, self-consistent potential based on first principles density functional theory and tight-binding potential to describe the interaction between atoms. Aiming at these primary potential models, scholars designed various methods to simulate and study the ground-state of atomic cluster. However, no matter which kind of method is applied, a NP-Hard question is the only problem to solve finally. Nowadays, the most frequently used algorithms consist of montecarlo method, simulated annealing algorithm, genetic algorithm, molecular dynamics simulation, etc.Because of not only a potential application prospect in catalyzer, medicine and colloid chemistry, but also being expected to play an important role in formatting new nanometer material and devices, research on Au clusters has very important significance. By chosing the representative Lennard-Jones potential of empirical potentials to describe the interaction between atoms of Au cluster, the quasi-physical and quasi-sociological algorithm is able to simulate and research its ground-state structure. Several kinds of quasi-sociological off-trap strategies is improved with the intelligence from the human society . And a kind of bran-new density diffusing quasi-physical off-trap strategies is designed by the inspiration from the gas self-diffusion phenomenon in nature. Finally, a global optimization algorithm for predicting the ground-state structure of Au cluster, that is, a absolute quasi-physical algorithm, is obtained. The algorithm has been verified a great improvement not only on the pace but also on the precision, and to have came up to the advanced international level by means of computing test and compared with the Benchmark, i.e.