改进的模拟退火算法用于团簇结构优化

Simulated Annealing Algorithm Based on Seed Strategy for Cluster Structure Optimization

为了解决团簇结构的优化问题,提出一种改进的模拟退火算法,称为基于种子策略的模拟退火算法(SS-SA),该算法在基本模拟退火算法的基础上,增加种子策略,提高算法性能。经实验论证,该基于种子策略的模拟退火算法对原子数目为3到10的团簇进行结构优化时,能求得其稳定结构,对原子数目为11到14的团簇进行结构优化时,能求得其近似最优结构。

In order to solve the optimization problem of cluster structure, proposes an improved simulated annealing algorithm, which is called Seed Strategy-Simulated Annealing （SS-SA）. The algorithm is based on the basic simulated annealing algorithm, increases seed strategy and im- proves algorithm performance. Experiments show that this seed-strategy-based simulated annealing algorithm can obtain a stable structure when the clusters with atomic numbers of 3 to 10 are optimized. When the structure of the cluster with 11 to 14 atoms is optimized, the ap- proximate optimal structure can be obtained.