并行遗传算法用于氩原子簇的结构优化

The Geometry Optimization of Argon Atom Clusters Using a Parallel Genetic Algorithm

本文提出了一种并行的遗传算法(称为IS-GA)并应用于氩原子簇的结构优化,该算法结合了智能搜索(IS)策略,并应用扩展分布式并行模型(EDGA)将该遗传算法并行化。对原子数为N=2…30的氩原子簇(Ar)k的优化结果表明,IS-GA具有较强的优化能力和效率,该算法能得到原子个数N<27的氩原子簇的最优结构,对于原子个数大于27小于30的氩原子簇可得到其近似最优结构。

A parallel genetic algorithm with intelligent search strategy (IS- GA) is proposed, and it is programmed in a parallel mode of extended distributed genetic algorithm (EDGA) . The intelligent search strategy can detect and avoid the local optimization and makes the method to be easy to find the final exact optimization. This algorithm also has the merits of parallel genetic algorithm in high efficiency. Applying the method to optimization of argon atom clusters, it was shown that the exact global optimal configuration of argon clusters with atom number N < 27 can be found in a reasonable computing time, and approximate optimization can be obtained for clusters with N = 27 - 30 .