Paper Info
Title
Global optimization of Lennard-Jones clusters by a parallel fast annealing evolutionary algorithm.
Abstract
A parallel fast annealing evolutionary algorithm (PFAEA) was presented and applied to optimize Lennard-Jones (LJ) clusters. All the lowest known minima up to LJ(116) with both icosahedral and nonicosahedral structure, including the truncated octahedron of LJ(38), central fcc tetrahedron of LJ(98), the Marks' decahedron Of LJ(75-77), and LJ(102-104), were located successfully by the unbiased algorithm. PFAEA is a parallel version of fast annealing evolutionary algorithm (FAEA) that combines the aspect of population in genetic algorithm and annealing algorithm with a very fast annealing schedule. A master-slave paradigm is used to parallelize FAEA to improve the efficiency. The performance of PFAEA is studied, and the scaling of execution time with the cluster size is approximately cubic, which is important for larger scale energy minimization systems.
Year
DOI
Venue
2002
10.1021/ci020004i
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES
Keywords
Field
DocType
lennard jones,evolutionary algorithm,global optimization
Population,Combinatorics,Evolutionary algorithm,Global optimization,Algorithm,Maxima and minima,Truncated octahedron,Tetrahedron,Genetic algorithm,Mathematics,Energy minimization
Journal
Volume
Issue
ISSN
42
5
0095-2338
Citations 
PageRank 
References 
4
0.93
3
Authors
3
Name
Order
Citations
PageRank
Wensheng Cai16810.88
Haiyan Jiang2153.13
Xueguang Shao36310.81