Strategies for increasing the efficiency of a genetic algorithm for the structural optimization of nanoalloy clusters

Abstract

An improved genetic algorithm (GA) is described that has been developed to increase the efficiency of finding the global minimum energy isomers for nanoalloy clusters. The GA is optimized for the example Pt12Pd12, with specific investigation of: the effect of biasing the initial population by seeding; the effect of removing specified clusters from the population ("predation"); and the effect of varying the type of mutation operator applied. These changes are found to significantly enhance the efficiency of the GA, which is subsequently demonstrated by the application of the best strategy to a new cluster, namely Pt19Pd19.

Optimization of bimetallic Cu–Au and Ag–Au clusters by using a modified adaptive immune optimization algorithm

“…From the structures of the 55-atom Ag-Au or Cu-Au clusters, it can be found that Au 55 is a fcc-like structure and Cu 55 and Ag 55 are icosahedra, but it is surprising that a gradual transition from fcc-like structure to icosahedron is not observed when the number of Cu or Ag atoms increases in the clusters. Even one Cu or Ag atom is contained in the cluster, the overall structure changes.…”

“…This can be accounted for by the fact that too much mutation will spoil the population in GA-like algorithms. 55 It may be worth mentioning that atom exchange is an efficient approach for the homotop problem, but this operation has no effect on the optimization of the cluster configuration. A combination of the atom exchange with high efficient optimization methods may be a good choice for the optimization of bimetallic clusters.…”

“…Furthermore, the structures of Cu 1 Au 54 and Cu 12 Au 43 clusters are found to be in agreement with that obtained by canonical Monte Carlo simulations at 100 K, 4 but the structures of others cannot be found in literatures. At first, because different parameters (as listed in Table 1) are used in calculation of the potential listed in Table 5, the potential of the optimized structure of Au 55 is different from that in Table 4, but it is still a fcc-like structure as shown in Figure 2a. However, compared with the amorphous structure of Figure 4h, the central atom is Cu, 11 Cu atoms and a Au atom construct the inner-shell of an icosahedron, and the other 10 Au atoms are evenly distributed in the outer-shell, having a fivefold symmetry.…”

First principles global optimization of metal clusters and nanoalloys

“…Thus, a gradual improvement of the trial solutions will be achieved by applying GAs to GO problems. There are additional strategies which have been introduced to improve the performance of the GA. One strategy which can help to accelerate the GM search is biasing the GA by introducing a seeded initial population [105]. Another strategy is to ensure a certain degree of diversity in each population in order to avoid stagnation (that is to avoid being trapped in a sub-optimal local minima).…”

“…The results are usually a set of numbers or arrays of dissimilarity measures indicating how the individuals differ from each other. Thereafter a 'predator operator' removes certain members of the population, those which are too similar or identical according to the chosen descriptors [105][106][107][108][109][110][111][112]. In the case of clusters, commonly used descriptors for dissimilarity measures are: energies, moments of inertia, radial distribution functions, connection tables, variance of atomic distances, etc.…”