STUDY OF THE LOWEST ENERGY STRUCTURE OF ATOMIC CLUSTERS USING A GENETIC ALGORITHM

An optimization approach using a genetic algorithm in the search for g lobal minima will be described. The method is based on the use of cont rol variables to form the genotypes in each generation. This procedure allows an accurate representation of the control variables leading to a high resolution determination of the optimum solution. A set of gen etic operators, appropriate for the operation on chromosomes represent ed by real numbers, is introduced. As an example, the method is used t o obtain the lowest energy structures of rare gas microclusters, R-n ( n = 4-10). Comparison of these results with published data shows an ex cellent agreement. It is shown that, in the present application, the g enetic algorithm based method converges to the global minima much more rapidly than the simulated annealing approach. The role of the select ion procedure used and the relative importance of the various operator s will be discussed.