Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
DY: Fachverband Dynamik und Statistische Physik
DY 40: Poster - Quantum Systems/ Stat. Phys./ Diffusive Process
DY 40.11: Poster
Donnerstag, 3. April 2014, 17:00–19:00, P3
Efficient Implementation and Application of the Artificial Bee Colony Algorithm to Low-Dimensional Optimization Problems — •Guido Falk von Rudorff1, Christoph Wehmeyer1, and Daniel Sebastiani1,2 — 1Dahlem Center for Complex Quantum Systems, Freie Universität Berlin, Arnimallee 14, 14195 Berlin, Germany — 2Institute of Chemistry, Martin-Luther-Universität Halle-Wittenberg, von-Danckelmann-Platz 4, 06120 Halle, Germany
We adapt a swarm-intelligence-based optimization method (the artificial bee colony algorithm, ABC) for the prediction of global minima on potential energy surfaces of molecular geometries to enhance its parallel scaling properties and to improve the escaping behavior from deep local minima. Specifically, we apply the approach to the geometry optimization of Lennard-Jones clusters. We illustrate the performance and the scaling properties of the parallelization scheme for several system sizes (5-20 particles) and different atomic interaction potentials. Deriving optimal parameters for the algorithm is a highly non-trivial problem. We present a strategy for finding ranges of the parameters of the ABC algorithm which yield maximal performance for Lennard-Jones clusters and Morse clusters. We evaluate small carbon clusters using the Tersoff potential to illustrate which kind of potential energy surfaces can be searched with this algorithm in a timely manner. The suggested parameter ranges turn out to be very similar for these different interaction potentials; thus, we believe that our reported values are fairly general for the ABC algorithm applied to chemical optimization problems.