A Global Optimisation Toolbox for Massively Parallel Engineering Optimisation - Computer Science > Distributed, Parallel, and Cluster ComputingReport as inadecuate




A Global Optimisation Toolbox for Massively Parallel Engineering Optimisation - Computer Science > Distributed, Parallel, and Cluster Computing - Download this document for free, or read online. Document in PDF available to download.

Abstract: A software platform for global optimisation, called PaGMO, has been developedwithin the Advanced Concepts Team ACT at the European Space Agency, and wasrecently released as an open-source project. PaGMO is built to tacklehigh-dimensional global optimisation problems, and it has been successfullyused to find solutions to real-life engineering problems among which thepreliminary design of interplanetary spacecraft trajectories - both chemicalincluding multiple flybys and deep-space maneuvers and low-thrust limited,at the moment, to single phase trajectories, the inverse design ofnano-structured radiators and the design of non-reactive controllers forplanetary rovers. Featuring an arsenal of global and local optimisationalgorithms including genetic algorithms, differential evolution, simulatedannealing, particle swarm optimisation, compass search, improved harmonysearch, and various interfaces to libraries for local optimisation such asSNOPT, IPOPT, GSL and NLopt, PaGMO is at its core a C++ library which employsan object-oriented architecture providing a clean and easily-extensibleoptimisation framework. Adoption of multi-threaded programming ensures theefficient exploitation of modern multi-core architectures and allows for astraightforward implementation of the island model paradigm, in which multiplepopulations of candidate solutions asynchronously exchange information in orderto speed-up and improve the optimisation process. In addition to the C++interface, PaGMO-s capabilities are exposed to the high-level language Python,so that it is possible to easily use PaGMO in an interactive session and takeadvantage of the numerous scientific Python libraries available.



Author: Francesco Biscani, Dario Izzo, Chit Hong Yam

Source: https://arxiv.org/







Related documents