The purpose of this project is to develop an algorithm that speeds up large scale simulations of many-body systems. A numerical method is implemented that simulates non-equilibrium phenomena on a mesoscopic time scale. A system is perturbed by an external force, and time averages of variables renormalized in space are calculated numerically, using results of linear response theory, as the system relaxes to equilibrium. The coarse-grained variables evolve slowly in time, allowing one to advance them on a mesoscopic time scale. / The algorithm was tested on two physical systems: a lattice confined ferromagnetic Ising model and an off-lattice Argon-like molecular system. The method simulated accurately the non-equilibrium phenomena studied. It was found that the algorithm is most efficient when it is applied to a process occurring on at least two time scales. This allows one to integrate out the fast, microscopic time scale in order to study long-time, macroscopic behaviour. Through the study of diffusion in a molecular system, it was concluded that the proposed method is computationally faster than solving the microscopic equations of motion and more accurate than solving the macroscopic equations.
Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.80902 |
Date | January 2003 |
Creators | Zhabinskaya, Dina. |
Contributors | Grant, Martin (advisor) |
Publisher | McGill University |
Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
Language | English |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Format | application/pdf |
Coverage | Master of Science (Department of Physics.) |
Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
Relation | alephsysno: 002095180, proquestno: AAIMQ98767, Theses scanned by UMI/ProQuest. |
Page generated in 0.0018 seconds