Konferensartikel

Parallel Simulation of PDE-based Modelica Models using ParModelica

Gustaf Thorslund
ThorslundTech AB, Sweden

Mahder Gebremedhin
Department of Computer and Information Science, Linköping University, Sweden

Peter Fritzson
Department of Computer and Information Science, Linköping University, Sweden

Adrian Pop
Department of Computer and Information Science, Linköping University, Sweden

Ladda ner artikelhttp://dx.doi.org/10.3384/ecp17142660

Ingår i: Proceedings of The 9th EUROSIM Congress on Modelling and Simulation, EUROSIM 2016, The 57th SIMS Conference on Simulation and Modelling SIMS 2016

Linköping Electronic Conference Proceedings 142:96, s. 660-666

Visa mer +

Publicerad: 2018-12-19

ISBN: 978-91-7685-399-3

ISSN: 1650-3686 (tryckt), 1650-3740 (online)

Abstract

The Modelica language is a modelling and programming language for modelling cyber-physical systems using equations and algorithms. In this thesis two suggested extensions of the Modelica language are covered. Those are Partial Differential Equations (PDE) and explicit parallelism in algorithmic code. While PDEs are not yet supported by the Modelica language, this article presents a framework for solving PDEs using the algorithmic part of the Modelica language, including parallel extensions. Different numerical solvers have been implemented using the explicit parallel constructs suggested for Modelica by the ParModelica language extensions, and implemented as part of OpenModelica. The solvers have been evaluated using different models, and it can be seen how bigger models are suitable for a parallel solver. The intention has been to write a framework suitable for modelling and parallel simulation of PDEs. This work can, however, also be seen as a case study of how to write a custom solver using parallel algorithmic Modelica and how to evaluate the performance of a parallel solver.

Nyckelord

OpenModelica, ParModelica, PDE, parallel Computing, GPU, GPGPU

Referenser

Inga referenser tillgängliga

Citeringar i Crossref