iGen: A program for the automated generation of models and parameterisations

Tang, D.F. and Dobbie, S. (2011) iGen: A program for the automated generation of models and parameterisations. [MIMS Preprint]

Warning
There is a more recent version of this item available.
[thumbnail of iGenMims1.pdf] PDF
iGenMims1.pdf

Download (181kB)

Abstract

Complex physical systems can often be simulated using very high-resolution models but this is not always practical because of computational restrictions. In this case the model must be simplified or parameterised, but this is a notoriously difficult process that often requires the introduction of `model assumptions' that are hard or impossible to justify. Here we introduce a new approach to parameterising models. The approach makes use of a newly developed computer program, which we call iGen, that analyses the source code of a high-resolution model and formally derives a much faster parameterisation that closely approximates the original, reporting bounds on the error introduced by any approximations. These error bounds can be used to formally justify use of the parameterised model in subsequent numerical experiments. Using increasingly complex physical systems as examples we illustrate that iGen has the ability to produce parameterisations that run typically orders of magnitude faster than the underlying, high-resolution models from which they are derived and show that iGen has the potential to become an important tool in model development.

Item Type: MIMS Preprint
Uncontrolled Keywords: Model reduction, parameterization, parameterisation, climate modelling, symbolic analysis, approximation theory
Subjects: MSC 2010, the AMS's Mathematics Subject Classification > 41 Approximations and expansions
MSC 2010, the AMS's Mathematics Subject Classification > 65 Numerical analysis
MSC 2010, the AMS's Mathematics Subject Classification > 68 Computer science
Depositing User: Mr D.F. Tang
Date Deposited: 30 Mar 2011
Last Modified: 08 Nov 2017 18:18
URI: https://eprints.maths.manchester.ac.uk/id/eprint/1595

Available Versions of this Item

Actions (login required)

View Item View Item