Coupling CAD and CFD codes within a virtual integration platform

Gatchell, S. and Whitfield, R.I. and Duffy, A.H.B. and Abt, C. (2008) Coupling CAD and CFD codes within a virtual integration platform. In: Marine CFD 2008: International Conference on Marine Computational Dynamics, 2008-03-26 - 2008-03-27.

[thumbnail of strathprints013462]
Preview
Text. Filename: strathprints013462.pdf
Accepted Author Manuscript

Download (1MB)| Preview

Abstract

The Virtual Integration Platform (VIP) is an essential component of the VIRTUE project. It provides a system for combining disparate numerical analysis methods into a simulation environment. The platform allows for defining process chains, allocating of which tools to be used, and assigning users to perform the individual tasks. The platform also manages the data that are imported into or generated within a process, so that a version history of input and output can be evaluated. Within the VIP, a re-usable template for a given process chain can be created. A process chain is composed of one or more smaller tasks. For each of these tasks, a selection of available tools can be allocated. The advanced scripting methods in the VIP use wrappers for managing the individual tools. A wrapper allows communication between the platform and the tool, and passes input and output data as necessary, in most cases without modifying the tool in any way. In this way, third-party tools may also be used without the need for access to source code or special modifications. The included case study demonstrates several advantages of using the integration platform. A parametric propeller design process couples CAD and CFD codes to adapt the propeller to given operating constraints. The VIP template helped eliminate common user errors, and captured enough expert knowledge so that the casual user could perform the given tasks with minimal guidance. Areas of improvements to in-house codes and to the overall process were identified while using the integration platform. Additionally, the process chain was designed to facilitate formal optimisation methods.