Version 3.0 of QPAC, Quintessa’s in-house general-purpose modelling tool, was released in May. QPAC differs from traditional computer codes, in which the mathematical model to be solved is hard-wired into the software, by adopting a flexible ‘model as input’ approach.
This method requires the user to input the equations to be solved along with the model parameter values and spatial discretisation. In this way complex coupled nonlinear models can be developed, where parameters can be time?dependent and variables can depend on each other, at a range of scales from detailed process models to high-level systems models. Having such a flexible tool at our disposal is invaluable to Quintessa’s modellers, allowing us to perform rapid prototyping exercises and investigate conceptual model uncertainty.
A number of modules have been developed for use with QPAC, which encapsulate commonly used physical processes such as multi-phase flow, radionuclide transport through porous media, heat transfer and geochemical processes. These allow QPAC to operate in a mode more akin to a traditional modelling code, where the user simply specifies input parameter values, boundary conditions and the system discretisation. However, where QPAC differs from these traditional codes is that additional user-defined processes can easily be coupled to existing modules.
New features in QPAC 3.0 include:
- QPAC Player, providing a simple user interface to QPAC models that have been developed by modellers at Quintessa. This allows clients and collaborators to reproduce results and perform their own sensitivity analysis, without requiring knowledge of the QPAC input language.
- Enhanced tools for system discretisation, including the use of unstructured grids which allow holes to be punched into Cartesian or polar grids or irregular edges to be specified.
- Weak as well as strong couplings can now be included, by reloading results obtained from existing QPAC simulations.
- Faster run times.