The aim of the booklet is to explain why validation and calibration of CFD for particular applications are important aspects of CFD modelling, and to recommend procedures to be followed to minimise and quantify uncertainty when CFD is used. The choice of parameters or variables to be used in validation and calibration studies should be discussed, and the design of experiments and tests against which to assess CFD should be included. Methods of making the best use of existing or sparse data should be considered, and the role of sensitivity studies in evaluating the quality of a CFD solution when test data are unavailable should be described. The extrapolation of validation from simple to complex flow problems, and the advantages and pitfalls of using published measurements as a source of validation data could also be covered.
The target readership is new and improving CFD users working in industry.
Booklet: How to make a CFD project fit for purpose (working title)
CFD is used for increasingly complex industrial applications. In spite of the improvement in computing resources and CFD software functionality, almost all calculations of real-world flow problems must be idealised to some extent, for
- in the simplification or omission of geometrical details, and the exploitation of symmetry or periodicity
- in the use of mathematical sub-models to describe particular aspects or features of the flow or fluid properties
- in the prescription of appropriate boundary conditions
- in the treatment of unsteady and turbulent behaviour of the fluid
The aim of the booklet is to examine the issues that must be considered when making decisions about how to idealise or simplify a flow problem for CFD, and to show how to maximise the usefulness of results with optimum use of analysis resources. Particular emphasis should be given to the relationship between the purpose of the calculation and the way the model is set up. The document is expected to cover geometry, mesh, mathematical models, boundary conditions, use of sensitivity studies, and any other relevant factors. Details of comparative examples should be included.
The target readership is new and improving CFD users facing a range of applications and model complexity.