| Popis: |
Fluid mechanics underpins many aspects of food engineering. The capacity to model and simulate these has grown primarily via developments in codes and numerical methods in other fields. Newtonian flows, including free surfaces, are now handled routinely. The complexity and diversity of foods - particularly multiphase materials - present significant challenges in terms of (i) capturing detail at appropriate length scales; (ii) rheology; and (iii) devising reduced-order models that are both tractable and capture key features of the flow. Multiscale modelling approaches offer one route. Machine learning algorithms offer opportunities to handle large datasets, and reduce the dimensionality and order of modelling in fluid mechanics. Whether these can accurately predict physically meaningful flow phenomena even for well-defined problems remains to be seen. |
