Development of a neural-networks based surrogate model for modelling stirred tank reactors
Thesis type MODELING AND SIMULATION
Description In this thesis project computational fluid dynamics (CFD) simulations will be performed with the objective of investigating fluid flow in stirred tank reactors with different operating conditions and geometries (e.g. stirring rate or fluid viscosity, size/type of the impeller).
The CFD results will be employed to train machine learning algorithms (both fully-connected and convolutional neural networks) to create surrogate models with very low computational cost: these surrogate models will be useful for parametric exploration and for process optimization purposes.
Deadline 02/12/2023 PROPONI LA TUA CANDIDATURA