Data-Driven Fluid Mechanics

Learning Outcomes

Subject Specific Intellectual and Research Skills

Having successfully completed this module you will be able to:

• Select appropriate methods to quantify and characterize flow behaviour, discussing principles and limitations of the techniques employed

Full CEng Programme Level Learning Outcomes

Having successfully completed this module you will be able to:

• The coursework and the Blackboard tests ask students to analyse complex fluid flows to reach substantiated conclusions regarding, e.g., their low-dimensional behaviour and stability characteristics. These tasks involve utilising a range of analytical and computational techniques of increasing sophistication, and require using domain specific knowledge to evaluate the limitations of the techniques employed.
• The module relies extensively on computer labs to practice first-hand the data analysis techniques and other computational methods covered in the lectures. The skills they are taught when attending these labs are key to successfully complete the coursework (80% of the module mark).
• Coursework for this module consist in preparing 10-page reports introducing the techniques utilised, and presenting the results obtained. Part of the marking scheme for the coursework assesses how effectively the students can communicate their technical work, including clarity and depth of writing, preparation of high-quality and readable figures, and whether results are analysed critically.
• For one of the coursework, students prepare a 10-page report summarising the analysis of a fluid dynamics dataset provided to them. Students are asked to review technical literature on the topic, either on the data analysis technique covered in the lectures or on the particular fluid problem under investigation and include some of it in the introduction section.
• The coursework and the Blackboard tests assesses how well students can utilise a range of computational techniques grounded on fundamental concepts of linear algebra, dynamical systems theory, fluid dynamics and optimisation, to analyse fluid systems and solve complex problems such as model-order reduction, data assimilation and flow reconstruction. The module is heavily informed by the latest technical literature and the syllabus include newest developments and trendy research topics.
• A major focus of the module, and of the assessment, is to utilise a variety of advanced techniques to model turbulent flows, to tackle for instance, model order reduction and flow reconstruction problems. This requires students to compare different methods, and evaluate the difference in modelling accuracy/performance.

Knowledge and Understanding

Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:

• Fundamental concepts of flow stability, dimensionality reduction and reconstruction for fluid flows
• Data-driven methods for flow analysis and modelling

Subject Specific Practical Skills

Having successfully completed this module you will be able to:

• Apply data-driven methods to large fluid dynamics data sets, utilising suitable computational techniques and tools

Transferable and Generic Skills

Having successfully completed this module you will be able to:

• Study and learn independently, including engaging with technical literature
• Manipulate and analyse large datasets and present key information in digestible ways

Model Order Reduction

• Modal analysis, dimensionality reduction and coherent structures in turbulent flows
• Manipulation and handling of numerical and experimental data
• Proper Orthogonal Decomposition: derivation, implementation and interpretation
• Fundamental concepts of flow stability
• Linearisation of the equations of motion, eigenspectra and stability modes
• Dynamic Mode Decomposition: derivation, implementation and interpretation

Flow Reconstruction

• Cost functions and adjoint-based optimisation
• Data assimilation methods for solving inverse problems and identifying unknown parameters
• Introduction to neural networks and their relationship to proper orthogonal decomposition
• Flow reconstruction with physics-informed neural networks
• Neural network achitectures and applications in flow modelling

The module features a series of lectures where data handling and computational techniques are introduced and motivated in the context of specific fluid flow phenomena. Case studies are used to illustrate how specific techniques can be utilised to gain fundamental understanding of flow behaviour and characteristics. The lectures are supported by laboratory sessions where practical data manipulation and flow analysis techniques are demonstrated on benchmark problems. Background reading, self-study and peer-to-peer learning will complement your learning.

Summative

This is how we’ll formally assess what you have learned in this module.

Referral

This is how we’ll assess you if you don’t meet the criteria to pass this module.

Repeat

An internal repeat is where you take all of your modules again, including any you passed. An external repeat is where you only re-take the modules you failed.