College of Engineering Centenary PhD Scholarship: New Generation of Real-Time Response Models Based on Deep Learning
Subject areas: Engineering
Project description: Complex large-scale dynamical system often has a vast number of degrees of freedom. It is not possible to model the system sufficiently quickly to predict behaviours in real time. This project will extract effective data to construct a fast model using deep learning methods. Deep learning is part of a broader family of machine learning methods and has been applied to a number of fields.
The proposed PhD project offers the unique opportunity to develop a data-driven reduced order model (ROM) using deep learning methods. It provides a fast way of performing computationally intensive tasks with real-time speed for fluids problem. The proposed new ROM is constructed from a number of simulations representing different parameters such as different initial or boundary conditions. The deep learning methods are used to extract features from those simulations. After constructing the ROM, it is able to run the simulations with several orders of magnitude speedup.
Project supervisor: Dr Dunhui Xiao
UK/EU candidates: The scholarship covers the full cost of UK/EU tuition fees and an annual stipend of £14,296.
International candidates: The scholarship covers the full cost of tuition fees only.
There will be additional funds available for research expenses for UK/EU Candidates.