|The award will cover the tuition fees at the UK rate £4,712, plus a tax-free stipend of £18,622 per annum for 3.5 years of full-time study.
|4th December 2023
|5th January 2024
Qualification: Doctor of Philosophy in Engineering (PhD)
Start date: 1st April 2024
Funding for: 3.5 years
Supervisor: Dr Emmanouil Kakouris & Dr Lukasz Figiel
Early detection of damage in materials is crucial, as cracks reduce local stiffness, can affect structural integrity, and accelerate the ageing process of physical assets. This project will help predict damage degradation in materials and enable mitigation measures to prevent potential failure of structural components, which are critical for ensuring safety and achieving societal objectives.
The aim of the project is to exploit the recent advances in machine learning (ML) and multiscale modelling for simulating damage in materials. With the increasing complexity of emergent materials, predictive structural damage models require mechanistic understanding across different material scales, i.e. multiscale modelling. This research will focus on modelling the link between the micro-material properties of engineering materials and their macroscale mechanical behaviour while retaining adequate precision and accuracy. ML tools will be utilised to pre-process massive amounts of data, integrate and analyse it from different input modalities and different levels of fidelity, identify correlations, and infer the non-linear response of the overall system. The project will focus on developing both deterministic and probabilistic frameworks to predict the response of structural components undergoing damage in real time. The probabilistic model will capture the uncertainties present in the data as well as in the ML-driven physics-based model.
We are looking for candidates to work at the confluence of structural mechanics, uncertainty quantification, and ML, towards addressing the safety and resilience challenges of an ageing, growing, and changing critical infrastructure.
The successful candidate should have an interest in and/or an excellent understanding of computational materials modelling, simulations, probabilistic machine learning, and mathematics for solving partial differential equations, linear algebra, geometric analysis and numerical integration. The candidate will have good programming skills e.g. Python, MATLAB, C++, FORTRAN, an ability to work in a project team and take responsibility for their own research goals.
PhD studentship fully funded to commence in October 2023 or as soon as possible thereafter. The successful candidate would be based in the School of Engineering at the University of Warwick.
The award will cover the tuition fees at the UK rate £4,712, plus a tax-free stipend of £18,622 per annum for 3.5 years of full-time study. International candidates are welcome to apply but would be required to meet the fee difference.
UK candidates with a first-class or 2.1 honours degree at BSc or MSc in engineering disciplines, applied mathematics, physical science or computational science and a strong interest in computational materials modelling, simulations, applied mathematics, probabilistic machine learning & Bayesian statistics, and data science. International students are welcome to apply but must meet the fee difference themselves.
How to apply:
Candidates should submit a formal application, details of how to do so can be found via the 'Apply' button above.
Application form 'Course search':
Department: School of Engineering
Academic Year: 2024/25
Type of Course: Postgraduate Research
In the application form funding section, enter: Source: EK-Early Detection Machine Learning
If you wish to discuss any details of the project informally, please contact Dr Emmanouil Kakouris at Emmanouil.Kakouris@warwick.ac.uk.
The University of Warwick provides an inclusive working and learning environment, recognising and respecting every individual’s differences. We welcome applications from individuals who identify with any of the protected characteristics defined by the Equality Act 2010.
Type / Role: