PhD Studentship – Automated and Defect-free Forming of Complex Composite Parts with Machine Learning

University of Bath – Mechanical Engineering

Supervisor name: Evros Loukaides & Vangelis Evangelou (Department of Mathematical Sciences)

Research Centre: Materials and Structures (MAST)

Overview of the research
Major penalties arise from the lack of predictive modelling and the uncertainty associated with current design and manufacturing methods in composites, impacting on production cost and environmental footprint. This leads to over-conservativism in design, resulting in unnecessarily high operational costs and greenhouse-gas emissions; it also necessitates over-stringent manufacturing quality requirements associated with ’defect-free’ part production policies.

This project aims to explore the use of an actuated tool to improve formability and reduce defects in challenging geometries. The optimal tool path will be calculated using machine learning methods, while training and testing data will be provided by a combination of laboratory work and numerical modelling. Appropriate digital representations of the geometry will be developed, both for the tool path trajectory and for the shape of the part. These will respectively serve as the input and output of the resulting algorithm.

We are looking for a mathematically-able student with good skills in applied mechanics and an interest in the modelling and testing of composites, to develop computational models and analytical methods. Our partners in the project, GKN Aerospace (UK) and Fokker (Holland), will provide industrial context and realistic problems for analysis and testing, and you will be expected to liaise with practicing engineers and industrial researchers.  To work on this project you will need to have, or be able to develop, good programming and experimental skills; you can expect to be working in a multidisciplinary team within a fast-developing field in a high-technology industrial sector, where new ideas are valued and implemented.

A Home/EU award will cover tuition fees, a training support fee of £2,000/annum, and a tax-free maintenance payment of £17,000 per year (2017-8 rate) over at least 3.5 years.

Funding is provided by GKN Aerospace (UK)

Preferred start date: 2nd October 2017


Leave a Reply