About the project
The occurrence of adhesive wear between sliding surfaces can cause high friction, vibration, material transfer between surfaces and even seizure of components. This wear process is known as galling and is the least well understood wear mechanism induced by tribocontacts. Various palliatives are often used to minimise wear and prevent wear or seizure in sliding contacts. Systematic studies have been undertaken with different alloy systems, but the stochastic nature of galling has often left researchers unable to pinpoint to the drivers for galling or explain why certain systems work and others not.
This project aims to look at adhesive wear resistance of metal surfacing using a data centric approach. The data from three sources: open literature, RR testing and Soton testing will be used. The model will deploy the latest machine learning algorithms that can be constrained by known physics and are capable of predicting a stochastic process. It will use test conditions, surface roughness and surface composition to build a predictive data driven model for alloy selection and operation.
This project will be sponsored by Rolls-Royce Submarines, Derby and the student will receive invitations to their annual sponsored student conference. The student would have support to attend two international conferences and would gain skills in surface engineering, advanced experimental techniques, machine learning/AI, modelling and research-industry interactions.
