Condition monitoring of high performance automotive gearbox bearings

This exciting project, in collaboration with a world leading motorsport (Formula 1) company, is concerned with the detection and classification of bearing defects within high performance motorsport transmission gearboxes. The project builds on previous projects investigating gear damage detection, which involved the construction of an advanced gear testing rig. This rig will be used during this project to test bearings, both with artificially seeded defects and subsequently in life tests where bearings are allowed to fail naturally.

A range of monitoring techniques will be investigated, including vibration, acoustic emission and novel magneto-elastic torque monitoring. Appropriate implementations of simple metrics and more advanced signal analysis techniques would be used to analyse both the test rig data, and data collected by the sponsor on state of the art dynamometer facilities using actual gearboxes. The database of test results will also be used to develop and train advanced defect detection algorithms, which the student will develop using machine learning (ML) techniques to identify and classify defects based on multiple metrics extracted from sensors (e.g. frequency based metrics or statistical moments).

The idea is to fuse all available data into damage severity (quantification) and damage type (classification) parameters for a range of bearing defects. This will be made possible by the selection of appropriate architectures of Artificial Neural Network. Furthermore, as machine learning techniques become more established within the sponsor’s organisation, there is scope for the student to evaluate commercial software tools using the bearing data.

This opportunity would suit a student interested in careful experimental work and the development of novel condition monitoring techniques, in support of the sponsor’s continuing quest for performance and reliability in their motorsport operations.

Requirements: 2:1 or above undergraduate degree in Mechanical, Integrated or Systems Engineering, or a related subject.