A funded PhD studentship is available within the Autonomous and Cyber Physical Systems Centre at Cranfield University, Bedfordshire, UK. Read more Read less

As aerospace platforms go through their service life, gradual performance degradations and unwarranted system failures can occur. There is certain physical information known a priori in such aerospace platform operations. The main research hypothesis to be tested in this research is that it should be possible to significantly improve the performance of extreme learning and assure safe and reliable maintenance operation by integrating this prior knowledge into the learning mechanism.

The integrating should enable to guarantee certain properties of the learned functions, while keep leveraging the strength of the data-driven modelling. Most of, if not all, the traditional statistical methods are not suitable for big data due to their certain characteristics: heterogeneity, statistical biases, noise accumulations, spurious correlation, and incidental endogeneity.  Therefore, big data demands new statistical thinking and methods. As data size increases, each feature and parameter also becomes highly correlated. Then, their relations get highly complicated too and hidden patterns of big data might not be possible to be captured by traditional modelling approaches.

This implies that mathematical modelling of such data is infeasible. The data-driven modelling approach could resolve this issue and we could use obtain data-driven models using machine learning algorithms such as artificial neural networks, reinforcement learning, and deep learning.  A typical caveat of data-driven modelling using learning algorithms as Extreme Learning Machine (ELM) is that training data should cover the entire domain of process parameters to achieve accurate generalization of the trained model to new process configurations. In practice, this might not be possible, that is the sample data could cover only some space, not entire space, of process parameters. Integrating prior knowledge into the learning could enable accurate generalization of the data-driven model even when the space of system parameters is only sampled sparsely.

Consequently, it will improve the performance of the learning. Integration of the prior knowledge of the system into the learning procedure will be quite challenging since the key enabler of its very powers is the universal approximation capabilities.  Sampled data are generally noisy, outliers occur, and there always exist a risk of overfitting corrupted data. Therefore, the learned function may violate a constraint that is present in the ideal function, from which the training data sampled. This PhD study will address this research challenge.

CRANFIELD UNIVERSITY

Cranfield is the largest academic centre for postgraduate studies in Science and Technology in the UK. Focused on developing applied research to meet the demands of industry. Ranked in the top five UK universities for commercial income for research, Cranfield is able to invest in and has become home to many world-class, large-scale facilities, which enhance our teaching and research.

Our specialist areas of focus, or Cranfield themes, are where we bring a range of academic disciplines together in order to tackle the grand challenges facing the world within a range of industrial and commercial sectors. These are Water, Agrifood, Energy and Power, Aerospace, Manufacturing, transport systems, Defence and Security and management.

At a glance

  • Application deadlineOngoing
  • Award type(s)PhD
  • Duration of award3 years
  • EligibilityEU, UK
  • Reference numberCRAN1148

Supervisor

Prof Tsourdos and Dr Shin

Entry requirements

For this research studentship, we are seeking a talented graduate, having (or be expected to obtain) at least an upper second class honours degree (first class honours preferred), MSc or equivalent in Mechanical, Electrical Engineering, Control Engineering, Aerospace or Computer Science. Good mathematical background and experience with Matlab/Simulink, C++ and real time implementations and programming would be most desirable.

Funding

The studentship is for three years and provides payment of tuition fees at the UK/EU rate (£9,000 per annum) plus a £14,300-18,000 per annum contribution to living expenses.

We are looking for a UK national student (or a European resident in the UK for last three years) only. 

Cranfield Doctoral Network

Research students at Cranfield benefit from being part of a dynamic, focused and professional study environment and all become valued members of the Cranfield Doctoral Network. This network brings together both research students and staff, providing a platform for our researchers to share ideas and collaborate in a multi-disciplinary environment. It aims to encourage an effective and vibrant research culture, founded upon the diversity of activities and knowledge. A tailored programme of seminars and events, alongside our Doctoral Researchers Core Development programme (transferable skills training), provide those studying a research degree with a wealth of social and networking opportunities.


How to apply

To apply send a CV and a covering letter by email to: Dr Hyo-Sang Shin, h.shin@cranfield.ac.uk

For further information contact us today:
T: 44 (0)1234 758540
E: study@cranfield.ac.uk