PhD Application of Artificial Intelligence for Gas Turbine Performance
PhD @Cranfield University posted 1 month agoJob Description
To improve the availability and reduce the overhaul and maintenance (O&M) costs of gas turbine propulsion systems, the gas turbine industry is moving from scheduled maintenance to condition-based predictive maintenance. Gas turbine diagnostics and prognostics has been progressed quickly in recent years and are crucial technologies to predict the health of gas turbine systems and support the predictive maintenance. However, current technologies are relatively slow and not capable enough to provide quick performance, diagnostic and prognostic predictions for real time applications. With the rapid development of artificial intelligence (AI) nowadays, it has become possible to develop a fast-response AI-based condition monitoring system for gas turbine engines.
The objective of the project is to develop novel AI-based performance diagnostic and prognostic technologies and a digital-twin system to support condition-based predictive maintenance of gas turbine engines.
The project will be partially funded by Cranfield University.
Applications are invited for a PhD studentship in the Centre for Propulsion and Thermal Power Engineering, Cranfield University, in the area of gas turbine performance, diagnostics and prognostics to support condition-based predictive maintenance for gas turbine engines.
Cranfield has developed unique physics-based technologies on gas turbine performance simulations, diagnostics, prognostics and lifing methods in the past. A piece of comprehensive computer software, Pythia with the corresponding capabilities have been developed and tested successfully in several industrial applications. The software can be used effectively as a performance digital twin to generate high-quality engine performance models and produce required training data for the proposed project. This could be a good starting point for the project.
The objective of the proposed research project will be the development of AI-based digital twin technology for performance diagnostic and prognostic predictions of gas turbine engines. A novel Machine Learning-based diagnostics and prognostics digital twin system will be developed, aiming to provide fast and reliable predictions of the health of gas turbine engines.
Non-confidential operational gas turbine sensor data, if available, will be utilized to validate the developed digital twin in order to estimate non-measurable health parameters of major gas path components, including compressors, combustor and turbines. It is aimed to make a step forward in gas turbine performance, diagnostics and prognostics in terms of prediction accuracy, real-time applicability and computational efficiency. As for the AI perspective, Machine Learning will be novelly used in order to save time from the training process and improve the robustness and accuracy of the predictions.
The Centre for Propulsion and Thermal Power Engineering has a key focus and a proven track record on gas turbine performance, gas path diagnostics and prognostics, lifing, etc., which have been built up over the last half century. This provides a unique capability to assist researchers and engineers in gas turbine performance and health monitoring in gas turbine and power generation industry.
The history of gas turbine performance engineering at Cranfield dates back to 1946 and the foundation of the Institution. The Centre for Propulsion and Thermal Power Engineering contributes and focuses on gas turbines for aerospace, marine and power generation applications, the sectors where Europe and Britain are world leaders and major exporters. This high technology global industry is worth more than £30 billion per annum. Current challenges are arising from the need to address environmental issues, more cost-effective operations, maintenance and overhauls and the changing economic climate. These challenges have created an environment where a large return can be accrued from an investment in gas turbines and related power system research and education.
It is expected the research will generate new AI-based methods and knowledge for more advanced gas turbine health monitoring. The new knowledge will be very useful to guide condition-based operation and predictive maintenance of future gas turbine engines.
The student will be based within the Centre for Propulsion and Thermal Power Engineering within the Faculty of Engineering and Applied Sciences. Cranfield is a wholly postgraduate university and there are a wide range of MSc and Professional Development Short Courses throughout the year. The student may have opportunities to access some of the MSc courses and CPD short courses relevant to the research and attend and publish papers in international conferences.
Cranfield operates a substantial Doctoral Researchers Core Development programme (DRCD) for its research students. This programme provides a generic structured training programme which is constructed to support the researcher as the PhD progresses with specific courses aimed at the different phases of a PhD. For example, the programme includes aspects such as research methods, technical report writing, presentation skills, data management, leadership skills, professional development planning, intellectual property, publishing, etc. Such knowledge, skills and capability will enhance the student’s employment opportunities in both academic institutions and industry.
This is an exciting opportunity for a suitable candidate where he or she will be exposed to the latest technology, learn from experts working in the area and prepare for an exciting career in either academia or industry.
