Data-Driven Physics-Informed Reliability Prediction of Power Electronics for Net Zero Applications

About the Project

Start date: 30 September 2026

Power electronics are essential for converting power in many net-zero applications, including electrified transportation and renewable energy systems. However, they account for many system failures, driven by interrelated factors such as high electrothermal stress in power semiconductors.

To address this challenge, this project—delivered in partnership with Toshiba Europe—will develop a new physics-informed, data-driven approach for accurately modelling the reliability of power electronics in net-zero applications.

It will create new knowledge of power electronics reliability modelling and prediction, uncovering an accurate correlation between components’ electrothermal stress, failure mechanisms and lifetime. The project will also develop high-quality lifecycle datasets (i.e. through experimental tests and industry input), enabling robust statistical learning and hybrid modelling approaches for reliability prediction.

The approach will be further extended to create a real-time, multi-modal health monitoring and prognostics framework, integrating electrothermal, operational, and potentially hybrid data streams of power electronics. This will enable responsive lifetime prediction under diverse mission profiles representative of real-world net-zero systems.

Beyond reliability prediction, the project will explore creative extensions of prognostics modelling, such as sustainability-aware modelling (e.g. lifecycle efficiency and degradation-aware operation), as well as multi-modal data fusion for enhanced device failure prediction and early warning mechanisms.

These innovations aim to uncover deeper correlations between electrothermal stress, failure mechanisms, and system lifetime, ultimately reducing failure rates and enhancing the performance and sustainability of renewable energy systems.

The student will be based in the Autonomous Systems and Connectivity (ASC) Division at the University of Glasgow, a dynamic and collaborative research environment with strong international links and expertise in advanced engineering and applied sciences.

ASC Website: https://www.gla.ac.uk/schools/engineering/research/asc/

The student will receive close supervision through weekly meetings and will benefit from both academic and industrial mentorship. As part of the project, the student will have the opportunity to undertake a placement of at least six months with the industrial sponsor, with possible extensions.

The research outcomes will be disseminated through leading international conferences and engagement with communities such as the IEEE Power Electronics Society (PELS UK&I Chapter), as well as through industry collaboration. Developed prototypes will be evaluated in real operational environments to deliver impact beyond academia.

Supervision team

Dr Sheng Wang
https://www.gla.ac.uk/schools/engineering/staff/shengwang/

Dr Wenjuan Song
https://www.gla.ac.uk/schools/engineering/staff/wenjuansong/

Prof. David Flynn
https://www.gla.ac.uk/schools/engineering/staff/davidflynn/

Industry supervisor:
Dr Jenny Feng
https://www.linkedin.com/in/zhengyang-jenny-feng/

Exemplar papers

• Machine Learning Pipeline for Power Electronics State of Health Assessment and Remaining Useful Life Prediction (DOI:10.1109/ACCESS.2024.3460177)
• A Machine Learning Degradation Model for Electrochemical Capacitors Operated at High Temperature ( DOI: 10.1109/ACCESS.2021.3057959)
• AI-Driven Maintenance Support for Downhole Tools and Electronics Operated in Dynamic Drilling Environments (DOI:10.1109/ACCESS.2020.2990152)
• A Fast and Adaptive LSTM-based Surrogate Model for Predicting Limitation Performance of SFCLs in Electric Aircraft Systems (DOI: 10.1109/TASC.2026.3673790)
• Modelling and optimal design of a multifunctional single-stage buck-boost differential inverter (DOI:10.1109/OJPEL.2024.3445313)
• Prediction of Active Gate Drivers' Performance Using GAN-Augmented Data (DOI:10.1109/ECCE-Europe62795.2025.11238950)

Learning and Development Opportunities

• Close collaboration with the industrial sponsor, including co-supervision and mentorship
• Training in advanced software, hardware, and technologies for power semiconductor, power electronics, and AI applications.
• Opportunities to attend workshops, conferences, and seminars.
• Opportunities to lead or contribute to IEEE PELS student activities in Scotland.
• Access to unique facilities, including the CryoElectric Research Lab at the University of Glasgow and Power Electronics Labs at Toshiba Europe.

Academic Criteria

Candidates should hold or expect to gain a first-class degree or a good 2.1 (or their equivalent) in Engineering or a related subject.

Desirable skills:

• Modelling and analysis of power semiconductor devices and applications
• Reliability modelling and lifetime prediction
• Computational intelligence and machine learning methods
• Experience with MATLAB/Simulink, SolidWorks, PSpice, or equivalent tools
• Hardware experience in power electronics systems (e.g. power/thermal cycling test, double pulse test, power converter operation and prototyping)

Applicants whose first language is not English will be required to demonstrate proficiency in the English language (IELTS 6.5 or equivalent)

Academic Programme

Electrical and Electronics Engineering (full time)

How to Apply:

If you are interested in this opportunity, please contact Dr Sheng Wang (Sheng.Wang@glasgow.ac.uk) with a CV and a one-page cover letter to discuss your suitability. Interview will be done on the rolling-basis during the advertisement period.

Funding Notes

This scholarship is funded under an EPSRC IDLA partnership with Toshiba Europe. This scholarship will cover your tuition fees (home rate) and provide a competitive stipend (over £22,000) for four years.

International applicants would have to cover the tuition fee difference between Home and International rates. This may be met through external funding or self-funded top-up.