PhD Studentship: AI-driven fault diagnosis for wind turbine generators

PhD Studentship: AI-driven fault diagnosis for wind turbine generators

Quick Decision

This UEA project combines drivetrain physics with AI to improve fault detection and classification in wind-turbine generators. It is a close intellectual fit with Dr Salman Abdi Jalebi's work on electrical machines, power electronics and fault diagnosis, and UEA has published both recent machine-diagnosis studies and earlier work on wind-generator electrical faults. The central decision constraint is financial: the opportunity is explicitly self-funded and states no stipend or tuition award. UEA's 2026/27 overseas research-fee table lists £26,400 per year for the relevant school category, while the 2027/28 fee and total candidature cost are not yet established. Before applying, confirm the current deadline, application route and your full fee exposure.

Recommendation: Promising with material uncertainties. The research question is relevant and the supervisor's background is aligned. Applicants should proceed only after determining whether they can fund tuition and living costs and after clarifying what data, equipment and experimental validation are actually available.

Research Problem and Doctoral Work

Wind-turbine condition monitoring commonly relies on vibration measurements, yet drivetrain faults can be difficult to identify early and electrical faults may not be captured well by vibration-first systems. The advertised project targets the drivetrain—gearbox, generator and power electronics—and proposes hybrid methods that combine analytical models grounded in drivetrain physics with AI-based data analysis. The intended outcome is better detection and classification; the description does not promise a particular model or level of improvement.

The listed work includes analytical studies, computer simulation, finite-element analysis, and experimental development and validation. A direct UEA precedent illustrates one possible path: a 2021 study used three-phase stator-current signatures, wavelet processing and a Naïve Bayes classifier to distinguish simulated wind-generator short-circuit faults. It examined a 1,659 kW drivetrain model, so it offers a concrete methodological reference while remaining a simulation study of defined electrical faults. UEA wind-generator fault study

Recent group research gives complementary rather than identical evidence. A 2025 paper applies projection-based SVM to induction-motor bearing faults, while another compares classical algorithms with CNN and Transformer models on a public vibration-bearing dataset. These papers show active work on machine diagnosis and model comparison. They do not demonstrate that those algorithms transfer to wind-turbine electrical signals or that they will be used in this PhD. 2025 projection-SVM study · 2025 classical/deep-learning comparison

The strongest project-specific scientific question is how to connect mechanical fault mechanisms with electrical measurements in realistic operating conditions. Possible research choices include comparing physical-model features with data-only baselines, testing robustness across operating states, and measuring false alarms as well as detection accuracy. These are evidence-informed possibilities, not a disclosed work-package plan. The project record does not identify its generator type, current or voltage channels, vibration sensors, data source, fault taxonomy, AI architecture, test rig, or quantitative validation criteria. Those details will determine both feasibility and the scientific contribution.

School, Programme and Field Context

The degree is hosted and awarded by the University of East Anglia, in its School of Engineering, Mathematics and Physics in Norwich. The school's Sustainable Energy group includes Abdi Jalebi as group lead and works across electrical machines, power electronics and renewable generation. Sustainable Energy group · School research profile

QS lists UEA at =381 in the 2027 overall World University Rankings. UEA reports that 91% of its REF 2021 research was assessed as world-leading or internationally excellent. Both are institution-wide signals; neither establishes an engineering-specific ranking or the quality of this individual project. QS ranking profile · UEA research and REF profile

The public opportunity information allows full-time or part-time study but does not state the registered PhD duration, formal training schedule or dedicated laboratory access. Ask how the research programme will be structured and which experimental facilities are available to this student. University research activity is context, not a reservation of equipment or external data.

UEA's publicly accessible 2021/22 PhD award regulations describe School approval, an appointed supervisory team, approved research, a thesis and oral examination; they give normal periods of three years full-time and six years part-time. These are a historical general-rule reference, not a confirmed duration for this project or evidence that the same version governs a 2027 start. No project-specific taught training plan is published. Confirm the current regulations, the approved period of study and registration, and any required training with UEA before accepting an offer. UEA PhD award regulations (2021/22)

Supervisor and Lab Assessment

UEA identifies Dr Salman Abdi Jalebi as Associate Professor in Electrical Engineering Design and group lead for Sustainable Energy. His research profile lists electrical-machine design and optimisation, power electronics and drives, control, instrumentation, electric-vehicle power trains and renewable generation. This profile maps well to a project that must relate drivetrain physics to machine signals. Supervisor profile

Recent outputs add useful method context. The 2025 bearing-diagnosis comparison tests classical and deep-learning approaches on a public vibration dataset; the 2025 equivalent-circuit review concerns physical modelling of brushless doubly-fed machines; and 2024 work addresses improved modelling of an inductive displacement sensor. Together they support a plausible mix of physical models, measurement interpretation and data-driven diagnosis. Their application domains differ, so applicants should ask which research direction and collaborators are specifically assigned to this doctorate. Equivalent-circuit review · Inductive-sensor modelling study

Public pages do not identify a second supervisor, project-specific research group, guaranteed industrial partner, student cohort, test bench or equipment allocation. Confirm the supervisory split and how the team will provide generator data and experimental validation.

Funding, Application, and Living Cost

The opportunity is marked self-funded. No stipend, fee waiver, research-cost allowance or scholarship is stated. UEA's 2026/27 overseas postgraduate research fee table lists £26,400 per year for the School of Engineering, Mathematics and Physics category. This is a useful current reference for the listed February 2027 start, but it is not a confirmed quote for this exact programme or for the later 2027/28 fee year. Confirm your assessed fee status, annual fee increases, any additional research charges and whether a project award is available. UEA 2026/27 overseas research fees

UEA's 2026/27 accommodation list shows selected campus rooms at roughly £178.50–£209.50 a week, depending on room and licence. These rates cover housing only. A complete budget must also cover food, utilities where excluded, transport, visa and health-related charges, equipment or fieldwork costs, and other personal expenses. UEA 2026/27 accommodation prices

The opportunity record lists a 30 November 2026 application deadline and a planned 1 February 2027 start. Because no project-specific UEA page or full application checklist was located, applicants should confirm both dates and the submission process directly through UEA's postgraduate research route. UEA research-degree application guidance

Applicant Fit and Preparation

The stated minimum is a 2:1 degree in electrical and electronic engineering, mechanical engineering or physics. The opportunity lists full-time and part-time modes. It does not publish a degree duration or a separate preferred-skills checklist. The project methods make machine theory, signal processing, numerical modelling, finite-element analysis, programming, AI and experimental measurement useful preparation, but these should not be presented as additional formal requirements.

An application can make its fit legible by linking relevant modules or research to rotating-machine behaviour, sensor or signal analysis, condition monitoring, simulation and experimental validation. If the applicant has worked only with generic machine-learning datasets, explain how they would test domain shift and measurement quality rather than assuming benchmark accuracy will transfer. The current public record does not specify the documents, references or language-evidence route, so check the application checklist before submitting.

Field Outlook and Career Context

Wind-sector reliability research sits within a growing UK deployment and skills context. The Department for Energy Security and Net Zero's methodology estimates up to 145,000 direct and indirect onshore and offshore wind jobs across Great Britain by 2030 under its assumptions. The Clean Energy Jobs Plan and ORE Catapult's workforce report discuss skills demand and sector scenarios through 2030–35. These figures depend on deployment, supply-chain and policy assumptions; they are not forecasts for this research niche or guarantees of a graduate job. DESNZ wind-job methodology · Clean Energy Jobs Plan · ORE Catapult workforce foresighting

The field opportunity is strongest where reliability analysis can be validated against operational equipment and used to support maintenance decisions. The project will need to show how its methods manage rare faults, changing load and operating conditions, false alarms, and the gap between simulation and measured data. The published project information does not say which of these settings is included.

Next Actions and Questions to Verify

Before Applying

• Confirm whether 30 November 2026 remains the deadline and which documents, references and online application route apply.
• Request the annual tuition fee for the candidate's fee status, the expected 2027/28 rate, any available studentship or stipend, and the expected total period of registration.
• Confirm whether the advertised 1 February 2027 start can be supported for both full-time and part-time applicants.

At Interview

• Which generator architecture, electrical and mechanical signals, fault cases and datasets or test rigs will the PhD use?
• How will analytical modelling, FE analysis, AI methods and experiments be divided and validated?
• Which laboratory resources, industrial partners, computing resources and supervisory roles are committed to this project?

Before Accepting an Offer

• Obtain written confirmation of tuition, any stipend or award, annual fee changes, registration length, part-time terms and international-student conditions.
• Ask how experimental data access and validation milestones will be protected if equipment or partner access changes.

Evidence and Sources

Institution and doctoral application

• UEA supervisor profile
• UEA Sustainable Energy group
• School of Engineering, Mathematics and Physics
• UEA research profile and REF 2021 summary
• QS World University Rankings 2027 profile
• UEA postgraduate research application guidance

Supervisor and research

• Projection-based SVM for induction-motor bearing faults
• Classical and deep-learning bearing-diagnosis comparison
• Brushless doubly-fed machine circuit-model review
• Inductive displacement sensor modelling
• Rotational iron losses in brushless doubly-fed machines
• Sustainable Energy research outputs
• Wind-generator short-circuit diagnosis
• Eccentricity fault detection in doubly-fed machines
• Fault diagnosis of brushless doubly-fed machines

Costs and sector context

• UEA overseas postgraduate research fees 2026/27
• UEA accommodation price list 2026/27
• DESNZ methodology for wind-job estimates
• UK Clean Energy Jobs Plan
• ORE Catapult workforce foresighting
• The Crown Estate Offshore Wind Report 2025