PhD Studentship: Scalable Digital Twins with Advanced Machine Learning Integration for Offshore Wind turbine Operation and Maintenance

Qualification Type: PhD

Location: Cranfield University

Funding for: UK Students

Funding amount: A bursary will be provided of up to £20,000 tax-free plus fees for three years

Hours: Full time

Closes: 19/06/2025

Supervisors: Dr Ravi Pandit   

Cranfield University, in collaboration with Semtronics UK and international academic partners, is offering a PhD studentship focused on enhancing the efficiency and reliability of offshore wind turbines through advanced digital twin and machine learning technologies. This project will investigate existing digital twin and machine learning models and find knowledge gaps, leverage public and industrial datasets to develop scalable machine learning based digital twin models to improve performance, decision-making and reduce costs. A bursary will be provided of up to £20,000 tax-free plus fees for three years.

While digital twins are emerging in various sectors, their application in wind turbines remains underexplored, marked by critical knowledge gaps in integration with advanced machine learning. Research in comprehensive digital twins for wind farms, integrating diverse data sources, is limited. Current use of machine learning in predictive maintenance lacks depth in advanced algorithms like deep learning, essential for complex data. There’s also a gap in real-time data processing methods for immediate operational adjustments. Furthermore, the use of synthetic data for digital twin 4alidationn, especially against real-world conditions, is not well-developed. Finally, the scalability and adaptability of these models across different wind farm conditions is a significant challenge, with most models being turbine specific and not universally applicable. Addressing these gaps is crucial for enhancing wind turbine performance and reducing costs.

This PhD project aims to bridge key gaps in wind energy optimization by integrating digital twin technology with advanced machine learning. It involves developing a holistic digital twin model that incorporates various data from wind farm operations, serving as a platform for applying and honing sophisticated machine learning methods like deep learning for enhanced predictive maintenance. The project prioritizes real-time data processing for dynamic operational adjustments and employs both synthetic and real sensor data for thorough model validation, ensuring robust performance in diverse conditions. Additionally, the project will explore the potential benefits of using digital twins to optimize the performance of wind farms, while identifying any associated limitations or challenges. 

Entry Requirements

Applicants should have a 1st or 2.1 UK degree or an equivalent in a discipline related to electrical engineering, energy, or computer science. The ideal candidate should have background of electrical and computer and have strong programming experiences for wind turbines. The candidate should be self-motivated, possess good communication skills for regular interaction with other stakeholders, with an aptitude for industrial research.

Applicants should be UK Nationals. 

Funding & Sponsorship

Sponsored by EPSRC and Cranfield University.

Advert information