ÖØ¿ÚζSM

The renewable energy sector is rapidly evolving, with wind energy playing a crucial role in global decarbonization efforts. However, inefficiencies in wind turbine control and maintenance lead to increased operational costs and reduced energy output. Traditional maintenance methods rely on reactive or time-based servicing, which can result in unexpected failures and downtime. This research addresses the urgent need for adaptive, AI-driven solutions that enhance wind turbine performance, reliability, and longevity while supporting the Net Zero 2050 target.

This PhD project will develop an AI-enabled framework that optimizes wind turbine control and predictive maintenance. Using Deep Reinforcement Learning (DRL), the system will dynamically adjust turbine parameters such as yaw, pitch, and torque to maximize Annual Energy Production (AEP) while minimizing component stress. Additionally, a hybrid predictive maintenance model integrating Machine Learning (ML) with physics-based degradation modelling will enhance early fault detection, reducing unplanned downtime.

This PhD is hosted at ÖØ¿ÚζSM University, a global leader in renewable energy, AI-driven engineering, and industrial research. ÖØ¿ÚζSM’s expertise in wind energy systems, predictive maintenance, and AI applications provides an ideal environment for cutting-edge research. The project is supported by EPSRC Doctoral Landscape Awards (DLA) and RES Group, a leading renewable energy company committed to advancing sustainable wind farm operations. RES Group support this research and provide potential secondment opportunities and access to high-resolution SCADA data, industry mentorship, and real-world testing opportunities.

1. A novel DRL-based wind turbine control system that dynamically adjusts parameters for improved efficiency and reduced mechanical stress.

2. A Hybrid AI-Predictive Maintenance model that integrates machine learning with physics-based failure modelling to enable real-time fault detection and predictive maintenance strategies.

3. Validation of AI models with real-world SCADA data, ensuring industry relevance.

4. A digital twin framework for safe, simulation-based validation before deployment in operational wind farms.

1. Collaboration with Industry – The student will work closely with RES Group, gaining hands-on experience with real-world wind farm operations.

2. Secondment and Training Opportunities – The student will undertake a three-month secondment at RES Group and participate in specialized training at Complutense University of Madrid, Spain, focusing on wind turbine predictive control and AI applications.

3. Conference Attendance and Networking – Funding is available for leading conferences such as WindEurope, IEEE Power & Energy Society, and the European Wind Energy Conference, providing exposure to cutting-edge research and industry networking opportunities.

4. Access to High-Fidelity Simulations – The project will use OpenFAST, FAST.Farm, and Digital Twin simulations for AI model validation.

✔ Artificial Intelligence (AI) & Deep Reinforcement Learning (DRL) for energy optimization

✔ Predictive Maintenance & Failure Analysis using Machine Learning and Physics-Based Modelling

✔ Data Science & Advanced Analytics with real-world SCADA data

✔ Renewable Energy Systems & Wind Turbine Engineering

✔ Industry Engagement & Research Commercialization

✔ Scientific Writing & Presentation Skills, preparing research for high-impact journals and international conferences

Along with transferable skills in project management, technical writing, teamwork and public engagement, graduates will be well-positioned for careers in a variety of sectors including academia and industry.