PhD Studentship in Aeronautics: Reinforcement Learning for controlling nonlinear physical systems: application to wind farm optimization (AE0057)

Start Date: Between 1 August 2025 and 1 July 2026

Introduction: Reinforcement Learning (RL) offers a powerful approach for tackling the complexities of controlling nonlinear physical systems. These systems are known for their intricate behaviors, where small variations in input can lead to significant shifts in output, or conversely, large inputs may only result in minimal changes. Additionally, nonlinear systems can exhibit multiple solution paths, where identical conditions yield different outcomes, or demonstrate ‘memory’ through path dependence.

The proposed research project will explore the potential of RL for optimising the power outcome of wind farms for a wide range of operational conditions. Wind farms have become a significant source of renewable energy, yet optimising their energy output remains a challenging task due to fluctuating wind conditions, turbine wear, and other environmental factors. Traditional control strategies often fail to account for the highly dynamic nature of wind speeds and the complex interactions between turbines. This project aims to harness RL to maximise the energy efficiency of wind farms, improving their overall power output while minimising operational costs.

Objectives: The objectives are:

• Develop a reinforcement learning model to learn and optimise the power output of individual turbines in a wind farm.
• Incorporate environmental and operational constraints (e.g., wind speed, turbine wear, wake effects) into the RL model to ensure safe and sustainable operation.
• Evaluate the performance of the RL-based optimisation in comparison to traditional control methods under various wind scenarios.
• Minimise operational costs and maximise turbine lifetime through efficient management of turbine control actions, such as pitch and yaw adjustments.

Supervisors: Professor Sylvain Laizet, expert in computational fluid dynamics and high performance computing (www.turbulencesimulation.com) and Dr Georgios Rigas, expert in flow control and reinforcement learning.

Learning opportunities: You will develop knowledge and expertise in high performance computing, reinforcement learning, computational fluid dynamics and turbulent flow control.

Professional Development: You will have access to engaging professional development workshops in areas such as research communication, computing and data science, and professional progression through our Early Career Researcher Institute.

Duration: 3.5 years.

Funding: Full coverage of tuition fees and an annual tax-free stipend of £21,237 for Home, EU and International students. Information on fee status can be found on our Fees and Funding webpages.  

Eligibility: You should have a keen interest and solid background in computational fluid dynamics, programming, machine learning and/or in high performance computing. You must possess (or expect to gain) a First class honours MEng/MSci degree or Distinction in a Master’s level degree in a relevant scientific or technical discipline.

How to apply: Submit your application via our Apply webpages. When submitting your application, you will need to use the following details:

• Search course/Programme: Aeronautics Research (PhD)
• Research Topic: Please use reference number AE0057
• Research Supervisor: Prof. Sylvain Laizet and Dr Georgios Rigas
• Research Group: Aero

For further information: For questions about the project, please email Sylvain Laizet, Professor in Computational Fluid Mechanics: s.laizet@imperial.ac.uk

For queries regarding the application process, email the PhD Administrator.

Application deadline: 9 January 2025

Equality, Diversity and Inclusion: Imperial is committed to equality and valuing diversity. We are an Athena SWAN Silver Award winner, a Stonewall Diversity Champion, a Disability Confident Employer and are working in partnership with GIRES to promote respect for trans people.