user_image

Andrew Mole

I am a Research Associate in the Turbulence Simulation Group at Imperial College London. My current work is on the AI for Net Zero Project investigating Physics Aware Machine Learning for Wind Farm Optimisation.

EXPERIENCE

  • 2023 – Now

    Research Associate, Imperial College London

  • 2022 – 2023

    Research Assistant, University of Manchester

  • 2018 – 2022

    PhD Researcher, University of Manchester

  • 2018 – 2021

    Graduate Teaching Assistant, University of Manchester

  • 2021 – 2021

    CFD Methodology PhD Intern, McLaren Formula One Team

RESEARCH INTERESTS

  • Computational Fluid Mechanics
  • Machine Learning
  • Multi-fidelity Methods
  • High Performance Computing
  • Bayesian Optimisation
  • Reinforcement Learning

SKILLS

  • Python
  • Fortran
  • C++
  • XCompact3D
  • OpenFOAM
  • Star-CCM+

PROJECTS

Optimised Wind Farm Wake Control

01/2024 –

Reinforcement Learning for Dynamic Wind Farm Control
Reinforcement learning is applied to a LES wind farm environment to find dynamic control stratagies for improving power output.
Optimised Wind Farm Wake Control

07/2023 – 12/2024

Multi-Fidelity Bayesian Optimisation for Wind Farms
A multi-fidelity Bayesian optimisation method is applied to wind farm wake steering for increasing the power output.
Natural convection flow

11/2022 – 07/2023

Multi-Fidelity Methods for Natural Convection
Developing a 1D-3D multifidelity framework for application to the simulation of natural convection flow.
Diagram showing a multi-fidelity neural network

10/2021 – 10/2022

Multi-Fidelity Surrogate Modelling
Developing a multi-fidelity surrogate modelling framework for application to CFD data from RANS and LES.
Vortices around front wing.

01/2021 – 10/2021

Formula One Front Wing
Application of Embedded LES approach to a McLaren Front wing geometry to capture the vortex dynamics.
a slice of the velocity around tandem cubes using ELES

07/2018 – 10/2021

Embedded Large Eddy Simulations
Developing a nested Embedded Large Eddy Simulation approach for turbulent flow simulations.

RECENT PUBLICATIONS

[All Publications]
  1. Reinforcement Learning Increases Wind Farm Power Production by Enabling Closed-Loop Collaborative Control
    Mole, A, Weissenbacher, M, Rigas, G, and Laizet, S
    arXiv, 2025
  2. Multi-Fidelity Bayesian Optimisation of Wind Farm Wake Steering using Wake Models and Large Eddy Simulations
    Mole, A, and Laizet, S
    Flow, Turbulence and Combustion, 2024
  3. Multi-Fidelity Surrogate Modelling of Wall Mounted Cubes
    Mole, A, Skillen, A, and Revell, A
    Flow, Turbulence and Combustion, 2023