Intelligent wind turbines for optimised energy production
Elie Kadoche, PhD candidate at Télécom Paris, LTCI lab, Institut Polytechnique de Paris, June 2024.
Key words: low consumption, energy saving, optimisation
Elie Kadoche talks to us about digital technology in the service of the energy transition, and more specifically about the subject of his thesis on intelligent wind turbines for optimised energy production.
He was able to sum up the subject in three minutes, winning 3rd prize from the jury of the French national « My thesis in 180 seconds » competition, out of 16 finalists. He had previously won the IP Paris regional final.
Interview by Isabelle Mauriac, in French with English subtitles
Elie Kadoche works on control algorithms for optimising large wind farms.
In this video, he explains how the use and development of a reinforcement learning method enables wind turbines to orientate themselves with the wind.
Video Michel Desnoues, Télécom Paris
References
- Kadoche, E. et al. How to craft a deep reinforcement learning policy for wind farm flow control (2025) arXiv : 2506.06204 [ cs.LG ]
- Tuhfe Göçmen, Jaime Liew, Elie Kadoche, Nikolay Dimitrov, Riccardo Riva, Søren Juhl Andersen, Alan W.H. Lio, Julian Quick, Pierre-Elouan Réthoré, and Katherine Dykes. Data-driven wind farm flow control and challenges towards field implementation: A review. In: Renewable and Sustainable Energy Reviews 216 (2025), p. 115605. issn: 1364-0321
- Elie Kadoche, Pascal Bianchi, Florence Carton, Philippe Ciblat, and Damien Ernst. On the importance of wind predictions in wake steering optimization. In: Wind Energy Science 9.7 (2024), pp. 1577–1594
- Elie Kadoche, Sébastien Gourvénec, Maxime Pallud, and Tanguy Levent. MARLYC: Multi-Agent Reinforcement Learning Yaw Control. In: Renewable Energy 217 (2023),p. 119129. issn: 0960-1481
