EPJ E Highlight - Machine learning could help kites and gliders to harvest wind energy

Powering a ship with a kite

Using trial-and-error, machine learning algorithms could enable flying wind harvesters to dynamically adjust their orientations, allowing them to account for unpredictable turbulence and improve their performances.

Airborne wind energy (AWE) is a lightweight technology which uses flying devices including kites and gliders to harvest power from the atmosphere. To maximise the energy they extract, these devices need to precisely control their orientations to account for turbulence in Earth’s atmosphere. Through new research published in EPJ E, Antonio Celani and colleagues at the Abdus Salam International Center for Theoretical Physics, Italy, demonstrate how a Reinforcement Learning algorithm could significantly boost the ability of AWE devices to account for turbulence.

With far lower construction costs than traditional wind turbines, AWE could prove immensely valuable in expanding the reach of wind power to poorer, more remote communities. To extract wind energy, flying devices are either tethered to a ground station, where power is converted into electricity, or used to tow a vehicle. The main challenge faced by this technology is to maintain its performance in widely varying wind and weather conditions. To do this, researchers currently use computer models to predict the future state of the atmosphere, allowing kites and gliders to dynamically adjust their orientations. However, since turbulence requires an immense amount of computing power to approximate precisely, it is often ignored in existing models, leading to suboptimal performances in AWE systems.

In their study, Celani’s team addressed the issue using Reinforcement Learning: a machine learning algorithm which uses trial-and-error interactions with the surrounding environment to calculate which orientation of a kite or glider will extract the maximum possible energy from the atmosphere. As a proof of concept, the researchers applied the algorithm to a simulated ship being towed along by a kite. When issued with a simple set of manoeuvring instructions, the kite used Reinforcement Learning to tow the ship over long distances, even with no prior knowledge of the turbulence it would encounter. With the early success of their approach, Celani and colleagues now hope that the use of Reinforcement Learning could soon enable the reach of AWEs to expand even further in the future.

Orzan, N., Leone, C., Mazzolini, A. et al. Optimizing airborne wind energy with reinforcement learning. Eur. Phys. J. E 46, 2 (2023).

This was our first experience of publishing with EPJ Web of Conferences. We contacted the publisher in the middle of September, just one month prior to the Conference, but everything went through smoothly. We have had published MNPS Proceedings with different publishers in the past, and would like to tell that the EPJ Web of Conferences team was probably the best, very quick, helpful and interactive. Typically, we were getting responses from EPJ Web of Conferences team within less than an hour and have had help at every production stage.
We are very thankful to Solange Guenot, Web of Conferences Publishing Editor, and Isabelle Houlbert, Web of Conferences Production Editor, for their support. These ladies are top-level professionals, who made a great contribution to the success of this issue. We are fully satisfied with the publication of the Conference Proceedings and are looking forward to further cooperation. The publication was very fast, easy and of high quality. My colleagues and I strongly recommend EPJ Web of Conferences to anyone, who is interested in quick high-quality publication of conference proceedings.

On behalf of the Organizing and Program Committees and Editorial Team of MNPS-2019, Dr. Alexey B. Nadykto, Moscow State Technological University “STANKIN”, Moscow, Russia. EPJ Web of Conferences vol. 224 (2019)

ISSN: 2100-014X (Electronic Edition)

© EDP Sciences