Noise Steering: Exploiting Wake Steering and Acoustic Directivity for Noise-Constrained Wind Farm Control via Reinforcement Learning
Abstract
Existing noise mitigation strategies for wind farms usually consist in reducing turbines power output, without accounting for the wake interactions between them and the noise emission directivity. This leads to over-conservative policies and unexploited wind power. Yet, turbine yaw can simultaneously affect wake interactions and the direction of noise emission, offering an additional degree of freedom for farm-level noise control. We therefore propose Noise Steering, a model-free reinforcement learning (RL) strategy that jointly controls turbine yaw angles and power setpoints to maximise farm power while satisfying a local noise constraint in a defined region. To enable training at a reduced computational cost, a low-fidelity acoustic model is coupled with the steady-state FLORIS simulator. The RL agent effectively learns how to redirect wakes and noise directivity away from the protected region while redistributing power across the farm. On a six-turbine test farm, the policy satisfies the 40 dB noise constraint throughout the full range of turbulent winds, with an average power deficit of only 2% relative to greedy unimpeded control, and up to 7.7% at high wind speeds. In comparison, the industry-standard noise-reduced operation (NRO) baseline here generates an average power loss of 17%. When deployed in the higher-fidelity FASTFarm simulator, the policy largely retains its performance under steady and turbulent inflow, despite differences in simulation fidelity. These results demonstrate the potential of combining yaw control and power derating through RL to improve the power-noise trade-off in wind farm operation.
Related
- Assessing Basin Amplification and Rupture Directivity Effects Using Physics-Based Ground Motion Simulations in Wellington
- Steering Diffusion Priors with Sparse Observations for High-Resolution Temperature Downscaling
- Multi-frequency seismo-acoustic measurements of synthetic ice with fluid inclusions: implications for seismic and through-ice acoustic estimation of sea-ice thickness
- Effects of waveguide tube geometry and surrounding medium conditions on acoustic emission signal attenuation
- DEMONet: Underwater acoustic target recognition based on multi-expert network and cross-temporal variational autoencoder
BibTeX
@misc{frutos2026noise,
title = {{Noise Steering: Exploiting Wake Steering and Acoustic Directivity for Noise-Constrained Wind Farm Control via Reinforcement Learning}},
author = {Martín de Frutos and Baptiste Corban and Claire Bizon Monroc and Ana Bušić and Donatien Dubuc and Jiamin Zhud and Esteban Ferrer},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.36888},
eprint = {2610.36888},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.36888},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Noise Steering: Exploiting Wake Steering and Acoustic Directivity for Noise-Constrained Wind Farm Control via Reinforcement Learning
AU - Martín de Frutos
AU - Baptiste Corban
AU - Claire Bizon Monroc
AU - Ana Bušić
AU - Donatien Dubuc
AU - Jiamin Zhud
AU - Esteban Ferrer
PY - 2026
DA - 2026/10/28
PB - arXiv
DO - 10.48550/arxiv.2610.36888
UR - https://arxiv.org/abs/2610.36888
AB - Existing noise mitigation strategies for wind farms usually consist in reducing turbines power output, without accounting for the wake interactions between them and the noise emission directivity. This leads to over-conservative policies and unexploited wind power. Yet, turbine yaw can simultaneously affect wake interactions and the direction of noise emission, offering an additional degree of freedom for farm-level noise control. We therefore propose Noise Steering, a model-free reinforcement learning (RL) strategy that jointly controls turbine yaw angles and power setpoints to maximise farm power while satisfying a local noise constraint in a defined region. To enable training at a reduced computational cost, a low-fidelity acoustic model is coupled with the steady-state FLORIS simulator. The RL agent effectively learns how to redirect wakes and noise directivity away from the protected region while redistributing power across the farm. On a six-turbine test farm, the policy satisfies the 40 dB noise constraint throughout the full range of turbulent winds, with an average power deficit of only 2% relative to greedy unimpeded control, and up to 7.7% at high wind speeds. In comparison, the industry-standard noise-reduced operation (NRO) baseline here generates an average power loss of 17%. When deployed in the higher-fidelity FASTFarm simulator, the policy largely retains its performance under steady and turbulent inflow, despite differences in simulation fidelity. These results demonstrate the potential of combining yaw control and power derating through RL to improve the power-noise trade-off in wind farm operation.
ER -