Noise Steering
A reinforcement learning-based wind farm coordinated control strategy that jointly adjusts yaw angles and power setpoints to manage both wake effects and noise directivity, maximizing total wind farm power while satisfying noise constraints.
Timeline (last 90 days)
The Noise Steering study proposes jointly controlling yaw angle and power setpoints via reinforcement learning, which can substantially reduce power loss while satisfying noise constraints.
In experiments on the six-turbine test wind farm, the Noise Steering strategy satisfied the 40 dB noise constraint across the full range of turbulent wind speeds.
Relative to unconstrained greedy control, the Noise Steering strategy incurs an average power loss of only 2%.
Under high wind speed conditions, the power loss of the Noise Steering strategy is kept within 7.7%.
When the strategy was deployed to the higher-fidelity FAST.Farm simulator, it largely retained its original performance under both steady and turbulent inflow conditions.
Current validation is still based on a small-scale six-turbine test wind farm; the gap between the low-fidelity acoustic model and real sound propagation, as well as the complexity of wake coupling in larger-scale wind farms, are questions that need to be answered in follow-up work.
All Facts (6)
The Noise Steering study proposes jointly controlling yaw angle and power setpoints via reinforcement learning, which can substantially reduce power loss while satisfying noise constraints.
50%UnverifiedIn experiments on the six-turbine test wind farm, the Noise Steering strategy satisfied the 40 dB noise constraint across the full range of turbulent wind speeds.
50%UnverifiedRelative to unconstrained greedy control, the Noise Steering strategy incurs an average power loss of only 2%.
50%UnverifiedUnder high wind speed conditions, the power loss of the Noise Steering strategy is kept within 7.7%.
50%UnverifiedWhen the strategy was deployed to the higher-fidelity FAST.Farm simulator, it largely retained its original performance under both steady and turbulent inflow conditions.
50%UnverifiedCurrent validation is still based on a small-scale six-turbine test wind farm; the gap between the low-fidelity acoustic model and real sound propagation, as well as the complexity of wake coupling in larger-scale wind farms, are questions that need to be answered in follow-up work.
50%