Ensemble-based, large-eddy reconstruction of wind turbine inflow in a near-stationary atmospheric boundary layer through generative artificial intelligence
This provides a more accurate method for validating turbine dynamics in field experiments, though it is incremental as it builds on existing atmospheric modeling with machine learning enhancements.
The paper tackles the problem of reconstructing time-resolved wind inflow for wind turbines using a large-eddy simulation-based technique with a diffusion model, achieving Pearson correlation coefficients of 0.20 to 0.85 for streamwise velocity in synthetic tests and 0.25 to 0.75 in real-world validations.
To validate the second-by-second dynamics of turbines in field experiments, it is necessary to accurately reconstruct the winds going into the turbine. Current time-resolved inflow reconstruction techniques estimate wind behavior in unobserved regions using relatively simple spectral-based models of the atmosphere. Here, we develop a technique for time-resolved inflow reconstruction that is rooted in a large-eddy simulation model of the atmosphere. Our "large-eddy reconstruction" technique blends observations and atmospheric model information through a diffusion model machine learning algorithm, allowing us to generate probabilistic ensembles of reconstructions for a single 10-min observational period. Our generated inflows can be used directly by aeroelastic codes or as inflow boundary conditions in a large-eddy simulation. We verify the second-by-second reconstruction capability of our technique in three synthetic field campaigns, finding positive Pearson correlation coefficient values (0.20>r>0.85) between ground-truth and reconstructed streamwise velocity, as well as smaller positive correlation coefficient values for unobserved fields (spanwise velocity, vertical velocity, and temperature). We validate our technique in three real-world case studies by driving large-eddy simulations with reconstructed inflows and comparing to independent inflow measurements. The reconstructions are visually similar to measurements, follow desired power spectra properties, and track second-by-second behavior (0.25 > r > 0.75).