Numerical evaluation of the smearing effect in a spinner-mounted lidar for wind field reconstruction

This study assesses the impact of the smearing effect on full-rotor longitudinal wind speed reconstruction for a spinner-mounted lidar, applying two reconstruction methods: Inverse Distance Weighting (IDW) interpolation and a Least-Squares fit with Proper Orthogonal Decomposition (POD-LSQ). The smearing effect arises as the spinnermounted lidar beam, constrained to the hub, rotates with the rotor, causing the beam to sample across an arc during the sampling period. This results in a spatial and temporal integration of wind speeds, referred to in this study as the smearing effect. We evaluate a numerical single-beam pulsed lidar mounted on the hub of the DTU 10MW reference wind turbine using the HAWC2 v13.1 lidar sensor to generate synthetic lidar measurements. The lidar data is sampled from a turbulence field within an atmospheric boundary layer simulated using Large Eddy Simulation (LES), featuring a mean wind speed of 16.17 m/s and a turbulence intensity of 8%, over a 10-minute period. Reconstructed fields are analyzed with and without smearing effects across different sampling times. Contrary to our initial hypothesis that smearing would degrade reconstruction accuracy, results show that it improves accuracy by mitigating errors introduced by cross-contamination effects. Smearing reduces the Mean Absolute Error (MAE) in reconstructed wind fields, particularly for longer sampling times, with reductions of up to 2.9% for IDW and 8.8% for POD-LSQ for a sampling time of 0.5 s. Additionally, smearing improves Rotor Average Wind Speed (RAWS) estimation, reducing RAWS MAE by 25% for POD-LSQ—from 0.049 to 0.037 m/s. These findings suggest that, rather than degrading lidar-based wind field reconstruction, smearing serves as a beneficial averaging mechanism that enhances accuracy of the evaluated reconstruction techniques by mitigating errors associated with the assumption that lateral and vertical wind components are negligible in line-of-sight velocity when estimating the longitudinal wind speed for the reconstruction.