LiDAR measurements and machine learning make wakes in wind energy science
LiDAR measurements and machine learning make wakes in wind energy science lead image
When wind turbines generate power, they create an area with lower wind velocity and higher turbulence trailing behind their structure. In an array, this wake can impact the downstream turbines and cause them to produce significantly less power. The fluctuations from the wake also result in more maintenance and shorter turbine lifetimes.
Numerical simulations have explored wind turbine wakes but are limited because they idealize the complicated behavior of nature. Iungo et al. obtained LiDAR measurements of the wind field around turbines in North Texas and analyzed the data with machine learning algorithms to obtain a real-world picture of the wakes.
The team used a laser-based instrument to measure wind speed upstream and downstream of the turbine. The data was collected under different atmospheric conditions and other uncontrollable turbine parameters, such as rotor velocity and pitch angle, making it difficult to calculate flow statistics. Instead, machine learning techniques clustered the measurements to identify the most representative realizations of wind turbine wakes.
“The novelty of this paper, besides the experiments, is really how this data is post-processed and analyzed,” said author Giacomo Valerio Iungo. “It’s really a synergy of the engineering and the mathematical algorithms.”
Not only do these results inform big picture models and array design and deployment, but they can also be used to correct individual turbines. The group found systematic yaw misalignment that would have been unidentifiable with previous data analysis techniques.
The scientists are expanding the approach to monitor a large number of turbines in Oklahoma for the AWAKEN experiment. They are also beginning to investigate offshore turbines, which will present new challenges because of their accessibility and size.
Source: “Machine-learning identification of the variability of mean velocity and turbulence intensity for wakes generated by onshore wind turbines: Cluster analysis of wind LiDAR measurements,” by Giacomo Valerio Iungo, Romit Maulik, S. Ashwin Renganathan, and Stefano Letizia, Journal of Renewable and Sustainable Energy (2022). The article can be accessed at https://doi.org/10.1063/5.0070094 .
This paper is part of the Preparatory Work for the American Wake Experiment (AWAKEN) Collection, learn more here .
