The importance of round-robin validation when assessing machine-learning-based vertical extrapolation of wind speeds

Bodini, Nicola; Optis, Mike

doi:https://doi.org/10.5194/wes-5-489-2020

Articles | Volume 5, issue 2

https://doi.org/10.5194/wes-5-489-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/wes-5-489-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 5, issue 2

Research article

|

17 Apr 2020

Research article |

| 17 Apr 2020

The importance of round-robin validation when assessing machine-learning-based vertical extrapolation of wind speeds

Nicola Bodini and Mike Optis

Supplement

https://doi.org/10.5194/wes-5-489-2020-supplement

Data sets

Atmospheric Radiation Measurement (ARM) user facility. Doppler Lidar (DLAUX) R. Newsom and R. Krishnamurthy https://doi.org/10.5439/1374838

Atmospheric Radiation Measurement (ARM) user facility. Eddy Correlation Flux Measurement System (30ECOR) R. Sullivan and E. Keeler https://doi.org/10.5439/1025039

Atmospheric Radiation Measurement (ARM) user facility. Carbon Dioxide Flux Measurement Systems (30CO2FLX4M) S. Biraud, D. Billesbach, and S. Chan https://doi.org/10.5439/1025036

Download

Article (5123 KB)
Full-text XML

Short summary

An accurate assessment of the wind resource at hub height is necessary for an efficient and bankable wind farm project. Conventional techniques for wind speed vertical extrapolation include a power law and a logarithmic law. Here, we propose a round-robin validation to assess the benefits that a machine-learning-based approach can provide in vertically extrapolating wind speed at a location different from the training site – the most practically useful application for the wind energy industry.