skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Decreasing wind speed extrapolation error via domain-specific feature extraction and selection

Abstract

Model uncertainty is a significant challenge in the wind energy industry and can lead to mischaracterization of millions of dollars' worth of wind resources. Machine learning methods, notably deep artificial neural networks (ANNs), are capable of modeling turbulent and chaotic systems and offer a promising tool to produce high-accuracy wind speed forecasts and extrapolations. This paper uses data collected by profiling Doppler lidars over three field campaigns to investigate the efficacy of using ANNs for wind speed vertical extrapolation in a variety of terrains, and it quantifies the role of domain knowledge in ANN extrapolation accuracy. A series of 11 meteorological parameters (features) are used as ANN inputs, and the resulting output accuracy is compared with that of both standard log-law and power-law extrapolations. It is found that extracted nondimensional inputs, namely turbulence intensity, current wind speed, and previous wind speed, are the features that most reliably improve the ANN's accuracy, providing up to a 65 % and 52 % increase in extrapolation accuracy over log-law and power-law predictions, respectively. The volume of input data is also deemed important for achieving robust results. One test case is analyzed in depth using dimensional and nondimensional features, showing that the feature nondimensionalization drastically improvesmore » network accuracy and robustness for sparsely sampled atmospheric cases.« less

Authors:
 [1]; ORCiD logo [2];  [1]
  1. Univ. of Notre Dame, IN (United States)
  2. Univ. of Notre Dame, IN (United States); Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE; US Office of Naval Research (ONR); National Science Foundation (NSF)
OSTI Identifier:
1668333
Report Number(s):
PNNL-SA-150381
Journal ID: ISSN 2366-7451
Grant/Contract Number:  
AC05-76RL01830; 1565535; WFIFP2-SUB-001; N00014-17-1-3195
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Wind Energy Science (Online)
Additional Journal Information:
Journal Volume: 5; Journal Issue: 3; Journal ID: ISSN 2366-7451
Publisher:
European Wind Energy Association - Copernicus
Country of Publication:
United States
Language:
English
Subject:
Doppler lidar, Machince learning, Extrapolation Method

Citation Formats

Vassallo, Daniel, Krishnamurthy, Raghavendra, and Fernando, Harindra S. Decreasing wind speed extrapolation error via domain-specific feature extraction and selection. United States: N. p., 2020. Web. doi:10.5194/wes-5-959-2020.
Vassallo, Daniel, Krishnamurthy, Raghavendra, & Fernando, Harindra S. Decreasing wind speed extrapolation error via domain-specific feature extraction and selection. United States. doi:10.5194/wes-5-959-2020.
Vassallo, Daniel, Krishnamurthy, Raghavendra, and Fernando, Harindra S. Sun . "Decreasing wind speed extrapolation error via domain-specific feature extraction and selection". United States. doi:10.5194/wes-5-959-2020. https://www.osti.gov/servlets/purl/1668333.
@article{osti_1668333,
title = {Decreasing wind speed extrapolation error via domain-specific feature extraction and selection},
author = {Vassallo, Daniel and Krishnamurthy, Raghavendra and Fernando, Harindra S.},
abstractNote = {Model uncertainty is a significant challenge in the wind energy industry and can lead to mischaracterization of millions of dollars' worth of wind resources. Machine learning methods, notably deep artificial neural networks (ANNs), are capable of modeling turbulent and chaotic systems and offer a promising tool to produce high-accuracy wind speed forecasts and extrapolations. This paper uses data collected by profiling Doppler lidars over three field campaigns to investigate the efficacy of using ANNs for wind speed vertical extrapolation in a variety of terrains, and it quantifies the role of domain knowledge in ANN extrapolation accuracy. A series of 11 meteorological parameters (features) are used as ANN inputs, and the resulting output accuracy is compared with that of both standard log-law and power-law extrapolations. It is found that extracted nondimensional inputs, namely turbulence intensity, current wind speed, and previous wind speed, are the features that most reliably improve the ANN's accuracy, providing up to a 65 % and 52 % increase in extrapolation accuracy over log-law and power-law predictions, respectively. The volume of input data is also deemed important for achieving robust results. One test case is analyzed in depth using dimensional and nondimensional features, showing that the feature nondimensionalization drastically improves network accuracy and robustness for sparsely sampled atmospheric cases.},
doi = {10.5194/wes-5-959-2020},
journal = {Wind Energy Science (Online)},
issn = {2366-7451},
number = 3,
volume = 5,
place = {United States},
year = {2020},
month = {7}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:

Works referenced in this record:

The Perdigão: Peering into Microscale Details of Mountain Winds
journal, May 2019

  • Fernando, H. J. S.; Mann, J.; Palma, J. M. L. M.
  • Bulletin of the American Meteorological Society, Vol. 100, Issue 5
  • DOI: 10.1175/BAMS-D-17-0227.1

Spatial and temporal variability of turbulence dissipation rate in complex terrain
journal, January 2019

  • Bodini, Nicola; Lundquist, Julie K.; Krishnamurthy, Raghavendra
  • Atmospheric Chemistry and Physics, Vol. 19, Issue 7
  • DOI: 10.5194/acp-19-4367-2019

Spatial Variability of Winds and HRRR–NCEP Model Error Statistics at Three Doppler-Lidar Sites in the Wind-Energy Generation Region of the Columbia River Basin
journal, August 2019

  • Pichugina, Y. L.; Banta, R. M.; Bonin, T.
  • Journal of Applied Meteorology and Climatology, Vol. 58, Issue 8
  • DOI: 10.1175/JAMC-D-18-0244.1

Scaling, Anisotropy, and Complexity in Near-Surface Atmospheric Turbulence
journal, February 2019

  • Stiperski, Ivana; Calaf, Marc; Rotach, Mathias W.
  • Journal of Geophysical Research: Atmospheres, Vol. 124, Issue 3
  • DOI: 10.1029/2018JD029383

Atmospheric stability affects wind turbine power collection
journal, January 2012


Mesoscale model evaluation with coherent Doppler lidar for wind farm assessment
journal, June 2013


Airborne Observations of a Catalina Eddy
journal, September 2013

  • Parish, Thomas R.; Rahn, David A.; Leon, Dave
  • Monthly Weather Review, Vol. 141, Issue 10
  • DOI: 10.1175/MWR-D-13-00029.1

Forecasting wind with neural networks
journal, January 2003


Scalar-Flux Similarity in the Layer Near the Surface Over Mountainous Terrain
journal, June 2018

  • Sfyri, Eleni; Rotach, Mathias W.; Stiperski, Ivana
  • Boundary-Layer Meteorology, Vol. 169, Issue 1
  • DOI: 10.1007/s10546-018-0365-3

The Second Wind Forecast Improvement Project (WFIP2): Observational Field Campaign
journal, September 2019

  • Wilczak, James M.; Stoelinga, Mark; Berg, Larry K.
  • Bulletin of the American Meteorological Society, Vol. 100, Issue 9
  • DOI: 10.1175/BAMS-D-18-0035.1

Wind Speed Extrapolation Using Machine Learning Methods and LiDAR Measurements
journal, January 2018


Observations of offshore wind characteristics by Doppler-LiDAR for wind energy applications
journal, May 2016


Multi-step wind speed prediction based on turbulence intensity and hybrid deep neural networks
journal, April 2019


The internal boundary layer ? A review
journal, March 1990


Nonlinear model predictive control of wind turbines using LIDAR: Nonlinear model predictive control of wind turbines using LIDAR
journal, August 2012

  • Schlipf, David; Schlipf, Dominik Johannes; Kühn, Martin
  • Wind Energy, Vol. 16, Issue 7
  • DOI: 10.1002/we.1533

Field-test results using a nacelle-mounted lidar for improving wind turbine power capture by reducing yaw misalignment
journal, June 2014


Making the Black Box More Transparent: Understanding the Physical Implications of Machine Learning
journal, November 2019

  • McGovern, Amy; Lagerquist, Ryan; John Gagne, David
  • Bulletin of the American Meteorological Society, Vol. 100, Issue 11
  • DOI: 10.1175/BAMS-D-18-0195.1

Applying Neural Network Models to Prediction and Data Analysis in Meteorology and Oceanography
journal, September 1998


Vertical extrapolation of wind speed using artificial neural network hybrid system
journal, June 2016

  • Islam, Md. Saiful; Mohandes, Mohamed; Rehman, Shafiqur
  • Neural Computing and Applications, Vol. 28, Issue 8
  • DOI: 10.1007/s00521-016-2373-x

The MATERHORN: Unraveling the Intricacies of Mountain Weather
journal, November 2015

  • Fernando, H. J. S.; Pardyjak, E. R.; Di Sabatino, S.
  • Bulletin of the American Meteorological Society, Vol. 96, Issue 11
  • DOI: 10.1175/BAMS-D-13-00131.1

Impact of model improvements on 80 m wind speeds during the second Wind Forecast Improvement Project (WFIP2)
journal, January 2019

  • Bianco, Laura; Djalalova, Irina V.; Wilczak, James M.
  • Geoscientific Model Development, Vol. 12, Issue 11
  • DOI: 10.5194/gmd-12-4803-2019

CASPER: Coupled Air–Sea Processes and Electromagnetic Ducting Research
journal, July 2018

  • Wang, Qing; Alappattu, Denny P.; Billingsley, Stephanie
  • Bulletin of the American Meteorological Society, Vol. 99, Issue 7
  • DOI: 10.1175/BAMS-D-16-0046.1

Initial results from a field campaign of wake steering applied at a commercial wind farm – Part 1
journal, January 2019

  • Fleming, Paul; King, Jennifer; Dykes, Katherine
  • Wind Energy Science, Vol. 4, Issue 2
  • DOI: 10.5194/wes-4-273-2019

A comparison of ground-based LiDAR and met mast wind measurements for wind resource assessment over various terrain conditions
journal, November 2016

  • Kim, Daeyoung; Kim, Taewan; Oh, Gwanjun
  • Journal of Wind Engineering and Industrial Aerodynamics, Vol. 158
  • DOI: 10.1016/j.jweia.2016.09.011

Assessing State-of-the-Art Capabilities for Probing the Atmospheric Boundary Layer: The XPIA Field Campaign
journal, February 2017

  • Lundquist, Julie K.; Wilczak, James M.; Ashton, Ryan
  • Bulletin of the American Meteorological Society, Vol. 98, Issue 2
  • DOI: 10.1175/BAMS-D-15-00151.1

Improving Wind Energy Forecasting through Numerical Weather Prediction Model Development
journal, November 2019

  • Olson, Joseph B.; Kenyon, Jaymes S.; Djalalova, Irina
  • Bulletin of the American Meteorological Society, Vol. 100, Issue 11
  • DOI: 10.1175/BAMS-D-18-0040.1

Lidar uncertainty and beam averaging correction
journal, January 2015

  • Giyanani, A.; Bierbooms, W.; van Bussel, G.
  • Advances in Science and Research, Vol. 12, Issue 1
  • DOI: 10.5194/asr-12-85-2015

Sub-hourly forecasting of wind speed and wind energy
journal, January 2020


Validating precision estimates in horizontal wind measurements from a Doppler lidar
journal, January 2017

  • Newsom, Rob K.; Brewer, W. Alan; Wilczak, James M.
  • Atmospheric Measurement Techniques, Vol. 10, Issue 3
  • DOI: 10.5194/amt-10-1229-2017

The Nature, Theory, and Modeling of Atmospheric Planetary Boundary Layers
journal, February 2011

  • Baklanov, Alexander A.; Grisogono, Branko; Bornstein, Robert
  • Bulletin of the American Meteorological Society, Vol. 92, Issue 2
  • DOI: 10.1175/2010BAMS2797.1

Application of artificial neural networks for the wind speed prediction of target station using reference stations data
journal, November 2007


Quantifying error of lidar and sodar Doppler beam swinging measurements of wind turbine wakes using computational fluid dynamics
journal, January 2015

  • Lundquist, J. K.; Churchfield, M. J.; Lee, S.
  • Atmospheric Measurement Techniques, Vol. 8, Issue 2
  • DOI: 10.5194/amt-8-907-2015