Journal cover Journal topic
Wind Energy Science The interactive open-access journal of the European Academy of Wind Energy
Journal topic

Journal metrics

CiteScore value: 0.6
CiteScore
0.6
h5-index value: 13
h5-index13
Preprints
https://doi.org/10.5194/wes-2020-61
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/wes-2020-61
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

  13 May 2020

13 May 2020

Review status
A revised version of this preprint is currently under review for the journal WES.

Utilizing Physics-Based Input Features within a Machine Learning Model to Predict Wind Speed Forecasting Error

Daniel Vassallo1, Raghavendra Krishnamurthy2,1, and Harindra J. S. Fernando1 Daniel Vassallo et al.
  • 1University of Notre Dame, Indiana, USA
  • 2Pacific Northwest National Laboratory, Washington, USA

Abstract. Machine learning is quickly becoming a commonly used technique for wind speed and power forecasting. Many of these methods utilize exogenous variables as input features, but there remains the question of which atmospheric variables provide the most predictive power, especially in handling non-linearities that lead to forecasting error. This investigation addresses this question via creation of a hybrid model that utilizes an autoregressive integrated moving average (ARIMA) model to make an initial wind speed forecast followed by a random forest model that attempts to predict the ARIMA forecasting error using knowledge of exogenous atmospheric variables. Variables conveying information about atmospheric stability and turbulence as well as inertial forcing are found to be useful in dealing with non-linear error prediction. Wind direction (θ) and temperature (T) are found to be the most beneficial individual input features. Streamwise wind speed (U), time of day (t), turbulence intensity (TI), turbulent heat flux (w'T'), θ, and T are found to be particularly useful when used in conjunction.The prediction accuracy of the ARIMA-RF hybrid is compared to that of the persistence and bias-corrected ARIMA models. The ARIMA-RF model is shown to improve upon these commonly employed modeling methods, reducing hourly forecasting error by approximately 30 % below that of the bias-corrected ARIMA model.

Daniel Vassallo et al.

Interactive discussion

Status: final response (author comments only)
Status: final response (author comments only)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

Daniel Vassallo et al.

Daniel Vassallo et al.

Viewed

Total article views: 506 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
349 118 39 506 40 39
  • HTML: 349
  • PDF: 118
  • XML: 39
  • Total: 506
  • BibTeX: 40
  • EndNote: 39
Views and downloads (calculated since 13 May 2020)
Cumulative views and downloads (calculated since 13 May 2020)

Viewed (geographical distribution)

Total article views: 383 (including HTML, PDF, and XML) Thereof 383 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Saved

No saved metrics found.

Discussed

No discussed metrics found.
Latest update: 23 Sep 2020
Publications Copernicus
Download
Short summary
Machine learning is quickly becoming a commonly used technique for wind speed and power forecasting, and is especially useful when combined with other forecasting techniques. This study utilizes a popular machine learning algorithm, an artificial neural network (ANN), in an attempt to predict the forecasting error of a statistical forecasting model. Various atmospheric characteristics are used as ANN inputs in an effort to discern the most useful atmospheric information for this purpose.
Machine learning is quickly becoming a commonly used technique for wind speed and power...
Citation