Wind turbine Cpmax and drivetrain-losses estimation using Gaussian process machine learning

Research output: Contribution to journalConference Contribution

Abstract

In this paper it is shown that measured data in a wind turbine, available to the controller, can be formulated into a polynomial regression problem in order to estimate the turbine's maximum efficiency power coefficient, Cpmax, and drivetrain losses, assuming the latter can be well approximated as being linear. Gaussian process (GP) machine learning is used for the regression problem. These formulations are tested on data generated using the
Supergen Exemplar 5 MW wind turbine model, with results indicating that this is a potential low cost method for detecting changes in aerodynamic efficiency and drivetrain losses. The GP approach is benchmarked against standard least-squares (LS) regression, with the GP shown to be the superior method in this case.
LanguageEnglish
Article number032024
Number of pages8
JournalJournal of Physics: Conference Series
Volume1037
Early online date19 Jun 2018
DOIs
Publication statusE-pub ahead of print - 19 Jun 2018
EventThe Science of Making Torque from Wind 2018 - Politecnico di Milano, Milan, Italy
Duration: 20 Jun 201822 Jun 2018
http://www.torque2018.org/

Fingerprint

machine learning
wind turbines
regression analysis
power efficiency
turbines
aerodynamics
controllers
polynomials
formulations
coefficients
estimates

Keywords

  • polynomial regression problem
  • maximum efficiency power coefficient
  • Cp,max
  • Gaussian process (GP) machine learning

Cite this

@article{863ca78002d944468b76b12f91f474e3,
title = "Wind turbine Cpmax and drivetrain-losses estimation using Gaussian process machine learning",
abstract = "In this paper it is shown that measured data in a wind turbine, available to the controller, can be formulated into a polynomial regression problem in order to estimate the turbine's maximum efficiency power coefficient, Cpmax, and drivetrain losses, assuming the latter can be well approximated as being linear. Gaussian process (GP) machine learning is used for the regression problem. These formulations are tested on data generated using theSupergen Exemplar 5 MW wind turbine model, with results indicating that this is a potential low cost method for detecting changes in aerodynamic efficiency and drivetrain losses. The GP approach is benchmarked against standard least-squares (LS) regression, with the GP shown to be the superior method in this case.",
keywords = "polynomial regression problem, maximum efficiency power coefficient, Cp,max, Gaussian process (GP) machine learning",
author = "E Hart and Leithead, {W E} and J Feuchtwang",
year = "2018",
month = "6",
day = "19",
doi = "10.1088/1742-6596/1037/3/032024",
language = "English",
volume = "1037",
journal = "Journal of Physics: Conference Series",
issn = "1742-6588",

}

TY - JOUR

T1 - Wind turbine Cpmax and drivetrain-losses estimation using Gaussian process machine learning

AU - Hart, E

AU - Leithead, W E

AU - Feuchtwang, J

PY - 2018/6/19

Y1 - 2018/6/19

N2 - In this paper it is shown that measured data in a wind turbine, available to the controller, can be formulated into a polynomial regression problem in order to estimate the turbine's maximum efficiency power coefficient, Cpmax, and drivetrain losses, assuming the latter can be well approximated as being linear. Gaussian process (GP) machine learning is used for the regression problem. These formulations are tested on data generated using theSupergen Exemplar 5 MW wind turbine model, with results indicating that this is a potential low cost method for detecting changes in aerodynamic efficiency and drivetrain losses. The GP approach is benchmarked against standard least-squares (LS) regression, with the GP shown to be the superior method in this case.

AB - In this paper it is shown that measured data in a wind turbine, available to the controller, can be formulated into a polynomial regression problem in order to estimate the turbine's maximum efficiency power coefficient, Cpmax, and drivetrain losses, assuming the latter can be well approximated as being linear. Gaussian process (GP) machine learning is used for the regression problem. These formulations are tested on data generated using theSupergen Exemplar 5 MW wind turbine model, with results indicating that this is a potential low cost method for detecting changes in aerodynamic efficiency and drivetrain losses. The GP approach is benchmarked against standard least-squares (LS) regression, with the GP shown to be the superior method in this case.

KW - polynomial regression problem

KW - maximum efficiency power coefficient

KW - Cp,max

KW - Gaussian process (GP) machine learning

UR - http://iopscience.iop.org/journal/1742-6596

U2 - 10.1088/1742-6596/1037/3/032024

DO - 10.1088/1742-6596/1037/3/032024

M3 - Conference Contribution

VL - 1037

JO - Journal of Physics: Conference Series

T2 - Journal of Physics: Conference Series

JF - Journal of Physics: Conference Series

SN - 1742-6588

M1 - 032024

ER -