Improved linearised models of wind turbine aerodynamics and control system dynamics using harmonic linearisation

Richard C. Lupton, Robin S. Langley

Research output: Contribution to journalArticle

Abstract

Where non-linearities are not too strong, linearised frequency-domain approaches offer fast calculations, which can be valuable for preliminary design of wind turbine blades, foundations and floating platforms. But the aerodynamic and control system behaviour of a wind turbine is noticeably non-linear. Here we show for the first time that the technique of harmonic linearisation can reduce error in the approximation of aerodynamic and control system non-linearities, compared to the more common tangent linearisation. After deriving the linearised models, comparing linearised results to non-linear simulations for the NREL 5 MW turbine shows that: (1) harmonic linearisation captures aero-elastic effects and non-linearity in aerodynamic forces, giving a 2–4x reduction in error compared to the tangent linearisation; (2) harmonic linearisation can capture non-linear wake dynamics; and (3) the torque and pitch controller behaviour can be approximated with good results away from the rated wind speed but with some challenges when the two controllers interact. Further improvements in the linearised model of the control system have been identified. By improving the accuracy of linearised models, harmonic linearisation is a promising means to extend the applicability of frequency-domain approaches for initial design and optimisation of wind turbines.
LanguageEnglish
Pages148-162
Number of pages5
JournalRenewable Energy
Volume135
Early online date23 Nov 2018
DOIs
StatusE-pub ahead of print - 23 Nov 2018

Cite this

Improved linearised models of wind turbine aerodynamics and control system dynamics using harmonic linearisation. / Lupton, Richard C.; Langley, Robin S.

In: Renewable Energy, Vol. 135, 01.05.2019, p. 148-162.

Research output: Contribution to journalArticle

@article{a8c6de842ab1482eb85af0a21bb109b7,
title = "Improved linearised models of wind turbine aerodynamics and control system dynamics using harmonic linearisation",
abstract = "Where non-linearities are not too strong, linearised frequency-domain approaches offer fast calculations, which can be valuable for preliminary design of wind turbine blades, foundations and floating platforms. But the aerodynamic and control system behaviour of a wind turbine is noticeably non-linear. Here we show for the first time that the technique of harmonic linearisation can reduce error in the approximation of aerodynamic and control system non-linearities, compared to the more common tangent linearisation. After deriving the linearised models, comparing linearised results to non-linear simulations for the NREL 5 MW turbine shows that: (1) harmonic linearisation captures aero-elastic effects and non-linearity in aerodynamic forces, giving a 2–4x reduction in error compared to the tangent linearisation; (2) harmonic linearisation can capture non-linear wake dynamics; and (3) the torque and pitch controller behaviour can be approximated with good results away from the rated wind speed but with some challenges when the two controllers interact. Further improvements in the linearised model of the control system have been identified. By improving the accuracy of linearised models, harmonic linearisation is a promising means to extend the applicability of frequency-domain approaches for initial design and optimisation of wind turbines.",
author = "Lupton, {Richard C.} and Langley, {Robin S.}",
year = "2018",
month = "11",
day = "23",
doi = "10.1016/j.renene.2018.11.067",
language = "English",
volume = "135",
pages = "148--162",
journal = "Renewable Energy",
issn = "0960-1481",
publisher = "Elsevier",

}

TY - JOUR

T1 - Improved linearised models of wind turbine aerodynamics and control system dynamics using harmonic linearisation

AU - Lupton, Richard C.

AU - Langley, Robin S.

PY - 2018/11/23

Y1 - 2018/11/23

N2 - Where non-linearities are not too strong, linearised frequency-domain approaches offer fast calculations, which can be valuable for preliminary design of wind turbine blades, foundations and floating platforms. But the aerodynamic and control system behaviour of a wind turbine is noticeably non-linear. Here we show for the first time that the technique of harmonic linearisation can reduce error in the approximation of aerodynamic and control system non-linearities, compared to the more common tangent linearisation. After deriving the linearised models, comparing linearised results to non-linear simulations for the NREL 5 MW turbine shows that: (1) harmonic linearisation captures aero-elastic effects and non-linearity in aerodynamic forces, giving a 2–4x reduction in error compared to the tangent linearisation; (2) harmonic linearisation can capture non-linear wake dynamics; and (3) the torque and pitch controller behaviour can be approximated with good results away from the rated wind speed but with some challenges when the two controllers interact. Further improvements in the linearised model of the control system have been identified. By improving the accuracy of linearised models, harmonic linearisation is a promising means to extend the applicability of frequency-domain approaches for initial design and optimisation of wind turbines.

AB - Where non-linearities are not too strong, linearised frequency-domain approaches offer fast calculations, which can be valuable for preliminary design of wind turbine blades, foundations and floating platforms. But the aerodynamic and control system behaviour of a wind turbine is noticeably non-linear. Here we show for the first time that the technique of harmonic linearisation can reduce error in the approximation of aerodynamic and control system non-linearities, compared to the more common tangent linearisation. After deriving the linearised models, comparing linearised results to non-linear simulations for the NREL 5 MW turbine shows that: (1) harmonic linearisation captures aero-elastic effects and non-linearity in aerodynamic forces, giving a 2–4x reduction in error compared to the tangent linearisation; (2) harmonic linearisation can capture non-linear wake dynamics; and (3) the torque and pitch controller behaviour can be approximated with good results away from the rated wind speed but with some challenges when the two controllers interact. Further improvements in the linearised model of the control system have been identified. By improving the accuracy of linearised models, harmonic linearisation is a promising means to extend the applicability of frequency-domain approaches for initial design and optimisation of wind turbines.

U2 - 10.1016/j.renene.2018.11.067

DO - 10.1016/j.renene.2018.11.067

M3 - Article

VL - 135

SP - 148

EP - 162

JO - Renewable Energy

T2 - Renewable Energy

JF - Renewable Energy

SN - 0960-1481

ER -