TY - JOUR
T1 - A surrogate model of offshore wind farm support structures for wind farm design and financial valuation
AU - McWilliam, Michael K.
AU - Friis-Møller, Mikkel
AU - Pollini, Nicolo
AU - Dykes, Katherine
AU - Jensen, Morten
N1 - Publisher Copyright:
© Published under licence by IOP Publishing Ltd.
PY - 2022/6/2
Y1 - 2022/6/2
N2 - In the preliminary phases of offshore wind farm development, very little information on project design are available for supporting financial valuation and site design. In this work, we develop a surrogate model for offshore wind support structure mass for input to techno-economic analysis that is based on a small set of input parameters. Using reference turbines and a broad set of met-ocean conditions, a large design spaced is developed from which a sampling of conditions is used to estimate the dimensions and mass of monopile support structures. The results are parameterized using statistical methods to create a functional model of costs relative to high-level site and technical inputs. To preserve the transparency of the model input-output relationships, a statistical surrogate model is used based on quadratic functions of the inputs. Overall, the rated power and rotor diameter of the turbine has the greatest influence on the mass, followed by the specific power. The water depth was the next most important environmental parameter, followed by wave period. The full surrogate model captures 98.5% of the variance of the monopile mass as a function of the above inputs. We present results related to monopile foundations, but the methodology is flexible and can be applied also in the case of other types of support structures.
AB - In the preliminary phases of offshore wind farm development, very little information on project design are available for supporting financial valuation and site design. In this work, we develop a surrogate model for offshore wind support structure mass for input to techno-economic analysis that is based on a small set of input parameters. Using reference turbines and a broad set of met-ocean conditions, a large design spaced is developed from which a sampling of conditions is used to estimate the dimensions and mass of monopile support structures. The results are parameterized using statistical methods to create a functional model of costs relative to high-level site and technical inputs. To preserve the transparency of the model input-output relationships, a statistical surrogate model is used based on quadratic functions of the inputs. Overall, the rated power and rotor diameter of the turbine has the greatest influence on the mass, followed by the specific power. The water depth was the next most important environmental parameter, followed by wave period. The full surrogate model captures 98.5% of the variance of the monopile mass as a function of the above inputs. We present results related to monopile foundations, but the methodology is flexible and can be applied also in the case of other types of support structures.
UR - http://www.scopus.com/inward/record.url?scp=85131887632&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/2265/4/042048
DO - 10.1088/1742-6596/2265/4/042048
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.conferencearticle???
AN - SCOPUS:85131887632
SN - 1742-6588
VL - 2265
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - 4
M1 - 042048
T2 - 2022 Science of Making Torque from Wind, TORQUE 2022
Y2 - 1 June 2022 through 3 June 2022
ER -