Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Michele Martini (1) F P Raúl Guanche (1) José A. Armesto (1) Iñigo J. Losada (1)
(1) Environmental Hydraulics Institute of Cantabria, Santander, Spain
Printer friendly version: Print
Download poster(1.20 MB)
Presenter's biographyBiographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited
Mr. Martini has been working in the offshore renewable energy sector in the last years. He is currently Ph.D. student at the Environmental Hydraulics Institute of Cantabria, and researcher in the OceaNET project (Marie Curie Actions). He studied Energy Engineering at La Sapienza in Rome and Wind Energy at the Technical University of Denmark in Copenhagen. He then worked two years in the wave energy field at "Politecnico di Torino" and moved to Spain to increase his knowledge in O&M of floating systems. His research is focused in evaluating the influence of met-ocean conditions over the performance of such installations.
Long-term floating wind turbine performance under met-ocean conditions influence
In the last decade the interest of industries towards floating wind energy has rapidly increased, leading to the deployment of the first full scale prototypes in real seas. Results have been encouraging and the first floating wind farms are nowadays being planned. Due to their exposure, the performance of such systems may be heavily influenced by met-ocean conditions. Since there is only poor experience in running these installations, their long term dynamic behaviour still represents a partially unknown issue. In this investigation, a methodology for evaluating the performance of a floating wind turbine across its entire life span is proposed.
The approach chosen for this work aims at providing a valuable estimation of long term floating platform performance parameters, still keeping a good computational efficiency. This goal is pursued combining the use of metereological reanalysis databases with multivariate climate data selection techniques, with a time-domain floating wind turbine model and with scattered data interpolation methods. A case study is selected: the structure analysed is the OC4 semisubmersible platform, mounting the NREL 5 MW wind turbine, and the site chosen is located off the coast of Porto, Portugal.
Main body of abstract
For this location, met-ocean data for a period of twenty years are generated by means of meteorological reanalysis techniques. A subset of a thousand hourly conditions is selected by means of a maximum dissimilarity algorithm, which allows to span the climate variability. Wind and wave time series are then generated through Kaimal and Jonswap spectrums, respectively. Then, the wind turbine behaviour is simulated in the time domain for selected each met-ocean condition. The numerical model includes linear hydrodynamic theory, a quasi-static mooring model and a simplified quasi-static thrust aerodynamic model with instantaneous relative wind speed calculation. The platform motions and the electrical power are hence statistically analysed. The results are finally interpolated for all the life span by means of radial basis functions, which are appropriate when working with scattered data. This work aims at predicting the twenty years wind turbine capacity factor of the wind turbine, under the influence of met-ocean conditions. When any of its operating parameters exceeds, or is expected to exceed, a certain safety tolerance, the machine needs to be shut down. The occurrence of this events lowers the energy yield of the system. The effect of imposing different operational thresholds to the wind turbine tower inclination and hub acceleration is evaluated. Results reveal that the capacity factor decreases non-linearly, as the safety tolerances are decreased, and strongly depend on the statistical quantity (e.g. % quantile) chosen to evaluate the operating variables.
The developed tool could be used at different engineering stages. At the design stage, it can give important statistical parameters directly related, for example, to fatigue life of structural components of the wind turbine. At the operational stage, it can help in planning preventive shutdown of the machine, based on weather forecasts, in order to reduce the risk of damage or failure of any electrical or mechanical component of the system. The illustrated procedure may also help in investigating a possible economical trade-off between lower failure probability (by imposing lower operating thresholds) and lower power production.
The proposed methodology may represent a powerful integrated tool for estimating with accuracy fatigue loads of floating wind turbines components, for selecting appropriate operation strategies and for calculating the influence of met-ocean conditions over the long-term energy production of such systems.