Abstract
An analysis of different floating platform concepts in site conditions typical of the Mediterranean Sea is presented. A tension leg platform, a spar buoy, and a barge support substructures were investigated. In all the three cases, the National Renewable Energy Laboratory 5-MW machine was considered as wind turbine. This turbine model was also used to carry out a performance comparison between the selected offshore site and an adjacent onshore site, in order to estimate the advantages of the offshore solution in terms of annual energy production. The comparison among the different floating platform concepts considers the loads on the turbine induced by different wind and wave conditions. The simulations were performed using the fully coupled time domain aero-hydro-servo-elastic simulation tool FAST, made available by National Renewable Energy Laboratory. The wind and waves data used in the simulation, provided by the UTMEA-CLIM laboratory of ENEA, are typical for Mediterranean Sea conditions. The simulation results are reported and discussed. They represent a first contribution in helping to resolve basic design trade-offs among different floating platform concepts candidate for offshore wind energy deployment in Mediterranean Sea.
Introduction
The selection of offshore sites located at a proper distance from the shore mitigates the visual impact of the wind farms thus improving their social acceptance. In the North and Baltic Seas, this requirement can be satisfied in many cases using fixed-bottom support structures for the turbines because water depth is shallow also far from the shore, up to several tens of kilometers, the average distance and water depth being 43 km and 27 m, respectively (source: www.ewea.org/statistics/offshore/). Almost all the wind farms currently installed in the world use this type of foundation. Conversely, deep waters (even in the proximity of the shore) characterize the Mediterranean Sea. This implies the need to use floating support structures for the turbines. Floating wind technologies are in an early stage of development, and only few prototypes were installed around the world. Currently there are no offshore wind farms in the Mediterranean Sea.
As a consequence, the analysis of different floating technologies in typical Mediterranean Sea conditions takes on a great interest to evaluate the floating offshore wind energy potential.
Here, the issue of floating offshore wind turbines installed in site conditions typical of the Mediterranean Sea is addressed. Specifically, a load analysis for different floating platform types was performed using wind and waves data of a typical Mediterranean site. A state-of-art, public domain simulation code (FAST) 1 was used to calculate wind turbine loads. The same reference wind turbine, National Renewable Energy Laboratory (NREL) 5 MW (Jonkman et al., 2009) was used for all the different floating platforms. In the following section, the floating platforms in Mediterranean Sea were modeled, and then the result of the performed simulations were analyzed and discussed. Conclusions will close this work.
Floating platforms in typical Mediterranean Sea conditions: analysis of different concepts
Investigated floating concepts
As told before, three different floating platforms were considered, which represent the three primary floating platform classes: a tension leg platform (TLP), a spar buoy (Hywind concept), and a barge.
To perform a comparison of these three different concepts in the typical Mediterranean Sea conditions, results by the extended modeling work performed at NREL (Jonkman and Matha, 2011) were used. The NREL 5-MW system was modeled by the Massachusetts Institute of Technology/National Renewable Energy Laboratory (MIT/NREL) TLP, the OC3-Hywind spar buoy, and the ITI Energy barge, representing the three primary floating platform classes (Figure 1). All of these floating platforms were developed specifically to support the rotor, nacelle, and tower of the NREL baseline 5-MW system.
Platform schemes: left—tension leg, center—spar buoy, and right—barge.
MIT/NREL TLP
The MIT/NREL TLP is a joint design by MIT and NREL (Matha, 2009). This platform uses four legs with two taut mooring lines on each leg to provide a restoring force. The central spar part of the TLP is weighted on the bottom, which adds stiffness and also makes it possible for the platform to be towed without the lines or turbine attached.
OC3-Hywind spar buoy
A Norway-based company, Statoil Hydro, developed a spar buoy design that is currently supporting a Siemens 2.3-MW turbine in a floating demonstration project, the Hywind project. NREL modified this design to be compatible with the NREL 5-MW turbine, and the result is called the OC3-Hywind spar buoy (Jonkman, 2010). The spar buoy uses a heavy counterbalance at the base of the spar to move the center of mass below the center of buoyancy. This creates a restoring moment if the spar is pitched or rolled.
ITI Energy barge
The ITI Energy barge is a floating platform designed by the Department of Naval Architecture, Ocean & Marine Engineering at the University of Strathclyde, Glasgow, Scotland, UK, through a contract with ITI Energy (Sebastian and Lackner, 2012). The barge has eight catenary mooring lines, two coming off each corner. The mooring lines are added not only to tether the barge in place but also to provide some stiffness.
Site location and metocean data
The western Sardinia coast and the Sicily channel are found to be among the most productive areas in the whole Mediterranean. Then, the locations in the south-western coast of Sardinia (onshore) and further away from that in the Mediterranean (offshore) were selected as the reference sites for which to obtain environmental (metocean) data for the load analyses (CapoSperone area in the S.Antioco Island, see Figure 2). Since one is dealing with three different support platforms and each is designed for a specific depth of water, three different offshore sites had to be chosen, but all of them have the same latitude. The UTMEA-CLIM laboratory of ENEA provided the required wind and wave data.
Platform locations.
Results and discussion
Platform stability
A comparison between the natural frequencies of the platforms and the cut-in to cut-off wave frequency range of the incident waves contributes to identify potential sources of instability of the floating system. In order to do such a comparison, the first step is to find the natural frequencies. The natural frequencies can be obtained using free decay tests in surge, heave, pitch, and yaw. To carry out these decay tests, one runs FAST with no wind, no waves, and no currents. The platform was just displaced from its initial location. Thus, it is possible to determine the period (and frequency) of the oscillations resulting from this displacement, before the platform stabilized at its initial location.
This analysis is important in order to determine if resonance is susceptible to occur in a given site, and this is verified here. To determine which platform is the best for our site, one started comparing the natural frequencies of the selected installation with the frequency of the incoming waves (see Table 1). The incident waves in the selected location have the frequency range between 0.11 and 0.25 Hz, so the heave natural frequency of ITI Barge and the yaw natural frequency of OC3 spar buoy are within this range. In this study, the platforms were used as they were without any modification and any tuning or tailoring action aimed to customize the platforms to the specific site characteristics. These actions could be an outcome of the analysis that is finalized to detect the presence of problems like this (which can be solved introducing proper modifications in the platform configuration), but this exceeds the scope of this work.
Natural frequencies of the selected platforms.
TLP: tension leg platform.
Power curve and energy production
An estimate of the difference in the annual energy production between onshore and offshore installations and other relevant production indices (capacity factor, number of equivalent hours) is relevant in this onshore/offshore analysis. To perform this task, the estimate of the total annual energy production of 5-MW NREL wind turbine mounted on TLP support platform and onshore equivalent wind turbine is evaluated.
In order to get the annual energy produced for each case, we had to find the share of every wind speed over a year according to its probability of occurrence (365 days × 24 h × frequency of the wind speed), see Figure 3, and then multiplying this value to the corresponding generated power. As a result, it can be seen that the 5-MW wind turbine mounted on TLP platform produces 18.64 GWh/year, and the equivalent land-based wind turbine produces 13.94 GWh/year which means that the offshore wind turbine produces 34% more than onshore version per year. As a consequence, it is possible to evaluate the capacity factor and the number of equivalent hours for TLP and onshore configurations. The capacity factor would be the amount of energy produced by the wind turbine divided by the maximum energy it could produce with working constantly with its nominal power over the whole year. For our TLP, we get the value of 42%, while for onshore configuration, it is about 32%. The corresponding values for the number of equivalent hours (number of hours at nominal power which need to produce the same annual energy) are 3728 h for the land-based wind turbine and 2788 h for the offshore–TLP platform. These results are summarized in Table 2.
Annual energy productivity TLP/onshore.
Comparisons among onshore and offshore 5-MW turbine performance.
TLP: tension leg platform.
Load analysis
Load analysis requires verifying the structural integrity of a wind turbine by running a series of design load cases (DLCs) to determine the loads expected over the lifetime of the machine. The loads are examined within the primary members of the wind turbine, including the blades, drivetrain, nacelle, and tower, and for the floating system, the mooring lines.
Each International Electrotechnical Commission (IEC) design standard prescribes numerous DLCs. For these load analyses, it is considered not necessary to run all the DLCs prescribed by the design standards, and just one subset has been chosen, according to the preliminary level of this study. We consider a normal turbulence model (NTM) consisting of full-field three-component stochastic winds with a turbulence standard deviation given by the 90% quantile. This is based on the wind turbine turbulence category (B in this project) and the normal sea state (NSS). NSS models the irregular sea state as a summation of sinusoidal wave components whose amplitude is determined by the wave spectrum, each parallel (long-crested) and described by Airy wave theory. The sea state is derived from the JONSWAP spectrum, whose formulation is based on the given values of the significant wave height and peak spectral period.
The normal wind profile that is used in wind models should consist of a vertical power-law shear exponent of 0.2 for land-based wind turbines according to the IEC 61400–1 design standard and a value of 0.14 for sea-based turbines according to the IEC 61400–3 design standard. To facilitate the response comparisons, a value of 0.14 for both cases was selected.
Load analyses for each of the four system models were run according to the specifications, data, and procedures described above. The load analyses helped to perform a preliminary identification of problems with all system configurations, including instabilities and the susceptibility of excessive platform pitching motions incompatible with wind turbine operations. The main results obtained in the analysis are summarized below (see Figure 4).
Ratio between offshore and onshore configurations: 1, torque at blade root; 2, torque of lowest pitch shaft; 3, yaw force; 4, yaw torque; 5, torque at the tower root; and 6, generated power.
The barge will be more affected by the waves than by the wind because it will be more subjected to wave loading, and because of its relatively light weight, therefore, creates larger system responses, but since in our location the sea state is relatively calm, consequently, the ITI Energy barge system shows the least increase in loads, but still has the greatest motions in surge, sway, roll, and yaw. The excessive rolling motions of the barge bring about load excursions more extreme farther down the load path—from the blade, through the drivetrain and nacelle, to the tower.
The MIT/NREL TLP system has much less platform motion than the other two concepts, particularly in pitch and roll, due to the limited platform motions that do remain.
The OC3-Hywind spar system has the greatest pitch motion among all three concepts. This yields generally greater shaft and tower loads in the OC3-Hywind system than in the MIT/NREL TLP system, except for loads preliminary affected by platform yaw.
It can be noticed that there is a great increase in the power generated in the offshore case (most of which due to the different in wind speeds of offshore with respect to onshore case) but also there is a great increase in the loads, especially on the yaw bearing.
Conclusion
The results characterize the dynamic responses of the three primary floating wind turbine concepts, represented here by the MIT/NREL TLP, the OC3-Hywind spar buoy, and the ITI Energy barge system, together with the NREL 5-MW baseline wind turbine. The impacts brought about by the dynamic coupling between the turbine and each floating platform are presented, and comparisons between the concepts are quantified. In summary, almost all of the floating wind turbines show increased loads on turbine components as compared to the land-based system.
One conclusion is that among these three offshore support platforms, the TLP is more compatible with our site conditions because in other two platforms, there is risk of occurrence of resonance unless tuning or tailoring actions are implemented to adapt the structures to the selected site characteristics.
The second conclusion is that although we have a significant increase in the power generated in offshore wind turbine compared with onshore wind turbine, there is also a corresponding increase in the applied loads, especially in the yaw bearing. To reach a technically feasible concept, modifications to the design are required. Some forms of design modifications are possible. First, the turbine, especially the tower, could be strengthened to enable it to withstand the increased loading. However, this solution may not be cost-effective. Second, solution to improve the response of the floating system is to add design features that will increase damping to stabilize the pitch motion. Such solution acts as a wind turbine control system, which relies on the conventional wind turbine actuation of blade pitch, generator torque, and nacelle yaw to dampen the excessive pitch motion. A simpler solution for improving the pitch damping may be to introduce passive damping devices into the underlying design. Similarly, a tuned mass damper (TMD) could be placed at the top of the wind turbine tower; when tuned at the natural period of the rigid-body turbine plus platform pitch mode, such a system could dampen pitching (and rolling) motion. It may also be possible to dampen the platform rotational motions with the equivalent of passive anti-roll stabilizers installed within or on top of the platform. The platform’s hydrodynamic radiation damping and viscous drag could also be increased through the incorporation of damping orifices in the platform or horizontal damping plates and/or bilge keels positioned below the free surface.
Platform design can be also improved by adding degrees of freedom (DOFs) in or between the floating platform and the wind turbine to eliminate the direct coupling between the platform motions and turbine motions. For example, articulated joints may be installed in the floating platform, as in the Versabuoy offshore system, or between the wind turbine’s tower and nacelle, as in the Wind Eagle turbine (Kelley et al., 2005).
Footnotes
Acknowledgements
The authors thank Dr Adriana Carillo and Dr Gianmaria Sannino, researchers at the UTMEA-CLIM laboratory of ENEA, for assistance with metocean data for the Mediterranean Sea that generally improved the manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.