Characteristics

• Semi submersible platform with 3 columns
• Three turbines
• The platform is self orientating towards wind
• Mooring lines connect to a detachable turret
• Cable for power transmission is guided through the turret to the seabed

Advantages

Construction

• flexible layout - scalable
• Easy fabrication - all construction is performed at yard, including installation of the turbines
• Easy installation - the floater is tugged to offshore destination, connected to the mooring lines and ready for operation (Plug and Play)

Energy Production/Operation

• Cost effective solution - three turbines on one platform
• Self orientating towards the wind - wind and wave response is independent of each other

Maintenance

• Easy access for inspection and maintenance as well as space to perform such tasks
• Easily disconnected from the turret and tugged to the yard for modification or more extensive maintenance

Dimensions

Height upwind turbines above sea level: 71 m
Height downwind turbines above sea level: 90 m
Distance between upwind turbines: 103 m
Turbine power: 3,6MW each, total 10,8 MW - rotor dia: 104 m
Vessel draft: 23m (operation) 7m (at yard)

Development

History

• 2006 - 2008: Concept development
• 2008 - 2009: Validation of concept: Optimization of the concept.

Extensive calculations and analysis performed by WindSea and external specialist companies (RISØ DTU etc.)

Test performed in water tank: The marine performance of the vessel, power production at different sea conditions as well as the ability to self orientating towards the wind.

Test performed in wind tunnel: Power production including the effect of wake and turbulence.

Test performed in water tank with wind exposure (combined water ytank / wind tunnel test)

Results

• The main principal of the WindSea floater has been validated
• The total energy production for a given configuration has been calculated
• The power production of the system has been calculated and verified in test to be 93% of the theoretical maximum, based on comparison to three stand alone turbines
• Wake effect: The results from the test and WindSea`s calculations are in accordance with third party calculations (RISØ)

Total energy production and wake losses

The results presented are taken from the study performed by Risø. Those results have been corroborated by analyses performed by the project and by the model tests. The Risø analysis covers the whole range of wind speeds (from 5 m/sec to 27 m/sec) while our own analysies and model tests are performed only for specific wind speeds.

The power production versus the main wind speed at 10 m above sea level is presented in figure 5.1. The figure presents two curves. The higher curve is for the downwind turbine standing alone, while the lower one is for the downwind turbine accounting for the wake effect (shadow) from the two upwind turbines.

The loss of power due to the wake effect is most significant for the low velocity in the range of 7 m/sec to 13 m/sec, while for high velocities the power production is almost unchanged.
It should be noted that the curves are dependent on the turbine properties and on the relative positions of the 3 turbines. By a suitable choice of turbines and positions the gap between the two curves may be reduced.

The total power produced by the turbine in a year is obtained by the combination of the power produced for each wind velocity and the probability of occurrence of this velocity.

By doing this operation for the two curves and taken the ratio of the two obtained numbers, the total power production reduction due to the wake may be calculated.

Risø has calculated this ratio for the given configuration to be 75% for the rear turbine. For the two turbines in front, there is no negative interaction between the two, hence this is calculated to 100 % for both.

Based on those production curves, the environmental data from Ekofisk and down time of 15% (this value is documented in the System Engineering activity) the power production for the 3 turbines is calculated to be in the range of 14,8 GWh/year for the 2 upwind turbines and 11,7 GWh/year for the downwind turbine.

This gives a total production of 41,4 GWh/year.

The production value is 93% of the maximal production of 3 like turbines (3 x 14,8 = 44,5 GWh/year).

The figure shows the distribution of the power production as a function of wind velocity. Most of the power is produced at velocities between 9 m/sec and 18m/sec.

System Engineering

The electrical scheme for the power export of the three turbines has been outlined. No difficulties are expected for the realisation of the system.

The electrical export cable is fed through the turret. The cable can not be twisted. To avoid this, the WindSea vessel has to be rotated back to its original orientation. A so called swivel is designed to be installed. A free rotating swivel includes a torque free electric connector that can be rotated 360 degrees and makes it possible to transfer the electricity from the platform to the export cable even when the platform is rotating

The project investigated the market availability of a swivel with the required capacity. A swivel with the required capacity and characteristics are commercially available and suitable for use.

It is also concluded that a main WindSea design principle (i.e. the ability to rotate against the wind) has proven to be technically feasible.

High focal point for development has been on the downwind turbine.

The use of three upwind turbines has been investigated. The rear tower in this configuration is erected to be in a vertical position and hence an upwind turbine can be utilized without the turbine blades coming in conflict with the turbine tower. The investigations performed concludes that there are no negative effects technically or for the calculated and tested power production, based on the same wind data etc.

The analyses performed by Risø simulates the air flow behind the two upwind turbines including the general meandering of the flow and the local turbulences created by the turbine blades. The forces on the downwind mill are calculated, i.e. forces on the blades, the rotor, and the tower.

The analyses show that the effect of the turbulences created by the wake results in a marginal increase of the forces and fatigue damage to the main components of the mill.
It was expected that the effect would be larger. Further investigation to explain this result will be done in the future.

However, the main conclusion of the Risø investigation is that the required strength for the main components of the downwind turbine can be achieved using standard available components.

An important issue is the reliability of the turbine. Offshore environment make intervention and maintenance more difficult, and the availability of the turbine may be adversely affected.

By using the Markov theory, simplified availability analyses was performed. Fault rate statistics of high quality are hard to obtain. The availability analyses is based on a German statistic with historical data from 1994 to 2004 for onshore wind turbines. Wind turbines have gone through a major development during those years.

If shutdown of one upwind turbine results in a shutdown of the other upwind turbine for stability reasons, the considering generator capacity on board Windsea has an overall availability of 85,2 %. In those analyses the contingency in both fault rate and the prospected repair time is significant due to lack of operation experience of offshore wind turbines mounted on floaters.

Due to the three wind turbine design, the platform control system has to integrate the control systems of each turbine. Interaction of the turbines like common starting, shutdown etc. must be a part of the control system.

The control system can also be extended to control the turbines to stabilize the direction of the platform to ensure that the upwind turbines are headed against the wind.

Further operational analysis is planned.

Structural Engineering

Forces due to wave, current and wind, including inertia forces as well as the self weight and buoyancy, have been analyzed by a computer model of the platform. A beam element model in Sesam was used.

The model is restrained at the attachment to the mooring line.

The stiffness analysis has been performed for 28 waves distributed into 9 directions. The waves represent the 100-year event and is the event expected to give the maximum internal forces in the structure. In a later stage more waves should be investigated to verify that the maximum loads for each location have been verified.

All calculations are done according the NORSOK code.

The stresses in all part of the structure are also computed and the dimensioning of each main part of the vessel was performed.

No fatigue analysis for the overall structure has been performed thus far. However the fatigue sensitive locations have been identified as shown in the figure 5.3.

Fatigue analysis of the towers due to wind and inertia forces has been done, since those areas are anticipated to be the most fatigue loaded. The towers are subjected to different loading conditions: the turbines may be activated or not. The fatigue damage created when the turbines are not in use has been calculated for 20 years life time.

A probabilistic analysis method has been used and the damage calculated to 0,079.
In its analysis, Risø has estimated the fatigue damage when the turbines are in use.

Risø pointed out that the proposed design of the tower has the drawback that it has an eigenfrequency in the range of the frequency activated by the rotation of the rotor. Risø has therefore proposed to reinforce the tower to avoid this situation. Based on this new design the fatigue damage was calculated to 0,57.

Based on those evaluations it is concluded that the fatigue is not expected to be a major issue in the design of the vessel.

Finally, the vessel has been checked for its tolerance of a boat impact. The standard criteria as described in NORSOK have been used.

As a result a total steel weight has been calculated.

Total steel weight is about 4600 t exclusive of the nacelles but inclusive the towers.

Total displacement of the vessel at operating draft: 9200 t.

Motion and stability analyses

This activity has focused on the definition of the overall dimension of the vessel and on the prediction of the motion when the WindSea vessel is subjected to wave and wind activity.

The stability criteria are in the DNV rules for mobile units. Its leads to the main dimension for the vessel: distance between columns 75 m, columns diameter 9 m.

The pontoon diameter is given by the requirement of floating to a draft of 6-8 m for facilitating the entry into harbors

The motion has been calculated with the computer program Sesam (Wadam).

Two alternatives have been studied: one with skirts attached at the lower end of each columns and one without skirt.

The function of those skirts is to increase the damping of the system and consequently to reduce the motion of the vessel.

The study shows that the skirts do reduce the motion in heave (vertical motion) but have no significant effect for the pitch motion (rotation around an axis perpendicular to the wave direction).

To determine the anchor forces, non- linear in time domain analyses have been performed using the software Simo.

The effect of the wind on the vessel motion was more difficult to assess. In this case, contrary to the common floating devices, the wind forces are significant compared to the wave forces: typically 150 t for the wind compared to a total of about 700 t.

The analysis of the wind action on the turbines shows that the frequencies are in range of 0,2 Hz and above. The frequency for the pitch and surge motion are in the range of 0.05 Hz and below. Due to this large separation in the frequencies it is concluded that the two effects may be analysed separately and summarized using the RMS method.

Stability
WindSea satisfies the DNV rules and regulations with respect to stability during in-place conditions. Maximum static heel angle is 4.3° and 9.5° during normal working condition and 35m/s wind speed storm respectively. WindSea also has acceptable stability characteristics when flooding of compartments occur during operation.

During transit, it is assumed that the waterline is intersecting the pontoon (no ballast). In this configuration WindSea has high stiffness and good stability, and therefore satisfies the DNV rules and regulations for both intact and damaged condition.

Motion
The motion analyses shows promising results for the extreme (100 year event) condition. The following maximum responses have been estimated:
Heave 10.3 m
Pitch 6.2 degrees
Roll 5.4 degrees

The above values relate to WindSea without skirt dampers. The following maximum response was found when skirts were mounted on the platform columns:
Heave 11.7 m
Pitch 11.8 degrees
Roll 11.6 degrees

These results indicate that skirts do not necessarily improve the motion characteristics of the platform. It was, however, found that the air gap was reduced by the inclusion of skirts.

For the operating condition with wind, the following maximum responses have been estimated (without skirts only):
Heave 3.2 m
Pitch 5.2 degrees
Roll marginal, head sea/wind only

An important design criterion for a semi-submersible unit is the split force acting on the column and pontoons. Investigations of split forces show that a maximum occurs for a wavelength of about 115-120m. This wave length is some 10 m less than one normally expects for conventional semi-submersables using approximately twice the breadth measured between the outer the pontoons. The reason is purely geometrical; WindSea has a triangular footprint as oppose to traditional rectangular hulls.

Mooring
The mooring system design is governed by the small water depth required which calls for a complex configuration to obtain required flexibility of the system. An iterative design approach was performed by using different configurations to obtain a layout that would not influence the wave frequency motion of the platform too severe. Another limiting criteria was the assumption that the fiber rope should not have bottom contact. However, if this assumption can be invalidated, using fibre only may be feasible and would be an improvement in the overall mooring system

The motion response of the platform is very moderate with the current mooring system. This is the case for both operational and storm conditions.
Analyses show a maximum offset of about 14.5 m and a mooring load of 830 tons.

Self Up-righting Capacity
This report shows preliminary calculations of the mechanical self up righting system of the WindSea platform. After having investigated various shapes of the aft tower, the drop shape has been deemed compatible compared to other shapes. The reason are given below.

Up righting characteristics: due to the geometry, the up righting characteristics are significantly better than the semi aerofoil and the ellipse shape.

Fabrication: the drop shape is easy to fabricate as plates can be connected to a circular tower giving the outside drop shape.

Mooring and Turret design

Mooring
The mooring and anchor system required for the entire platform can be based on proven equipment and available technology. Consequently, there is no major technology risk related to the mooring system.

Optimization of the system can only be made when the installation site has been defined. Water depth and bed soil conditions have to be known. As soon as the soil conditions are known, the anchor type can be defined. Cost optimization will include anchor and mooring lines as a whole.

The prototype mooring system is designed with a total of 6 mooring lines. The design includes 3 clusters separated 120° and 2 lines in each cluster separated by 10°. This design is chosen to meet the requirement of "one line rupture without drifting off" (ISO and DnV).

The Ekofisk area is used in the validation and cost calculation, which means a 70 meter water depth and soil conditions where suction anchors are suitable.

A wind farm installation has a significant cost savings opportunity by combining one anchor for as much as 3 mooring lines. The challenge is related to the anchor design. In a wind farm with a total of three mooring lines for each platform, only two lines will remain in operation if one mooring line ruptures. In this situation, the platform will drift off until the two mooring lines take the load. The load acting on the anchor will change direction by approx 120°.

Solution to this drifting problem are possible by a modification of the suction anchor or other anchor types that are available in the market (DPA etc).

The DPA anchor (Deep Penetration Anchor) is a low cost option and can be used in a wide range of soil conditions, but not for Ekofisk. DPA is a relatively new development, now under final certification and testing. DPA is the most promising design to take load in different directions and thereby suitable for a 3 mooring line solution.

The DPA anchor should be considered also for the prototype installation if the soil conditions are within the acceptable criteria.

Turret
The turret system is the mechanical and electrical interface between the entire platform and the mooring-/anchoring system. The turret system will give the platform full capability to rotate while connected to a geostationary mooring system. The turret is also the entry point for electrical high voltage power cable to the platform.

The turret will have functionality for the disconnection and reconnection of the platform from the mooring system.

The turret required for WindSea can be based on proven design principles. Vendors having years of relevant experience will be able to design the WindSea turret.

Consequently, there is not seen any major technology risk related to the turret system.

Fabrication

Fabrication of the WindSea platform can be performed in a number of ways. The overall philosophy is that the most cost effective fabrication method fulfilling the fabrication criteria should be chosen. The following gives backup for this information.

Hull including Wind Mills
Fabrication of the hull can be split and carried out at a number of yards, and the final assembly performed at a central place. This also includes systems to be included in the hull and wind mills.

It should be noted that the fully equipped platform will be able to float at a draft of 6 to 7 m having a GM in the order of 12 m. This allows for the possibility of both dock and inshore fabrication/assembly.

Tow-out of the platform to field will be done by tug boats. Further ballasting of the hull to achieve a greater water depth is foreseen.

Helideck
The helideck is considered to be part of the hull equipment.

Wind mills
As for the Helideck the windmill towers steel structures are considered to be part of the hull equipment. The windmill supplier is assumed to be involved in installation of the equipment in the towers.

Anchors
The positioning of the anchors is considered not to be critical as installation within a range of 0.5 to 1 m should be easy to achieve and this will be well within the acceptable range.

Turret
The turret can be installed immediately after installation of anchors and mooring system. The turret will be fabricated and machined as one unit and fully tested for easy connection/disconnection with the platform at the assembly yard.

Maintenance

The Windsea platform is designed to be disconnected from the mooring lines and towed to shore for major maintenance work. This will be performed on a bi-yearly basis or when major equipment requires service or repair.

The following inspection and maintenance areas are foreseen:

• Anchors,
• Anchor lines
• Turret
• Hull
• Wind Mills
• Helideck
• General vessel equipment

For each item the level and the frequency of required maintenance has to be identify.

All needs have to be summarized and coordinated such that the number of intervention on the vessel may be minimized.

It is possible to conceive of a number of plausible Operation & Maintenance (O&M) strategies for Windsea installations.

There are two different types of maintenance action:

Preventive Maintenance (PM) aims to reduce the occurrence of failures

Corrective Maintenance (CM) that involves action only after a failure has occurred.

Any approach to maintenance can employ, either or a combination of both of these actions.

Each attempts to balance capital costs, operational costs, and energy production in a different way. In considering the ideas, it is important to remember that our objective is to minimize the cost of the electricity produced by the offshore farm. This is not the same as maximizing the energy production, and indeed the most economic scheme may be one which sacrifices a little electricity for a great reduction in maintenance costs.

In practice, all wind turbine concepts are likely to have teething troubles at their introduction. For a period immediately after the construction of a wind farm, say 6 months, a special commissioning maintenance regime would have to be in operation until the teething troubles were ironed out. For the subsequent mainstream operation, the following maintenance strategies have been identified:

• The no-maintenance strategy
• The only-CM-maintenance strategy
• The opportunity-maintenance strategy
• The PM & CM maintenance strategy
• The light-PM and light-CM maintenance strategy
• The periodic check maintenance strategy

It also of first importance to choice the O&M strategy at the project start since this will govern the choice of the technology used.

Model tests

Model test have been performed both in wind tunnel and in wave basin.

In the framework of the validation phase, the tests were conducted to verify the main principles of the WindSea concept and to validate the calculation methods used during design.

The model tests do not cover all ranges of wind speed and sea states, but are performed for specific data.