Piled foundations have been used throughout the world for supporting offshore oil and gas platforms and there exist well-established recommended practices and guidelines for the design of piles and grouted connections:
API RP2A, American Petroleum Institute, Recommended Practices for Planning Designing and Constructing Fixed Offshore Platforms
NORSOK N004 Design of Steel Structures.
Fixed offshore oil and gas platforms are generally supported by 3 or 4 legs with either a single pile driven through the leg or one or more skirt piles arranged around each leg, the piles connected to the leg by means of grouted sleeves. The piles are hollow steel tubulars ranging in diameter from 914mm to 2743mm.
In benign, shallow waters, a single pile has been used to support the topsides and as a conductor for drilling the well. In some cases, the conductor itself has been used to support the topsides. Conductors diameters are between 508mm and 914 and are normally either driven or drilled and cemented.
Nearshore marine construction of jetties and mooring dolphins has often used piles of greater diameter than those used offshore, but the depth of penetration and the means of installation have been different.
OWEC’s have been supported on single monopiles, effectively a downwards extension of the tower and generally using methods developed from marine construction. They have ranged in diameter from 2.1 m at Bockstigen (Gotland) to 3.7m at Lely and have been installed by driving or by drilling and cementing (rock socket).
Large diameter tubular piles are a well-established design as indicated above. However, unlike an oil platform, the foundation supporting an OWEC is subjected to a much larger proportion of live load compared to dead load. This means that the foundation experiences larger shears and bending moments and relatively small axial compression. The design of monopile foundations should consider cyclic loading of near-surface soils and the potential for loss of soil contact at the surface (post-holing). Rock-socketed piles are unlikely to be susceptible to this effect.
There are four main means of installing piles:
· Above-surface steam, hydraulic or vibration hammers
· Underwater hydraulic hammers
· Drill and grout
Pile driving is a faster and less weather sensitive means of installing piles than drilling and normally results in greater pile capacity than a drilled pile. There are however several disadvantages compared with drilling and grouting:
· The act of driving will sometimes damage the pile head and the pile may not be driven truly vertical. In order to connect the tower, this could entail cutting the head level and true and prepping it for either welding on of a flange or direct welding of the tower. This problem was overcome at Utgrunden by using a sleeve, incorporating the tower connection flange, that slid over the pile and could be adjusted to grade and level. Once in position, the annulus between sleeve and pile was grouted.
· During pile driving, accelerations both lateral and vertical of up to 50g will be observed. Any attachments to the pile will need to be designed for this or retrofitted. This would include access ladders and walkways, anodes, J-tubes etc.
Drill-drive would be slower than simply driving and would suffer all the disadvantages of driving. It is generally only used to assist driven piles in reaching target penetration in hard soils.
Drill and grout has been successfully used for some monopile foundations and is the only method if penetration of rock is required. The benefits of drill and grout are:
· More controlled placement of the pile without damage and to a tight tolerance is possible. This permits bolting on of the tower without top of pile preparation and eliminates the need to retrofit ladders, boat landings etc..
Gravity foundations or gravity base structures (GBS) have been used extensively in the Norwegian sector of the North Sea, mainly in deep water, for example Troll and Sleipner. The UK sector has also used gravity foundations in deep water, but more recently in shallower water: Ravenspurn and Harding.
GBS are generally buoyant for floatout, tow and installation and are then ballasted with water, iron ore or grout to provide sufficient on-bottom weight to resist overturning. The GBS normally consists of a series of open and or closed cells that form the base and one to four legs that are integral to the design, provide stability during temporary conditions and support the topsides.
To date gravity foundations for OWEC’s have been similar in appearance to onshore foundations with the connection to the tower raised above Highest Astronomic Tide. Examples are Middelgrunden, Vindeby and Tuno Knob
The gravity foundation has advantages for installation over a monopile in that the c omplete
OWEC can be assembled on-shore in a dry-dock as one unit and no drilling or piling equipment is necessary. However, the efficiency of the installation operation does depend on the dry-dock being located close to the OWEC’s site, thus minimising transport times. Additionally, a specially modified transportation/installation vessel is needed.
A variety of different configurations have been used to date and it is likely that optimisation for particular site-specific developments would result in more solutions. The likely future of gravity foundations as water depths increase are discussed below.
Solid concrete plate foundation – Middelgrunden, Vindeby
These are extensions of onshore foundations and are likely to increase significantly in weight as water depths increase, although the plate could be made to contain additional heavy ballast as an alternative to simply adding concrete mass.
Concrete box caisson (filled) – Tuno Knob
The caisson does not rely purely on the mass of concrete to provide stability and would probably not increase in mass quite so significantly as the solid plate.
Steel caisson – proposed
This would be similar in form to the plate foundation with provision for the heavy ballast.
The OWEC is dynamically sensitive to excitation caused by a complete rotation of the rotor and passage of the blades past the tower. This gives two periods that must be avoided to ensure that resonant response does not occur.
For example: for a three-bladed rotor with a rotation speed of 22 revs/minute the natural period T of the OWEC must be as given below.
· stiff-stiff natural period T < 0.8sec
· stiff-soft natural period 1.0sec < T < 2.4sec
· soft-soft natural period T > 3.0sec
It is normal to define the exclusion period as the calculated period +/- 10%
The natural period of the OWEC is critical as discussed above and depends on the following:
· Mass of the system
· Stiffness of the tower
· Stiffness of the combined substructure and foundation.
(Note: substructure is defined as the element between the tower and the seabed, foundation is defined as the element at seabed and below.)
The monopile is potentially the least stiff of the foundations options and, particularly in slightly deeper water, is likely to be of the soft-soft type. However, it was observed at Lely that the behaviour of two of the OWEC’s was stiffer than predicted, and that one was stiff-soft rather than soft-soft. It was fortunate that the exclusion period was avoided, although it must be noted that this was purely chance.
Multi-pile substructures are likely to have more predictable natural periods, being less dependent on the lateral stiffness of the surface and subsurface soils.
For any design, sensitivity studies must be undertaken to ensure that, even with upper and lower bound soil properties, the predicted range of OWEC natural periods does not fall within the exclusion period.
Scour of the seabed can also significantly affect the foundation stiffness. Scour protection will be necessary where granular surface soils exist in areas where the seabed can experience high currents or wave particle velocities.
Offshore structures generally have adequate fatigue resistance if their natural period is less than about 4 seconds. Above this level, design against fatigue is not impossible, but is more difficult.
Current demonstration OWEC projects: Middelgrund, Lely, Vindeby, Blyth are in very shallow and generally sheltered water (2m-10m) and the behaviour of the foundation is little influenced by wave dynamics.
In deeper water, and particularly with monopiles and monotowers, it is likely that the natural period of the OWEC will be greater than 3 seconds, a soft-soft foundation, and will be more susceptible to wave-induced fatigue damage. Aerodynamic damping is a result of rotor rotation and affects fore-aft first order motions. This will reduce the observed fatigue damage due to waves compared to that predicted using a theoretical undamped system.
Up to 20m water depth, it is likely that the drilled and grouted monopile will be the most cost-effective solution, with the concrete plate foundation as an alternative.
Above 20m, it is likely that the natural period of an OWEC on a monopile will exceed 4 seconds, with potential problems for fatigue resistance, although aerodynamic damping would help to reduce the dynamic response.
A concrete gravity structure is theoretically suitable for depths greater than 20m although the weight and cost of such a structure could be prohibitive. It could be designed either to be self-floating or barge transportable. The former would require the structure to be constructed in a dry dock, although it is noted that the Middelgrunden structures were constructed in a dry dock and were not self-floating.
Steel structures would be suitable for these depths and would probably not be excessively heavy. It is likely that they would be supported by small (36-48in) piles rather than gravity or suction foundations, although a heavily ballasted steel caisson may be cost-effective. Such structures could either be of lattice tower or monotower construction. A lattice tower would probably be lighter than a monotower, but because of the large number of members and joints, would be more expensive to fabricate and would require significantly more inspection and maintenance, particularly in the splash zone. The lattice tower is likely to have a higher natural period than a monotower, and could therefore be more fatigue-susceptible.
A monotower is a large diameter central tube supported by three or four small diameter piles. The piles are connected to the tube by means of grouted sleeves and tubular braces. The benefit of the monotower is its simple construction, but it would still have a higher cost per tonne compared with a monopile. The turbine tower would be bolted to the monotower, just as for a monopile, thus the operational experience at Lely, Vindeby and Blyth regarding O&M, access, control rooms, workrooms would be transferable. Separate provision would be necessary if a lattice tower were to be used.
An alternative monotower concept is to use a large diameter tube with pile sleeves attached closely to the tube with shear plates – similar to a large offshore platform ‘leg bottle’. It is anticipated that three 36in-48in piles would be suitable for this purpose, and they could be driven, speeding up the installation process. The cost per tonne would be between a monopile and a braced monotower. Pile weight would be lower than the monopile so overall cost should be less.
The optimum concept for a particular site should be assessed by detailed analyses of all concepts and their site-specific costs:
· CAPEX:- engineering, fabrication and installation.
· OPEX:- inspection, maintenance, repair, visit intervals, support and/or accommodation vessel/unit requirements.
Sea ice is a consideration in the Baltic but not in the UK or Dutch sectors of the North Sea. However, since the sea ice is annual ice up to about 600mm thick, structures can be designed to resist it by providing sloping faces to the substructure at sea level. This reduces the ice pressure by inducing bending in the ice and breaking sheets into small pieces.
At Bockstigen, the monopiles have an octagonal form of ice protection made of stainless steel and filled with concrete.
Foundations could be designed using conservative assumptions of the effects of breaking waves compared with non-breaking waves and this would probably not be a significant cost item for a 1 or 2 OWEC development.
However, the economics of large OWECS rely on economy of scale and optimisation of all aspects of design to remain economically attractive. Better understanding of breaking wave phenomena for generic and site-specific wave environments is therefore necessary.
Breaking waves can cause both local damage to offshore structures and impose significant global forces. A single column structure such as a monopile or even a monotower is more susceptible to global forces compared with a multiple legged jacket structure because the wave force is applied instantaneously to a single discrete element rather than to an array of elements. A phenomenon known as ‘ringing’; a dynamic response to the high frequency components of a wave train, has been observed on a single column concrete gravity structure in the Norwegian sector(Sleipner). It has been suggested that a similar phenomenon can be observed with breaking waves acting on a monopile in shallow water.(Structural Dynamics of Offshore Wind Turbines subject to Extreme Wave Loading – N Rogers – Border Wind)
At the EPSRC OWEN workshop ‘Structure and Foundations Design of Offshore Wind Installations March 2000, NDP Barltrop discussed breaking waves and their effect on shallow structures. The effects of breaking waves upon the Bockstigen monopile structure are investigated in this study.
It should be noted that the occurence of breaking waves is not applicable for existing Dutch offshore windfarms as they are located in inland water.
Because the behaviour of waves in shallow water is so dependent on local topology it may be difficult to predict whether waves would tend to break. There may well be local knowledge, existing model test information from coastal defence programmes or measurements that would indicate whether breaking waves had been observed.
Model testing would be a useful means of investigating the behaviour of waves at a particular site and with representative models of an OWECS give information on wave run-up, celerity, particle velocities and steepness. Current and wind can significantly alter the steepness of waves in shallow water, and should be considered in any testing programme.
Direct research into breaking waves in relation to offshore wind energy is currently being undertaken under the Engineering and Physical Sciences Research Council (EPSRC) Renewable and New Energy Technologies (RNET) ‘Dynamic Response of Wind Turbine Structures in Waves’ NDP Barltrop University of Glasgow et al.
At the Bockstigen demonstration project the monopile and tower are strain gauged and measurement of the dynamic behaviour the OWEC and metocean and meteorological measurements are underway.
Garrad Hassan are further developing Bladed for Windows and Germanischer Lloyd have undertaken development under Joule 1 (Jour 0072) Study of Offshore Wind Energy in the EC
The OWEN / ESPRC Workshop April 1999 identified research priorities in this area as:
A need to improve the prediction of environmental conditions for input to the design calculations, including:
· The relationship between extreme winds and waves.
· Improvement in metocean predictions for sites of interest
· Improved models of boundary layer, turbulence and machine wakes in maritime areas
· Predictions of wind and wave directions
· The determination of loading due to breaking waves and other shallow water effects
A decision as to whether components (namely turbine and support structure) are treated in an integrated way during design, reducing conservatism.
To develop improved understanding of the structural dynamics of offshore wind structures
To assess the reliability of existing spectral wave models
To assess importance of wave-driven fatigue on offshore wind structures
To investigate the suitability of different types of foundations for offshore wind energy applications, for example, their response under cyclic loads and their dynamic characteristics.
To routinely monitor the performance of offshore anemometry masts and wind turbine structures – with the data used to refine models and designs
To assess the available methods of determining and measuring dynamic soil properties
To investigate the economics of off-the-shelf foundation designs
Webdesign and hosting Starforze
Webcontent © 2002 www.offshorewindenergy.org
Updated September 2008