The Future of Floating Wind Turbines

A Review of the Status and Risks of Floating Wind Turbine Technology

The floating wind turbine platforms (FWTP) industry is at an early and exciting stage. There are a number of emerging designs and a mix of platform and wind turbine types. It is also encouraging to see high levels of collaboration, with many companies working alongside bodies such as the Energy Technologies Institute, National Renewable Energy Laboratory, NOWITECH, Risø and the University of Maine. This article is a summary of a major report conducted by the author, which provides an up-to-date review of the FWTP industry and its principle technical risks.

By Charles Briggs, Renewable Energy Consultant, SgurrEnergy, UK

{access view=!registered}Only logged in users can view the full text of the article.{/access}{access view=registered}There are a number of potential markets for FWTP technology – various companies have targeted the Mediterranean coast of France, Iberia, Italy, Norway, Scotland, Japan, Korea and the USA. Of these nations, the US State of Maine is the only place with a stated installation target: 5GW of FWTP by 2030. The State of Maine estimates that the FWTP offers up to US$ 20 billion of investment, and will create between 7,000 and 15,000 jobs.

While a move to FWTP may sound improbable, there is the precedent of the oil and gas industries’ move to deeper waters, albeit that the rationale was different. There are a number of reasons why FWTP may be preferable to nearshore, bottom-founded turbines:

• Winds are typically stronger and more consistent further offshore.
• Greater siting flexibility since depth is no longer a primary issue.
• Reduced visual impact from shore (particularly if over the horizon).
• Integrated hull and tower, eliminating the need for project-specific transition pieces.
• Mass production leading to lower installation costs, with the full assembly of the platform and turbines both onshore and inshore.
• Simplified offshore installation process.
• Potentially lower environmental impacts on wildlife and their habitats, for example sea mammals for which pile driving may damage hearing.

The long-term survivability of floating platforms has already been proven in the oil and gas industries – the first permanent floating platform was installed in the Argyll field in the North Sea in 1975. FWTP differ in one significant respect however; they combine large aerodynamic forces (from the turbine) with the hydrodynamic loading from wind and waves. These aerodynamic and hydrodynamic forces can combine to produce overturning moments which will be relatively larger than those on oil and gas platforms. It remains to be seen if FWTP can be sufficiently stable such that the turbine performance is not significantly impaired in comparison with bottom-founded turbines.

It holds, generally speaking, that as distance from shore increases, so availability decreases due to access issues, and operations and maintenance (O&M) costs will increase. This will be a key issue for floating wind farms. In 2002 a paper by Bulder et al. calculated that availability of a FWTP would be approximately 91%; this does not compare well with the onshore wind best practice figure of 95%.

There are three generic types of floating platform in common use in the offshore oil and gas sectors and these are being adapted for the FWTP market:

• Spar buoys. The Statoil Hywind design fits in this category, and was the first commercial-scale FWTP deployed, being installed in 2009 off Norway using a Siemens 2.3MW turbine.
• Tension Leg Platforms (TLP). Blue H and Glosten Associates are developing TLP designs, with Blue H having already deployed a 60kW prototype in waters off Italy back in 2007. Blue H is also noteworthy in that it is one of only three principle commercial designers developing their own wind turbine. The Blue H and Glosten Associates designs are shown in Figure 1.
• Semi-submersibles. This is the most common platform type under development for FWTP, with all designs using a triangular platform. The notable designs thus far are from Nenuphar (who are exceptional in proposing to use a vertical axis wind turbine, VAWT), Principle Power (WindFloat) and WindSea. The WindFloat design is unusual as it employs a fail-safe active trim optimisation system which keeps the mean tower position vertical by counteracting wind-induced heel. The WindSea platform is the only design proposing to mount more than one wind turbine on the platform, with a design for three turbines mounted on all corners of the platform. These three designs are shown in Figure 2.

The final platform which should be noted at this stage is the SWAY concept, seen in Figure 3 (alongside the Hywind). SWAY claim that this design reduces the fatigue load by 50% compared with an upwind spar, enabling a 5MW turbine to be installed on a lighter tower. This is made possible by placing the yaw at the bottom of the spar, allowing the whole tower to turn together with the turbine. This ensures that the side prone to the most fatigue, with wire stays, is kept upwind at all times.

In reality the optimum FWTP does not exist; the platform type must be selected according to the conditions prevalent at the site in question. However, the following conclusions can be made:

• Spars: In many respects the simplest of the design types, spars experience low heave movements but pitch and roll still exist. The required water depth for these designs limits the number of suitable sites, and requires sheltered waters for assembly purposes. Thus it is likely that the market for spars will be geographically specific; it is notable that the Hywind and SWAY concepts are designed by Norwegian firms. Fjords make for excellent sheltered water assembly sites. Spars also have large footprints, although a single tension leg anchor, such as proposed by SWAY, has the smallest footprints of all FWTP.
• TLP: These are probably the most stable of the platform designs, at least the passive designs. However, installation issues will be significant where used with vertical load anchors; ballasted gravity anchors however simplify such issues. TLP also appear to be the most sensitive to water depth, bottom conditions and scalability issues, but are likely to be smaller than semi-submersibles and have a lower footprint than both spars and semi-submersibles. TLP are more vulnerable to areas of large tides and storm surges than other platforms.
• Semi-submersibles: The design type most familiar to shipbuilders, semi-submersibles will be the easiest platforms to transport, deploy, maintain and decommission, but the most sensitive to wave loading. They will also be the platforms suitable to the largest range of depths but, like spars, have large footprints. The large amount of structure at the surface will also require significant anti-corrosion measures.

The author’s main report highlighted four priority areas for all FWTP:

• FWTP engineering standards and design codes: While IEC 61400-1 and IEC 61400-3 standards, in conjunction with local standards such as Norsok in Europe and the American Petroleum Institute (API RP-2A) in the USA, provide an established approach which can be extended to the particular technical challenges of FWTP, the writing of FWTP-specific standards must be considered a high priority. Det Norske Veritas (DNV) recently announced that it has started work on such standards.
• Quality assurance within certified shipyards: The proximity of suitable shipbuilders, oil and gas platform fabricators and offshore wind foundation fabricators that have the necessary welding experience to ensure quality control will be important. Thus any country that intends to deploy FWTP must work with the shipbuilding industry.
• Reduction in top weight: Another priority is the need to reduce the weight of the FWTP above the waterline. The moment of inertia, and hence pitch motion, is highly sensitive to weight distribution.
• Low cost anchors and moorings: The use of synthetic, most likely polyester, mooring lines rather than steel will make deployment simpler and the platforms cheaper to install, since they are lighter.

Weight reductions may manifest themselves in the turbine tower and nacelle through the use of composite materials. However, cost will be the determining issue; optimising the FWTP design to minimise weight may in turn complicate manufacturing and increase capital and maintenance costs.

It is possible to reduce the weight of the FWTP by relaxing the constraints on blade noise relative to onshore designs, enabling the use of more flexible and lighter rotors, leading to higher tip-speeds, lower input torque, lower gear ratios, smaller gearboxes and therefore a lighter nacelle. There are three principle design means to make lighter turbines:

• Downwind rotors: At present only one FWTP company is publicly active with a downwind design (SWAY), although it is developing this in conjunction with Areva Wind.
• Two-bladed rotors: Only one FWTP company (Blue H) is publicly active with a two-bladed design, and no offshore wind turbine companies are publicly considering two-bladed designs.
• VAWT: It remains to be seen if VAWT can make a breakthrough on FWTP having failed to establish a presence in the existing megawatt-scale market.

It is likely that developers will wish to minimise supply chain complexity and are therefore unlikely to embrace wide variations in turbine configuration (e.g. vertical and horizontal) and platform types. Another point of interest raised by some manufacturers was the possibility that the platform might be used for more than 20 years, the accepted benchmark for the lifespan of an offshore wind farm.

The review showed that fatigue rates will be an issue; the principle question is not one of extreme weather event survivability but rather whether the wind farms will last 20 years. The various opinions gathered through this review also showed that the economics of bringing platforms back to harbour for refit are uncertain. Sites with low significant wave heights will be best; hence one might speculate that the Mediterranean and Great Lakes are the most obvious markets to develop first.

The question of whether FWTP will be deployed in large numbers is dependent on a range of issues, but the primary one is economics. Platforms can be built to be very stable but at high costs. Likewise turbines can be combined with sophisticated motion control systems, but this pushes up the price and risk.

Platform design will require very detailed modelling software and engineering to ensure that the required factors of safety are met while making the platform as economic as possible. In this sense every aspect of the FWTP will need to be optimised, from anchor systems through to nacelle materials. Unnecessary conservatism cannot be afforded, and specific FWTP design standards must take account of this.

Acknowledgements
The following are thanked for their assistance:

• Heriot-Watt University, National Renewable Energy Laboratory (NREL), Norwegian University of Science and Technology (NTNU), Risø DTU and the University of Maine.
• ABS Consulting, Experia Consulting, Glosten Associates, Nenuphar, Principle Power, SgurrEnergy, SWAY, Statoil and WindSea.

Biography of the Author
This work was completed in conjunction with SgurrEnergy while Charles Briggs was studying for an MSc in Renewable Energy Engineering at Heriot-Watt University, 2009/2010. He was awarded the MSc with distinction on 2 October 2010, and is now working full time for SgurrEnergy.{/access}