Seven years ago, there was just one industrial-scale floating wind turbine in existence, Statoil's Hywind demonstrator off the Norwegian coast, using a 2.3MW Siemens machine.
Now, pioneering units totalling 15MW are installed off Norway, Portugal and Japan, and their performance is changing perceptions, convincing the broader industry that floating wind has a viable economic future.
Several pre-commercial arrays are close to realisation, while the first commercial facilities should be operating by 2025. Technological and economic challenges remain, but there is no doubt the sector is gaining momentum.
This was evident at a recent international conference on floating offshore wind turbines in Marseille, France. Floating offshore wind is "essential" to fill the gap between the fixed-foundation offshore pipeline and Europe's 2030 targets, Giles Dickson, CEO of WindEurope (formerly EWEA), declared.
The drivers behind floating wind energy are well-rehearsed: increasingly constrained onshore markets; limited shallow-water sites suitable for fixed-foundation turbines; and a huge, otherwise inaccessible resource, often close to major demand centres.
In response, more than 30 concepts are under development, mainly featuring spar, semi-submersible or tension-leg platforms (TLPs), alongside variants such as barges, hybrid wind-and-wave devices and multi-turbine platforms.
While it will depend on the market, most observers believe that fixed-foundation turbines will dominate in waters less than 40 metres deep and floating platforms in depths of over 60 metres, with a mix of solutions for the range between.
The prime technologies
Spar-buoys, which Statoil chose for its Hywind demonstrator, are inherently stable and relatively simple structures consisting of a cylindrical steel floater held upright by the weight of ballast and secured by catenary mooring lines.
The main disadvantage is the draft of at least 80 metres, which means the spar has to be upended and ballasted, and the turbine installed - using expensive heavy-lift vessels - in a sheltered deep-water area, preferably close to port.
While it can then be towed to its final location using conventional tugs, and towed back if needed, heavy-lift vessels are again required for major maintenance.
Semi-submersible platforms are a popular choice among early demonstrators, largely because they offer more flexibility in shallower waters and are easier to install.
With their low draft, they can be assembled and commissioned at the quayside using a large crane, towed out by standard tugs and back to port for major repairs.
The most notable downside is the need for a large and heavy structure to maintain stability, especially as turbine size increases. Principle Power's WindFloat platform, supporting a 2MW Vestas turbine, has been operating successfully off Portugal since 2011.
TLPs are submerged structures anchored by tensioned vertical tethers. The shallow draft permits quayside assembly, commissioning and repairs, while the limited movement allows a smaller, lighter structure than semi-submersibles.
However, the high-vertical-load anchors and tensioned tethering system add extra costs and complexity when installing and disconnecting. T
he first TLP supporting a wind turbine is scheduled to be installed by Gicon in the German Baltic Sea towards the end of the year.
A question of cost
As with all new technologies, development costs are higher than would be expected in a mature market, and everything now depends on bringing them down.
Recent studies by the UK-based Carbon Trust and research body Energy Technologies Institute (ETI) indicate that leading technologies deployed at commercial scale could achieve a levelised cost of energy (LCoE) of £85-95/MWh (EUR106-118) by 2025, with further reductions possible over time.
This would enable floating wind to compete with fixed-foundation turbines by the end of the 2020s, many believe.
"We need to show that floating wind can be competitive in the medium term, before 2030," Pierre Parvex, head of the offshore commission of French wind energy association FEE, said in Marseille. A LCoE of £85/MWh is "the minimum to stay in the race," he said.
There is plenty of scope to cut costs, initially by reducing the size, weight and complexity of the floating foundation while maximising its stability, and in simplifying the assembly and installation process. Anchors, mooring systems, electrical infrastructure, command-control systems and wake effects all need to be better understood and optimised.
Turbine size and availability will play an important role. With offshore work typically costing around ten times of that onshore, operations and maintenance is also key. Minor work will be carried out in-situ, but turbines will be towed into port (or sheltered waters for spars) for major repairs.
The cost-benefit of this will partly depend on distance from port, but developers need to "demonstrate that power cables and mooring systems can be connected and disconnected relatively quickly", says Charlie Nordstrom, a floating foundation expert at consultancy BVG Associates. (Nordstrom has since left BVG.) The potential for "plug-and-play" systems "will be a game-changer", he believes.
With each technology having its own advantages and disadvantages - and being largely untested - it is impossible to predict which will achieve the lowest cost of energy.
The TLP has "tremendous potential if they can solve the demanding mooring requirements", argues Johan Sandberg, segment leader for floating wind turbines at DNV GL, while Michael Guldbrandtsen, senior consultant at Make Consulting, believes that semi-submersibles have "a huge advantage" if designers can achieve a lightweight structure.
In reality, we will need a mix. "I am pretty convinced there is room for multiple solutions," says Nordstrom. "Market and site conditions will drive different answers in different regions."
Also crucial in cutting costs is political commitment, backed up by ambitious targets to drive commercial-scale deployment. "We need very large volumes and visibility," Dickson argues. Costs will then inevitably fall, thanks to industrialisation and economies of scale, alongside lower costs of capital and insurance.
The demonstrators are showing the way, with Statoil's Hywind recording a 37-52% load factor and 95% availability, while the spar floater has caused "no unexpected turbine downtime", Hywind says.
For what will probably be the world's first pre-commercial array - the 30MW Hywind Scotland project, expected to be operational in late 2017 - the spars will have a draft of 80 metres rather than 100 metres of Hywind I, despite supporting the bigger 6MW Siemens turbines.
Likewise, the 2MW Vestas turbine on Principle Power's WindFloat platform off Portugal exactly matched Vestas' power curve for the V80, reports Kevin Banister, Principle Power's senior manager of business development.
"Wind turbine performance was not degraded at all by being on the floater," he says. Principle Power expects the optimised WindFloat Atlantic array to be 60% cheaper per megawatt than the demonstrator. Comprising six 8MW turbines, it should be installed off Portugal in 2018 if all goes to plan.
Japan targets 2020
These and other pre-commercial arrays planned off France, Japan, Taiwan and the US will focus on testing wake effects and optimising assembly and installation methods, mooring and anchoring systems and maintenance strategies.
Work also needs to be done on the electrical infrastructure, including dynamic cables, floating substations and distributed transformers.
Even so, many believe the first commercial arrays could be online in the early 2020s, not necessarily in Europe. Despite cost overruns on the Fukushima Forward demonstrators, the Japanese government has announced that the 2020 Olympics will be powered (at least partly) by floating wind turbines.
"They want a representative project of renaissance," notes Bruno Geschier, chief sales and marketing officer for Ideol, which is building two demonstrators in Japan.
While deployment is largely government-driven in Japan, the country also has limited sites for conventional offshore turbines. High energy costs and good infrastructure are creating good conditions for the sector.
Hawaii is another contender for early commercial-scale deployment, again due to its deep water, high electricity prices and limited alternatives. Three projects of 400MW each, all featuring the WindFloat platform, are under development. If all goes well, the first turbines could be online in 2020, the developers say.
Mutual benefits for oil and wind
Others believe the oil and gas industry could offer the first commercial application for floating wind. The oil and gas industry moved from fixed to floating foundations in the 1990s.
Now, several firms are looking at floating turbines to power water-injection systems to squeeze more oil and gas from mature or marginal fields.
In certain circumstances, this will be "cheaper and much more flexible than conventional solutions", says Johan Sandberg, segment leader for floating wind turbines at DNV GL, which is leading the "Win Win" joint industry project, involving seven partners including oil and gas operators ExxonMobil and Statoil.
Importantly, this would be "a purely commercial market, with the huge advantage of not having to rely on subsidies, with its associated political risks," says Sandberg. That would significantly increase the pace of development and mobilise the oil and gas sector to bring its ingenuity and creativity to the table, he adds.
Wherever they are, these early movers will be "critical to de-risk the technology", says Rhodri James, an analyst at the Carbon Trust. And the long-term prospects are very promising. Make Consulting estimates around 3.4GW will be operational by 2030.
The speed of progress has taken many by surprise, with floating wind moving from being regarded as "an investor's folly" in the early 2010s to industry and government now taking it "very seriously", as Nordstrom says. That is a remarkable shift.