Oscillating Water Column (OWC)
The Oscillating Water Column (OWC) can be conceptualised as a hollow cylinder, the lower portion of which is in the water whilst the upper portion is in the air and capped. Part of each passing wave enters the device via an aperture at the bottom and rises up the tube in tandem with the rising wave front.
This pressurises the air contained in the capped upper part of the tube, pushing some of it out via an exit duct to an air-driven turbine that is connected to a generator. By ducting the air appropriately to the turbine, generation can also be maintained after the wave has peaked and its level is reducing again, sucking air out of the collection chamber and into the system via the associated turbine.
Normally an impulse generator or Wells-type generator is used so that rotation is maintained in the same sense whichever direction the air is flowing in.
So far so good. But an interesting feature of the wave power technology is that, by harnessing the natural dynamic effect of resonance, power extracted can be greatly magnified. Resonance occurs when the forcing impulse, the wave, impinges on the device in time with the latter’s own natural frequency.
A useful analogy is that of a child on a swing. Given an initial impulse to get it moving, the swing executes each to-and-fro cycle in a given time, or period. That period does not vary unless the system is physically altered, for example by shortening the chains on which the seat is suspended. By giving the swing a small push each time it completes a cycle and starts to reverse its motion, it can be made to go higher with little effort.
Similarly, an oscillating water column has its own inherent ‘swing’ period or natural frequency. Given an initial forcing impulse, the water within the column will tend to rise and fall back in a fixed time, this period being governed by various system parameters including the form and dimensions of the collection wave energy chamber, the volume of the enclosed air and the resistance presented by the turbine.
Without further forcing inputs, the column will continue to oscillate with decreasing amplitude until the initial energy has been dissipated and the column steadies. On the other hand, introducing further energy will increase the oscillation and, by applying wave induced pushes at the right frequency and time, the variation in the level of the water column can be made to exceed the wave peak-to-trough distance by a considerable margin.
This results in more air passing through the turbine than would otherwise be the case, and hence more electrical power being generated.
The periods of wave encounters at a given location vary, although frequencies tend to cluster within a given bandwidth. Moreover, wave trains may be simple and regular but are just as often superimposed over each other so that each wave can incorporate a number of frequencies and periods. These factors make designing for resonance a complex matter. An oscillating water column able to resonate over a relatively broad band of frequencies would be able to deliver usefully amplified power levels over a wide range of wave conditions.
Multi-resonant solution
One wave energy company aiming for such a result is Orecon Ltd, based in Cornwall, UK. Orecon’s floating multiple resonant cavity (MRC) device incorporates multiple chambers, each tuned – by virtue of its internal dimensions – to a particular range of wave frequencies.
In 2010, after several years of development and testing at model scale, Orecon is due to start manufacturing a pilot-scale multiple resonant cavity system for installation at North Cornwall’s WaveHub, which is projected to start operating and connecting offshore marine renewable energy devices to the UK national grid from 2011.
According to Orecon’s Business Development Director Ken Street, the steel-built multiple resonant cavity system (measuring 45 m x 31 m) will generate up to 1.5 MW. He explains that each multiple resonant cavity has three chambers that have the same cross-sectional area but different air draughts. The draughts are set so that each chamber can respond resonantly to a specific wave field:
“Between them, the three chambers can cover most of the frequency range typical of wave fields around the UK”, he says. “We considered building in arrangements to vary chamber draught, so that it would have been possible to tune our system more precisely, but this was ruled out on grounds of complexity and cost. As it is, we can extract three to four times the power that a non-resonant OWC can achieve. Our approach is a great improvement on single-chamber devices and should help make OWCs commercially viable.”
An advantage of the multiple resonant cavity, along with other oscillating water column devices, is that the power train and other active parts are out of the water above sea level, leaving only the lower passive structure exposed to the ravages of salt water. This both minimises and simplifies maintenance, which can be carried out on-station.
A water-borne system, the multiple resonant cavity is tension-moored to the sea bed so that it does not heave (rise and fall) with the passing waves. Nor does it adjust for tidal rise and fall, this being accommodated by a variation in chamber air draught that occurs as the tide ebbs and floods.
In operation, incoming waves encounter the device’s two smaller-draught chambers first, and these extract the most wave energy. Once they are saturated in higher sea states, further energy is extracted from the wave as it encounters the third, rear chamber. Each chamber has its own independent power train. Guide vanes direct air flow to the turbine rotor and ensure that it rotates always in the same direction. Hour glass shaped inlet ducts accelerate the air as it approaches the turbine, enhancing efficiency.
In extreme sea states, a shut-off off valve prevents entry of solid water to the turbine. A power converter allows the generator to rotate at optimum speed for the chamber conditions and independently of grid frequency. Output voltage is transformed on board to 33 kV to minimise transmission losses as the supply is conveyed ashore by cable to a local substation. Orecon has opted to use impulse turbines rather than the Wells-type turbine that has hitherto equipped many wave energy prototypes (see below).
The wave energy company, founded in 2002 by two University of Plymouth post-graduates, has raised private investment, along with a DTI Award (now BERR) and a Carbon Trust grant, enabling it to proceed with development and deployment of its device even in these times of financial stringency.
Technical Director Fraser Johnson says that the multiple resonant cavity is designed to be robust and reliable, using tried and tested components from other marine and industrial sectors. The steel-constructed multiple resonant cavity, expected to weigh about 1000 tonnes, is being built and tested to the standards of the Det Norske Veritas (DNV) classification society. The structure incorporates trimming tanks along with power-enhancing scoops, access ladders, towing points etc.
Although means of achieving resonance are becoming more sophisticated as oscillating water column technology develops, the effect has been utilised for some time. Wavegen’s LIMPET wave energy device, for instance, which has been operating since November 2000 on an exposed cliff edge in the Orkney Islands off northern Scotland, has its collector tilted so that the period of the internal water column matches the peak energy period of the waves, easing passage of water into the water column.
LIMPET’s collector is configured as three chambers, each with an aperture at the top. Air above the three water columns combines to feed the generation system. The latter comprises two Wells turbines, each connected to a 250 kW generator that feeds power to the Islay grid. The system is designed to perform optimally in average annual wave intensities of 15-25 kW/m.
Breakthrough
An alternative to the multi-resonant chamber approach is to have a large single chamber able to resonate across a range of wave conditions. Here, a finding by a research team at the Massachusetts Institute of Technology (MIT), with additional involvement of Portuguese academics, could have important repercussions for future oscillating water column designs. Some commentators have described the finding as a breakthrough. The key to successful resonant design, it seems, is the compressibility of air.
Chiang Mei, the Ford Professor of Engineering in the Department of Civil and Environmental Engineering at MIT has, with colleagues, developed numerical methods for simulating wave forces on various devices and the way the devices respond. These simulations can help designers to maximise energy capture.
MIT’s first detailed model was of a near-shore oscillating water column device, then being tested by researchers at the Technical University of Lisbon in Portugal. The Portuguese have amassed considerable expertise in marine wave power having had, in particular, a pilot-scale experimental oscillating water column system operating at the Azores island of Pico since 1990.
The more recent wave energy prototype tested and modelled by MIT proved insufficiently potent to attract commercial investors, so it was a case of back to the drawing board. Simulations conducted by Prof Mei’s team at MIT showed that if the diameter of the oscillating water column was greatly enlarged, it would be able to generate a resonant response to periodic waves, thereby magnifying the power generated to a level much more likely to attract investors:
As the Professor told Renewable Energy Focus, “most designers of OWCs have thought in terms of collection chamber diameters of three metres or less. We considered a 10 m chamber, consequently having a much larger volume of air above the water column. Our modelling has shown that a small air volume gives rise to a single resonant peak only, such that the system responds resonantly to just a single wave frequency. Since sea waves are essentially broadband, this is not an optimum situation.
“However, if the chamber volume is made considerably larger, the air enclosed in what is in effect a pneumatic chamber acts like a spring, and the spring force can be made to match the hydrodynamic inertia of the water column. At this condition, it turns out that there are two resonant peaks, giving something more akin to a broadband response. It is even possible to engineer a four-peak condition, making for a better response still. If you make the air volume too large, however, there is no resonance at all. Therefore the actual volume has to fall within finite limits, which can be modelled.”
Commenting on his own reaction to the findings, Prof Mei added, “It was exciting to discover that we can thank the compressibility of air in the chamber for broadening the bandwidth of extraction efficiency.”
Some control over the chamber’s resonance characteristics can be afforded at the design stage by choosing the pressure the enclosed air is subjected to. This can then be engineered by controlling the resistance to the passage of air through the turbines, by careful specification of the turbine characteristics. Whether it would be possible or even desirable to exercise control dynamically during system operation by altering the turbine characteristics – blade or guide vane angles for instance – or mediating the drag exerted by the attached generator, is a moot point and further research would be needed.
Prof Mei at MIT has had an on-going collaboration with Professors Antonio Falcao, Antonio Sarmento and Luis Gato of Instituto Superior Tecnico at the University of Lisbon, since the Portuguese likewise seek to harness resonance to enhance the efficiency of wave energy capture.
The Lisbon team had hoped to develop a 10 m version of MIT’s oscillating water column design for intended installation in the end of a breakwater on the Duoro River near the city of Oporto. Under the plan, three similar oscillating water columns integrated into a breakwater at the mouth of the river would generate up to 750 kW of wave power. The new wave energy absorber would exploit the effect of the large oscillating water column capture chamber to enable efficient energy capture across the broad spectrum of wave frequencies expected at the location.
It is believed, however, that further progress on this project is currently in doubt because of financial constraints.
Mei concedes that the large collection chambers indicated by his research may be rather more expensive to construct than those having more customary dimensions. He points out that his expertise is limited to hydrodynamic modelling and that he has not considered economic factors.
He further adds that the simulations his team has carried out relate primarily to an idealised ‘vertical tube’ oscillating water column and that more detailed modelling would be needed for actual devices, most of which would be of less regular form.
Indeed, be believes that, while wave energy devices have enormous promise for the future, much more research is needed before they can become a commercial reality. Moreover, because individual oscillating water columns can generate only modest amounts of power, he envisions a future in which multiple devices will be deployed in arrays – necessitating the investigation of interactions and spacings etc.
Nor is the oscillating water column the only candidate competing for the future wave energy mainstream. For example, Mei professes himself intrigued by the Manchester Bobber and has carried out work to model arrays of these devices. The Bobber was conceived some five years ago by a University of Manchester team led by Professor Peter Stansby and Dr Alan Williamson.
Stansby and Williamson envisaged a number of bobbing buoys deployed in an array from an offshore structure, which could be a disused rig. For each floating buoy, the oscillating motion caused by passing waves is converted to unidirectional motion by a clutch and flywheel system, and then to electrical energy by a connected induction generator. The flywheel also has an integrating effect, tending to iron out power extremes that would otherwise occur during the cycle.
In common with other bobbing wave energy converters, the Manchester Bobber can benefit from resonance. This involves designing the buoy/floats and associated mechanical system so that the total system’s natural period coincides with that of passing waves. By careful attention to buoy size, weight and hydrodynamic characteristics, designers can widen the wave band over which the system can respond resonantly.
The Manchester Bobber has been extensively tested at one hundredth and one tenth scales, and development is being carried forward by a Manchester University spin-off company with several commercial partners. Similar considerations apply to other heave-based devices such as Ocean Power Technologies’ PowerBuoy series and the FO2 Wave Energy Converter.
The turbine factor
Utilising resonance effectively will be key to improving the commercial prospects of wave energy capture devices, many of which have to date delivered disappointing amounts of energy. Wave-to-wire conversion efficiencies as low as 10% have been quoted for first-generation prototypes.
However, an equally important focus for wave energy converter designers is the turbine, the characteristics of which have a great influence on overall efficiency.
Turbines that reverse their rotation during each ‘blow-suck’ cycle of the driving air supply would be impractical. It is enough that turbines have to be reliable in the onerous marine environment without being subject to constant fatigue and failure inducing cycling reversals. Solutions such as variable-pitch blades or dual turbines add unwanted complexity and threaten system reliability.
One answer to this is to rectify the driving air supply so that turbine rotation is unidirectional, but this requires delicate check valve arrangements and, again, unwanted complexity.
An ingenious solution devised by Professor Alan Wells of Queens University, Belfast, in the late 1980s avoids the need to rectify the air supply. Conceptually, the Wells turbine resembles certain forms of vertical axis wind turbines that always revolve in the same sense irrespective of wind direction. Symmetrical aerofoil blades aligned end-on to the airflow achieve the required result, enabling the turbine of an oscillating water column to maintain its same sense of rotation through reversals of driving airflow. Wells turbines were used in many of the early oscillating water column wave energy prototypes.
Unfortunately, the Wells turbine has certain drawbacks. Its efficiency is lower than that of a turbine having asymmetric airfoils, due to higher drag. Furthermore, symmetric airfoils are prone to stalling when their angle of attack to the airflow is high. After stalling, the turbine cannot re-start itself. Performance can be improved with enhancements such as optimised blade settings and use of end plates, but essentially the range of conditions in which the Wells turbine can be efficient is limited.
An alternative that is well suited to the low-pressure, high-volume airflow regime of an oscilating water column is the impulse turbine, in which the blades, rather than producing airfoil lift, react to the impulse of the impinging airflow by deflecting it.
Bi-directional airflow can be accommodated by having two sets of static guide vanes located one on each side of the turbine rotor and at some distance from it. The blades are attached circumferentially to the inside of an hourglass-shaped duct, at the central, narrow, point of which is located the rotor. Air entering this duct at modest velocity is given a swirl motion as it passes through the inlet vanes and is accelerated by the hourglass shape.
Accelerating air drives the rotor, then decelerates as it travels on through the now expanding duct half and through the outlet vanes. When the airflow direction reverses, the ducts that were previously at the outlet become the inlet ducts and, due to their orientation, swirl the air in a direction that will enable the rotor to maintain its rotation in the same sense as before (see illustration)
Orecon says that the HydroAir bi-directional impulse turbine it uses on its multiple resonant cavity wave energy system is better suited than competing (Wells) types to the low-pressure, high-volume airflow regime of an oscillating water column. It has a lower rotation speed (500-600 rpm), is self-starting and has a wide operational bandwidth. By combining their technologies, turbine maker Dresser-Rand and Orecon aim to transform the performance of new oscillating water columns and boost confidence that these wave energy converters can make a viable contribution to a future sustainable energy infrastructure.