The UK Energy Review (Department for Trade and Industry, 2006, p. 15) concluded that:
“Over the next two decades, it is likely that we will need around 25 GW of new electricity generation capacity, as power stations - principally, coal and nuclear plants—reach the end of their lives and close. This will require substantial new investment and is equivalent to around one third of today's generation capacity.”
For both environmental and energy security reasons, there is an increased recognition that existing fossil fuel technology cannot continue to be as heavily used as in the past and there is a corresponding movement towards generation technologies that operate with low, or zero, carbon emissions. These include renewable technologies, such as hydro, on- and off-shore wind, and marine (wave and tidal) devices.
The use of wind technology to generate electricity has grown rapidly across the UK in the last decade. However, other renewable technologies, such as marine, have also received both financial support and political interest and the first generation of economically viable devices is now close to market.
Scotland is a leading player in the development of marine renewable technologies, being home to a large number of companies, academic experts, support facilities and strong government support for the sector. Within this set-up the European Marine Energy Centre (EMEC) provides a rigorous set of full scale conditions for the verification of wave and tidal devices, both in terms of the underpinning resource, but also in terms of survivability due to the seas that the sites experience.
The process of device development from design, through laboratory and tank testing to full scale grid-connected testing in the open sea presents enormous challenges. In the early stages of device development, the principle function of testing is to help the designer confirm the actual performance of the device in different conditions and gather engineering data that are otherwise difficult to obtain.
As well as having engineering significance, early test results are needed to give confidence that it is worthwhile investing funds to continue development activities. As development proceeds, developers need to raise funds from different sources and potential investors have different criteria for investment. Previous test results may meet these criteria, or the criteria could require further tests.
Different types of testing are appropriate at each stage of device development. As a design is developed, it will be necessary to consider not just the overall device, but also 'drill-down' into sub-systems and components (particularly where these are critical to overall performance, reliability or survivability). The step of moving from tank tests to full grid-connection in the open sea is a huge one, and is in part responsible for delays to developers’ initial deployment plans at EMEC.
It is apparent that an intermediate testing facility is required, which will provide the rigours of the open sea, but without the additional complexity of grid-connection. Such a facility would enable developers to gain key operational experience in an appropriate environment, as well as provide further validation of device mechanical performance. This approach also has the potential to meet the requirements of both institutional and private investors as it more closely mirrors the development pathway experienced in other developing technology environments. A scale mechanical test facility would also enable more technology options to be explored at significantly lower cost, thereby maximising the potential to progress towards design consensus.
At present, EMEC offers full scale berths for wave system and tidal developers. Typically, devices will have been tested at up to 1/10th scale in facilities such as those offered by the New and Renewable Energy Centre (NaREC) or at 1/50th scale in facilities offered by Universities or other research organisations.
The leap from 1/10th to full scale is a big one and necessitates extreme caution on behalf of the developer. This is, at least partly, as a result of inevitable uncertainties in the application of scaling practice between the essentially controlled environment of a 1/10th facility and the harsher operating conditions implicit in full scale testing.
There are vast operational differences between 1/10th and full scale tests. Tests at EMEC’s Billia Croo and Fall of Warness sites require sophisticated high risk and expensive marine operations the planning of which must dominate considerations of project development.
It makes sense to allow access to sea water in earlier stages of development, in such a way as to allow device related analysis, without the expense and constraints forced upon developers by ship and marine operations. In addition to developers, a low cost ¼ scale test facility would also be of direct interest to research institutions investigating fundamental concepts rather than focussing on the development of devices. Accepting the obvious case for intermediate scale testing as part of the technology development pathway, the development of a centralised national facility in support of the tidal energy industry has obvious merits:
Licensing and permitting requirements for operation in the coastal environment are potentially very onerous both economically and in delaying technology development. Adopting one centralised intermediate scale testing facility would ensure that an appropriate site was identified and properly legislated, removing this burden from individual developers.
The provision of support infrastructure in terms of resource and device monitoring equipment, appropriate boat access, on-shore facilities, etc. Would reduce development costs to the sector.
Continuous facility operation at one particular location would build a rich understanding and knowledgebase of data characterising the site, enabling developers to better understand the conditions impacting on their individual device compared with the alternative of developers each adopting their own separate location.
A centralised facility would enable test facility staff to amass appropriate operational experience which would be to the benefit of the development community.
The aim of this dissertation will be to review the potential for and requirements of scale test facilities for Marine Energy Convertors (MECs). It is intended to look at the physical requirements, any scaling issues and some potential locations of both wave and tidal test sites preferably within Orkney. The reason for the choice of area is to continue to promote Orkney as a centre of excellence in the marine renewables sector and to allow the utilisation of the skills and experience already in existence within the local community.
It is also intended to look at the mooring arrangements that could be used at such sites. Offering generic mooring arrangement for scale testing will allow developers to concentrate of proof of concept and reduce development costs.
As stated above, the main objective of this project is to investigate the potential for locating test centres off Orkney in what is a much less harsh environment than that experienced at the current EMEC sites. Initially, this involves assessing whether or not there are any suitable sites for such a development, taking into account issues such as the nature of the resource, the local bathymetry and the local infrastructure. If it transpires that the area does lend itself to hosting a tidal or wave test centre then the question arises of what this will involve, what changes or improvements will have to be made locally to accommodate it and what kind of impact it would have, both environmentally and economically.
This project will seek to address these issues at this early stage and provide a conclusion regarding whether or not there is potential for pursuing this development and to provide recommendations regarding the geographical areas that would be suited to it. It also aims to provide recommendations on what steps must be taken in both the immediate future and long term to ensure that the potential for exploitation of the tidal and wave resource in the region on a commercial scale is maximised.
3.1 The Development Lifecycle
The marine renewable energy technology industry is at a pre-commercialisation development stage of the research, development, demonstration and deployment cycle. The UK is widely acknowledged as a leading player in the development of marine renewable technologies, being home to a disproportionately large percentage of the technology companies, academic experts, support facilities, and with strong government support including finance mechanisms.
Two avenues for exploiting the marine resource are undergoing particularly rapid development, those being wave energy and tidal current energy. Technology progression from drawing board through to commercialisation of full-scale devices tends to be supported in both technology sectors by the development of numerical and physical modelling tools.
These tools enable a rigorous design and development approach through testing of the technology performance, design improvements, model improvement, and more testing to build confidence in the product. A marine energy technology developer will also simultaneously be developing bespoke numerical modelling tools and systems alongside their individual technologies. Some developers will also be able to construct in-house physical model testing capability (e.g. Wavegen’s scale wave basin for small scale testing and design development).
Similar facilities are housed in various universities throughout the UK which provide a pool of research facilities of which developers can avail themselves for small scale device testing and design development. The UK is also home to the European Marine Energy Centre (EMEC)at Orkney. EMEC provides full-scale testing facilities for both wave energy technologies at Billia Croo and tidal energy technologies at the Falls of Warness. At each of these locations a number of grid-connected ‘berths’ are available for technology developers to lease to use for performance testing of their device and to learn, develop and hone operational experience in installation, day-to-day operation, maintenance and decommissioning of the technology.
This is particularly beneficial to the development of the marine renewable energy industry as it provides a ‘ready made’ full-scale testing facility with much of the ancillary development issues facing a developer already taken care of – permitting from the necessary regulatory bodies (Crown Estate, etc.), appropriate resource monitoring equipment in place to support device development and performance assessment, archived quality data sets detailing the characteristics of the site, etc..
The leap from laboratory scale device experience to full scale technology deployment however is vast and necessitates extreme caution on behalf of the developer. This is, at least partly, as a result of inevitable uncertainties in the application of scaling practice between the essentially controlled environment of a laboratory facility and the harsher operating conditions implicit in full scale testing. Furthermore, the developer will have gained only the most rudimentary operational experience in the laboratory environment with regards to full-scale deployment and operational practices (e.g. most available laboratory facilities will only be able to provide suitable scaled testing conditions for wave or current conditions, where of course these two phenomena do not occur in isolation in the real environment). Additionally, EMEC provides a particularly extreme set of full scale conditions both in terms of the scale of the underpinning resource, but also in terms of survivability due to the propensity of severe storms and hence storm seas that the region experiences. EMEC therefore provides a particularly rigorous final test-site for developing technologies.
There are still fewer than 10 device-years of operational experience of wave and tidal current devices at sea. Most extant knowledge has accrued from testing small-scale models in tanks, which must continue to reduce the financial risk to the developer. At present, numerical modelling doesn’t provide reliable enough results to prevent the need for tank testing, and there is still much merit in and need for physical modelling in tanks from 1/100 scale to 1/10 scale (perhaps even at 1/3 scale) before full-scale testing in the open sea.
Tank testing is much faster and more repeatable than at sea and can emulate extreme events, at scale, again and again to allow understanding and mitigation of their effects with improved concepts and designs. Repeatability of sea states and emulation of extreme events in tank tests enables the developer to verify concepts, performance, design software, control strategies and investigate survivability. However, the time and investment required to go from concept at 1/100 scale to full scale in several stages can take a few years.
It must be highlighted that as a sector, the marine energy community has limited laboratory scale testing experience either in laboratory or open sea conditions. Additionally only three recent examples of large scale technology deployment in the open sea environment – Marine Current Turbines Ltd. (MCT) 300kW SEAFLOW demonstrator programme in the Bristol Channel, Hammerfest Strom AS’s 300kW demonstrator at Kvalsundet, Norway and Open Hydro’s ongoing turbine testing at EMEC. On the wave front, Pelamis has conducted scale model trials, deployment of a 750kW device at EMEC and an array of 3 x 750kW off Portugal.
The only developers that have conducted a full scale device deployment beyond this scale is MCT with their 1.2MW device deployed at Strangford Lough and Pelamis as noted above. Furthermore, the prescription of a ‘full-scale’ device is currently driven by the progression of alternative technologies such as wind turbine farms which now are typically in the size range 1-2MW per device.
Marine renewable energy development is striving to provide a similar rated electrical device for the market. This highlights an interesting contrast, as wind turbines have followed a continuous design progression over the last 25 years where significant investment has supported development of technology which over that time period has now reached a level of maturity where 1-2MW devices are the norm, and design consensus has been achieved – all wind turbines now follow the standard tower and 3 bladed turbine design typically with a gearbox and electrical generator. However perspective of a ‘full-scale’ device has progressed in the wind industry from 50kW in 1980, through 500kW by the early-mid 1990’s to reach today’s multi-MW potential devices. It is suggested that the marine renewable energy community is attempting to fast-track 25 years of technology innovation, investment and development observed in the wind industry and achieve similar performance in one giant leap by omitting intermediate development stages.
The European Marine Energy Centre (EMEC) in Orkney is at the international forefront of development in marine energy technology. The centre is the first of its kind anywhere in the world and offers a unique opportunity to test and develop the various fledgling marine energy technologies in an operational environment.
EMEC currently occupies facilities on two sites on Orkney, at Stromness and Eday. The main reason for situating the facilities in Orkney is because of the nature of the marine resource on the doorstep. In terms of wave energy, the wave site has uninterrupted exposure to waves of up to 15m whereas the tidal site experiences spring tide peak flows of 4 m/s – approximately 8 knots.
The location in Orkney also has other benefits that made it attractive for the development of a test centre. It offers relatively good access to grid connection compared to other more remote locations. Also, the location in Stromness is close to the local infrastructure, considerable local expertise and its long seafaring history.
The Wave Site
The wave site has been fully operational since October 2003 with 4 test berths situated about 2 km offshore at Billia Croo. The four berths are located on the 50m contour line and are connected to the national grid through the centre’s substation onshore.
The wave conditions on site are continuously monitored by two Waverider buoys at the test berths. These buoys are connected via radio link to the EMEC data centre where their readings are recorded. This arrangement enables EMEC to amass a comprehensive dataset of conditions on site and it is available to all developers who use the site. There is also comprehensive monitoring of weather conditions in the area and measurement of factors relating to power quality at the substation.
All of this serves to provide valuable real-time data for any device being tested on site, with the potential to evaluate how differing weather conditions and wave types translate into electrical power, usable or otherwise. This kind of information is vital in developing new technology, giving an opportunity to amend device design in order to maximise potential.
The wave test centre has already seen some success stories, most notably the Pelamis system, pioneered by Ocean Power Delivery (now Pelamis Wave Power). Pelamis began testing at Billia Croo in August 2004, when it became the first grid connected wave power generation system in the world. The device has also been exported to Portugal where a 2.25 MW array of three Pelamis devices has been trialled.
Other developers are constructing their machines to test at EMEC including:
Aquamarine Power (Oyster device). Currently being installed.
Ocean Power Technology (PowerBuoy).
The Tidal Site
The tidal site on Eday is located to the west of the island in an area known as the Fall of Warness and consists of five test berths. As with the wave site, each berth is connected to the grid via a substation onshore. There is a separate communications link with the office facilities in Stromness to relay telemetry from test devices.
The tidal site was completed in spring 2007 and became operational soon after. Depths vary between berths from the shallowest at 25m up to a 50m limit. There is monitoring of tidal conditions using Acoustic Doppler Current Profilers (ADCPs) deployed at the berths, offering comprehensive data regarding the tidal resource.
EMEC is home to the UK’s first grid connected tidal turbine at the tidal test facility off the island of Eday, Orkney. Installed Nov 06, the device generated on test for the first time in May 2008. The turbine test rig was put up by OpenHydro; an Irish company attracted to Orkney because of the test site built here. OpenHydro’s next step involved a seabed mounted device deployed by its own specialised installation vessel. This is a mechanical unit and is regularly monitored to see how it is surviving in the elements.
Other developers are in the process of bringing their machines to the site:
Tidal Generation Limited of Bristol are planning to complete the installation this year.
Lunar – 2010
Hammerfest Strom – 2010
Scotrenewables - 2010
Use of Mechanical Facilities
There is already demand from developers for facilities for mechanical testing. A low cost test facility would also be of direct interest to research institutions investigating fundamental concepts rather than devices, and seeking to ground-truth modelling work. Access to the sea for research is expensive and vessel hire can often swamp a typical research grant of around £500,000.
Funding bodies would be uncomfortable to see a grant dominated by facility or vessel hire costs. However, up to 25% of a typical research grant could be allocated to such costs, provided they were identified at the time of application. The Research Councils have indicated that they expect to spend up to £70million per annum on renewable energy research, with at least 10% of this on marine, so the potential market is large.
The marine renewable energy companies who have most closely followed this staged design pathway are probably those closest to commercialisation: namely, Pelamis Wave Power and Aquamarine Power for wave and Marine Current Turbines and OpenHydro for tidal current technologies. This clearly demonstrates the necessity for the facility, with high demand expected.
Wave and tidal current modelling techniques are improving all the time and complex 2D and 3D representations and visualisations are now produced by animation houses from numerical models, including extreme events. Hydrodynamic modelling of the marine devices and their interaction with the complex sea surfaces and flow patterns is also improving, but it is not yet able adequately to represent the non-linear conditions in the forces acting to produce power and on the device.
Numerical modelling techniques developed to analyse ship-slam (Qian et al., 2006), or forces on breakwaters and offshore structures (Shiach et al., 2004) need to be extended to floating tethered devices and their predictions validated by physical modelling and full-scale testing. Wire frame, surface visualisation and hydrodynamic models can now operate in combination to represent and numerically predict the response of concepts but, ideally, device modellers should be able to sweep through evolutions of shape to optimise collector form.
Experience in the sea will be the ultimate calibration of numerical modelling techniques and test of adequacy of design, although good or bad experience gained after deployment is most expensive. In addition to being able to forecast or predict the behaviour of the sea, it will be critical for investor confidence to be able to predict device response, performance and survival.
Wave energy technology has been developed since the mid 1970s, but with sporadic progress. This is due partly to government policy and R&D support being intermittently favourable, partly in response to variations in fossil fuel prices, which sent positive signals to private investors at some times and negative signals at others. Tidal stream energy technologies began to be developed during the 1990s after UK R&D programmes into tidal barrage schemes were discontinued.
To date, worldwide government R&D support for wave and tidal stream energy has been much less than other electricity generation and low carbon technologies, including other renewable Interest in marine renewables has picked up over the last few years, particularly in the UK. New concepts have been brought forward and old ones re-evaluated in the current political and economic context of increasing support for renewable energy to combat the threat of climate change, increase security of supply and create economic growth.
Figure 3 (overleaf) shows notable recent UK events. Currently, many different concepts of wave energy converter and tidal stream energy generator compete for support and investment in technology development. Some concepts are more advanced than others, both in the sophistication of the technology and development progress to date. Overall, devices are at early stages compared to other renewables and conventional plant, and crucially, optimal designs have yet to be converged upon. A few largescale prototypes have been built and tested in real sea conditions, but no commercial wave and tidal stream projects have yet been completed.
The development of marine current devices relies much more on numerical modelling as these devices do not scale in the same way as wave devices, because of the problems of scaling Froude and Reynolds numbers. Modelling techniques used for wind turbines can only broadly be applied to marine current turbines, because they do not adequately represent cavitation effects, shear profile, proximity of sea bed and the existence of the free surface. Batten et al. (2006) have developed models to show when the onset of cavitation will occur, its impact on performance and how it can be avoided by design of the blade and the use of pitching.
New blade coatings need to be developed to offer increased cavitation resistance, but it may be better to avoid cavitation through appropriate design. Marine current devices will have to operate in turbulent flows, and their presence will increase downstream turbulence through vortex shedding and wake effects. Numerical and physical modelling must take account of this, particularly in the move from single to multiple or arrays of devices.
While this describes some of the individual challenges, waves prevail on the surface at tidal current sites, and tidal currents run below wave power sites. The combined effects of both on device performance and survivability are beginning to occur as deployment increases.
Numerical and physical modelling of the combined effects of wave and tidal currents on floating devices whose size can compare with wave length, or on submerged devices in 20–30m of water need to be better understood and predicted. Salter (2001a) has proposed a combined wave and tidal current tank with fully circular actuation and absorption at a scale of around 1/30. Such a facility could allow the repeatable measurement of individual and combined response to complex wave and tidal currents, which will be invaluable for device and mooring design, leading to improved station keeping and survival.
The Role of Physical Model Tests in Determining Commercial Viability
There is a definite need for a cost-effective methodology for hydrodynamic testing of early-stage MECs -the goal being that this methodology could then provide performance data in a standardised format to show which devices have potential for commercial viability and are worthy of further development.
It is prudent to test marine energy converters (MECs) through increasing model scales since physical trials provide a mechanism for risk reduction, both technical and financial. This approach is achieved by progressively improving engineering and scientific knowledge from the initial verification of the concept(s) to the demonstration of the economics with large scale models. The rationale for this approach is to ensure due diligence can be successfully performed on the technology and business plan whenever private or public finance is sought to support further development and progression.
Technical output of a well constructed MEC testing programme can provide information for all of the following objectives:
verification of the concept;
validation and calibration of numerical models;
quantification of technical performance variables;
provision of environmental loading data to allow design(s) to be improved, including moorings and foundations;
identification and development of understanding of relevant hydrodynamics and other physics processes;
provision of data for optimized performance design;
generation of detailed information for the power take-off engineers;
evaluation of the economics;
qualification of the device’s seakeeping ability and general seaworthiness;
survival (large scale models @ benign sites);
environmental impact (large scale models @ benign sites).
Different investigations should be applied in specific phases of the development programme. An essential element of the test schedule is to ensure the correct science is investigated at the appropriate time. The decision as to which parameters should be addressed at each phase should be based on practical considerations and previous experience, especially towards the reduction of uncertainty in later phases.
Scaling of Test Results
The scaling laws relevant to predicting the hydrodynamic performance of fully submerged prototype tidal devices from scale model tests are based mainly on ensuring the equivalence of the Reynolds numbers between the scale model and the full-scale device that it represents. Engineering and cost equivalences also exist between models and full-scale prototypes.
For example, difficulties with building models down to weight, or distributing weight in a way that minimizes internal structural loads, always indicate that similar difficulties will exist at full scale. If erosion, sticking or damage to particular components occurs in model testing, these components will be similarly vulnerable at full scale. These become even more relevant when considering testing in a marine environment rather than a fresh water tank.
Furthermore, engineering complexities that will make one-off full-scale device construction difficult and expensive will be reflected in the model construction costs and methods. Once such engineering complexity is identified, the possible economy that could come with the manufacture of large numbers of identical devices can usually be estimated.
Thus, the methodology provides an indication of the comparative cost of manufacturing the device in small quantities, the economies that may be possible if the device is manufactured in large quantities and any design features that indicate critical maintenance or reliability issues-all contributing to determining the commercial viability of the full-scale device.
One of the primary reasons that scale model testing is conducted is that there are rules that enable it to be done. Physical laws of similarity and similitude can be applied such that the fundamental behaviour of a process can be observed, even when the theory governing the situation is not well defined or even poorly understood. The physical model behaves as an analogue computer that introduces all the parameters and interactions that would occur at full size. Most wave energy converters seem like simple devices but there are usually quite complex hydrodynamic interactions.
However, as with all engineering solutions, there are conditions that should be followed. In this instance the principle consideration is that not all physical processes scale to the same laws. The consequence of this fact is that the dominant relationships are compared and the appropriate similitude law adopted.
For wave energy physical modelling there are two main criteria to consider, Froude and Reynolds. Froude equates inertial forces whilst Reynolds relates viscosity. Because of their derivation both of these two criteria cannot be matched in the same model. Since inertia dominates the majority of the forces experienced in wave energy device operation Froude scaling is selected as the primary law to match between the model and the prototype.
It is, however, important to know that there will be some forces present that are not to scale and their influence on the results should be qualified if not quantified. This is particularly the case in small scale wave energy models, circa λ < 1 : 50, that have large wave frequency, first order, excursions, especially heave. Such physical combinations produce the worst deviations from full scale results and are the primary rationale for the progressive increase in scale through the phases. This approach not only applies checks to the scale effect but, since different facilities support different scales, the laboratory effects also.
Scale application (Froude)
The factors by which a physical parameter should be multiplied to convert from model to prototype scale can be read from Table 8. There are two important considerations when applying this:
• certain parameters have large multiplicands (e.g. power, λ7/2)
• the extrapolation is dependent on the scale, λ.
As part of this dissertation a survey of developers has been conducted. The results are displayed below .
A questionnaire was devised and sent out to approximately 100 known MEC developers. Currently, at the time of writing only 25% of those sent questionnaires have returned responses. Is this
Ensure it’s up to date
It’s easy to check the date of Yellow Pages. But what about a list of businesses on a website, or from your organisation’s database?
It’s surprising how many lists have no dates. So you may need to do some investigation to check the source of the list.
If it’s a year or more old, you will probably need to update it. It’s a good idea to check how the list was collated in the first place, so you can follow the same process.
2. Ensure it’s comprehensive
Check your list includes the full range of people you want to consult.
Suppose you need to contact a sample of people from the construction sector. Your list will need to include builders, joiners, electricians, plumbers etc. If for some reason your list has no plumbers, you will need to add them to the list, or your survey will not represent all the trades.
Check your list represents the time period you want feedback from.
If you seek feedback from your clients during the past two years, you need to be sure your list is complete for two years, so your sample will be a true representation of all your clients. If the first six months of your project are missing, your sample will not represent the past two years.
3. Eliminate any duplicates
When lists are updated, extended, or merged, there’s a real danger of introducing duplicates. If someone on the list is contacted twice they won’t have much faith in your project, and it’s a waste of your resources. Also, if you are drawing a sample from your list, your sample will be distorted by duplicates.
Similitude criteria and similarity conditions enable physical properties to be appropriately scaled and investigated at different model sizes. These rules should enable most aspects of a WEC to be investigated at a small scale (circa λ = 1 : 50–100) but practical considerations and non-scalable parameters result in the suggested progressively increasing size scheme.
Little empirical evidence yet exists to quantify the difference in results due to this phenomenon since few full scale devices have been deployed to provide data for scale comparisons. Theoretical studies suggest that if compressibility is to be an issue it will only be encountered at prototype (full) scale.
Model testing, particularly at a small scale, tends to produce conservative results. Viscosity, and in particular vortex shedding from edges, does not scale appropriately and can be overestimated during physical testing. Attention to detail during model construction can minimize this effect but the best solution is to increase scale when investigating critical parameters.
This factor often leads to difficulty matching physical and theoretical mathematical model results (see 3.2.2). Empirical results can be used to calibrate numerical models but, since this is still an art rather than a science, care and attention to detail should be applied.
Experience in the offshore engineering industry suggests that floating structures exhibit little viscous effects so coefficients are set to unity in mathematical models. However, these are usually very large bodies designed to move as little as possible. Conversely wave energy devices are relatively small (<1,000 t) and experience large excursions in at least one degree of freedom. They also have the added complexity of a power take-off system that feeds back into the modes of motion so the unity approximation might not be valid.
Mechanical friction should also be carefully dealt with during model construction since it too does not scale when Froude similitude is imposed. This is particularly important during phase 1 and phase 2 of the programme. Power take-off simulators can be a source of problems if friction is not minimized. Forces experienced at small scale will only be a few Newtons and even well engineered PTOs can have resistance such that losses are proportionally high percentages of absorbed power.
A very careful selection of simulator is required. Low friction dampers can alleviate some of these uncertainties.
In pneumatic devices the stiffness of the air spring is not scaled correctly by geometry. For static devices this can be compensated for by the use of an external air reservoir. In mobile buoyant units this is more difficult due to the practicality of not influencing the vessel motion. The degree to which this parameter will affect any test result depends on the level of applied PTO damping.
Objectives of the Test Methodology
The objectives of the methodology, which addresses the hydrodynamic performance of the device structure and prime mover, are deliberately restricted to minimise the cost of testing at a stage when developers are generally either self-funding or have minimal funding from other sources. The tests should produce just enough data to enable developers to make an informed decision about further development and will be expandable into a wider test program if deemed worthwhile.
The first objective is to provide an indication of the relative costs of manufacturing different devices in one-off, or small numbers.
Because any technical issues in construction of full-scale devices are usually reflected in construction of scale models, the cost of building a model is often indicative of the relative cost of the corresponding full-scale device. This is particularly useful for comparing the cost of different devices. For this exercise, the sizes of the models must result in constant blockage ratios in the test facility (i.e., the models must have the same effective hydraulic size).
The second objective is to provide an indication of relative installation costs.
The cost of installing and conditioning a model provides useful information about the corresponding costs for the full-scale device. If, for example, device alignment is critical, or a device produces exceptionally high mooring loads, these features will cause similar problems for the installation of both the model and full-scale prototype. Thus, by building models to a constant blockage ratio (in the same way as for the first objective), it will be possible to predict the relative installation costs of different devices at full scale.
The third objective is to provide an indication of relative economies of scale for mass-produced devices. The separation of fixed and variable costs is to some extent subjective, but the separation identified for models is likely to be replicated for full-scale devices, and their relationship provides an illustration of the economies possible when a particular device can be mass-produced. The fourth objective is to provide information indicating the relative reliability and maintenance requirements of different devices.
Reliability and maintenance issues arising during model tests will be reported, along with an assessment of whether or not similar issues are likely to arise at full scale.
The fifth objective is to quantify the efficiency of different devices. Efficiency is a measure of the relationship between power intercepted by a device of unit size and the useful power that appears at its output. As such, it is an indicator of reliability, environmental impact, the quality of a device's design and engineering and the magnitude of mooring or foundation loads. The level of efficiency also indicates the suitability of a device for standalone installations or for incorporation into large arrays.
Reliability will be affected by efficiency because low efficiency indicates that a device is absorbing energy that remains in the device, rather than transmitting it as useful output power. This retained power is likely to appear in potentially damaging forms, such as overheating, high mooring or structural loads or large amplitudes of motion, deflection or fatigue.
Environmental impact is also related to efficiency because a device with a relatively low efficiency extracts more energy from the environment per unit of useful output than a device of relatively high efficiency.
Low efficiency is indicative of features such as a large cross section of inactive support structure relative to active components or poor matching of impeller design to flow conditions and gear ratio/alternator revolutions per minute.
Scale Model Testing
Tidal stream devices can be tested in either a towing tank, in which the device is towed through still water to simulate the flow, or in a circulating water channel (CWC), in which the device is stationary and the water is pumped through it. The towing tank allows bigger models because it generally has a large cross section, but it does not allow continuous testing. On the other hand, the CWC allows continuous testing, but only of small models.
The forces and loads measured on small models are low, and hence errors (those caused by friction in bearings, for example) are more serious. Equivalence of flow conditions between model and prototype may not be achieved, owing to low Reynolds numbers, which means the flow over the model may be laminar rather than the turbulent flow prevailing around the full-scale device. This second problem is less apparent in a CWC, because the incident flow may already be turbulent.
The advantage of larger models in a towing tank is partly offset by the extremely low initial turbulence of the stationary water. Turbulence stimulators may be employed to help overcome this problem.
Where it is not possible to test models of large devices at properly scaled flow speeds, measurements made during model tests will give a good indication of prototype performance, provided that the flow is fully turbulent.
Conduct of Tests
The methodology requires measurements of the power present in the inflow to the device, the power present in the efflux and the power available to drive the device's electricity generator. The inflow and efflux powers will be deduced from measurement of the flow distribution through a cross section of the test facility ahead of and behind the position of the device, first in the absence of the device, and then with the device in place.
This flow data will be analyzed to calculate the input power to the device and the power extracted from the flow at a given operating point. The generated torque and shaft speed will also be measured.
As a cross check on the input power data that is deduced from the flow distribution at the device inlet and outlet the lost power may be calculated from the product of the average flow speed (or towing speed) and the drag on the device.
At a given flow speed, the load on the output shaft of the device will be varied in order to determine the speed/torque characteristics of the device and its maximum power output. It will be necessary to repeat measurements of the inlet and outlet flow distributions in each case.
These measurements will be repeated for a number of different nominal flow conditions (i.e., free stream flow speed and, where appropriate, angle of incidence to the flow) in order to identify any systematic speed or Reynolds number-dependent effects. It is important to identify the flow speed required to overcome static friction in the model, because this factor does not scale in the same way as other factors affecting device performance.
Data Analysis and Reporting
The raw data from the torque, shaft speed and free stream flow velocity measurements will be processed to calculate dimensionless values of power and flow at each point of operation investigated. These dimensionless factors are known as the power and flow coefficients, respectively.
The power coefficient can be determined by the equation C^sub p^=Tω/ρω^sup 3^D^sup 5^=T/ρω^sup 2^D^sup 5^, and the flow coefficient can be determined by the equation C^sub q^=V^sub o^/ωD, where T=torque, measured in newton meters; ω=rotational speed, in radians per second; D=rotor diameter, in meters; ρ=water density, in kilograms per cubic meter; and V^sub o^=free stream velocity, in meters per second.
The flow distributions at the device inlet and outlet will be integrated and processed to calculate the efficiency, η, at the corresponding operating point.
If the laws of scaling apply for all test conditions, the dimensionless data will fall on a single characteristic (for example C^sub p^ versus C^sub q^) that defines the operating characteristic of the device.2 These data may be scaled up to predict full-scale device performance and used to compare the performance of different devices, if the tests were conducted at a scale producing the same blockage ratio.
Any divergence from a single characteristic (outside the limits of experimental error) indicates that one or more factors that influence scaling have been overlooked, and the modelling technique must be refined if prototype performance is to be accurately predicted.
As previously explained, qualitative information on the predicted costs of manufacturing (including potential economies of scale), installation and maintenance will also be provided in the test report. The model manufacturer will be invited to contribute a section explaining cost-critical features of the design of the device and commenting on the prospects for more economical manufacturing. Alternative manufacturing methods will be suggested, explaining how the choice of any particular method will affect the distribution between fixed and variable costs.
It should not be regarded as a requirement for scale sea trials that a link to the local distribution network is required. The restricted power rating enables alternatives such as resistive load banks or simply some form of pump to be used to dissipate the energy, even the high peaks experienced in irregular seas.
Technically, even if the connection was made, it cannot be regarded as addressing multi-megawatt prototype grid issues. Other power electronic options, however, do scale sufficiently for the developer to investigate supply matters with the aim to improve the quality of the electricity output. It would be recommended that this work should be studied on dry test rigs prior to sea trial deployment if possible.
Site Selection criteria
There are several important issues concerning the selection of suitable test sites. These are:
accessibility, a local, convenient and available harbour for light service tasks;
a nearby port for launch and delivery to site of the model;
prearranged licences and consents for deployment;
pre-deployed wave measurement instruments;
distance to landfall;
correct water depth;
appropriate seabed and bathymetry;
acceptable wave climate;
suitable tidal currents.
Of these the most difficult to achieve are the last 2, the local wave climate and tidale current suitability. These should be as close as possible to the scale conditions found at the site the test station is representing. There can be more than one representation at different scales.
The aim of the study was to identify the best site for a tidal energy test centre in the Highlands and Islands region of Scotland. A phased approach was adopted; an initial high level screening of eight potential sites was undertaken, followed by further examination of the three best sites. The key criteria that the site must fulfil are listed below:
1. A tidal stream channel of sufficient width to accommodate 3 or, ideally, 4 developers, each of whom are anticipated to require to bring an initial device for testing and to follow up with further devices after a year or so to produce an array of up to 5 devices.
2. The spacing of devices is to be 50 metres (m) between the centre of each device across the stream. That implies a width of stream of circa 200 m per developer with a 5-unit array.
3. The water depth is required to be in the range of 50 m in one area of the stream to accommodate at least one developer and 30 - 40 m across the rest of the main part of the stream for the others.
4. Currents peaking ideally up to 6 to 7 knots (3 – 3.5 m/s) at spring tides are required to give the range of velocity to adequately test the devices.
5. A free channel of 500 m width for small boats, including leisure craft and creel or other small fishing boats is required for navigation. Also, the selected location should not be part of, or too close to, a shipping lane or a route for tankers, large ferries or similar vessels.
6. Cost-effective grid access is required such that grid development will maintain pace with the increasing generation capacity of the test site.
7. The site shall be in an area where a suitable level of marine support exists to offer opportunity for local services for the operational phase.
The following criteria have been used as the basis for determining the optimum site:
1. Physical resource
a) dimensions and topography of site
b) water depth range of adequate extent
c) tidal current conditions (resource)
2. Environmental constraints and stakeholder issues
a) protected areas
b) sensitive species (benthic ecology, birds, marine mammals, etc)
c) shipping lane proximity for ferries, tankers, etc
d) availability of 0.5 km channel for small boats.
3. Differential Costs and commercial issues
a) construction (offshore and onshore)
d) data communications, and
e) integration with existing EMEC sites.
Much of the coastline and intertidal areas in the study area are protected by both national (Sites of Special Scientific Interest (SSSIs), National Nature Reserves (NNRs)) and international (Special Areas of Conservation (SACs), Special Protection Areas (SPAs) and Ramsar Sites) legislation. SACs and SPAs are afforded European conservation status and protection under the EU Habitats Directive. SSSI status provides the site with legal protection under the Wildlife And Countryside Act 1981, and NNR sites are protected under the National Parks and Access to the Countryside Act 1949.
For any development passing through SPA and SAC sites, including any export cable to shore, the project would need to prove that development would not affect the conservation importance of the sites. The project Environmental Report would need to contain sufficient information for the appropriate assessment to be undertaken, and bird and benthic surveys for example are likely to be required as a part of the process.
Scottish Natural Heritage is a statutory consultee in the consenting process, and they would be likely to require certain conditions such as timing installation activities within the designated areas to avoid sensitive periods. The Highlands and Islands region also contains a number of areas, which could potentially be designated as marine SPAs and SACs in the future. The UK government has also implemented the EU Habitats and Birds Directives in offshore waters.
The main impacts on benthic communities would arise from installation of any structures, foundations or moorings in or on the seabed and installation of the power or data cables connecting to the shore. The level of impact depends on the types of structures and cables installed, and installation method. It is likely, however, that benthic communities in the area would be impacted both by direct displacement of species located in the immediate vicinity of installation operations, and indirectly through re-distribution of any sediment present into the water column.
This effect will be localised and temporary and benthic communities could be expected to recover over a medium-term period (the speed of benthic recovery being dependent on the type of community originally present). The presence of any permanent structures on the seabed would alter the physical environment, and therefore may affect benthic communities in the region. However, a possible effect is that any such structure could act as “artificial reefs” to encourage colonisation by benthic species. If the cables were buried, then no impacts would be expected on benthic communities during operation of the cable, unless, of course, any repairs/maintenance needs to be done on the cable.
Shipping and navigation
The review which follows gives a high-level appraisal of maritime factors, particularly
marine traffic, which should be considered as part of the final selection process.
The comments against each potential site are based upon Admiralty Charts and Pilots,
supplemented by local information held by Orkney Harbours on individual areas and upon common marine practices.
Each of the tidal sites discussed are, by definition, areas where there are tidal flows. Associated with such areas are eddies, overfalls and other disturbances to the flow as a result of the high velocities and obstructions. Short, steep seas when the tide is flowing against a strong wind can also be a feature of such areas.
Mariners tend to avoid such areas because of difficult manoeuvring and discomfort. However, mariners with local knowledge may use the tidal effects to their advantage, when the flow is in the direction they wish to go.
Commercial fisheries and shellfisheries
Primary fisheries potentially operating in the Orkney, Shetland and Pentland Firth areas are creel (baited traps) fishing for lobster (summer) and crab (winter), dredging for scallops and small amounts of trawling for sandeels and whitefish.
Historically, herring fisheries were strong in the Orkney Islands but are now largely insignificant. The creel fleet can be separated into two fleets: small inshore boats which make day trips and land into local homeports, and the larger offshore crabbers capable of making multiday trips. The fleet operates throughout the year, and it is thought that most available seabed is set with gear at least some time during the year.
The majority of industrial trawling takes places well outside of the Phase 2 sites, primarily east and west of the Shetland and Orkney archipelagos. Bottom trawling for Nephrops is carried out in areas with muddy seabed and concentrated to the northwest of the OrkneyIslands. The small amount of trawling for whitefish which takes place within the Orkney Islands is most likely located in areas of smaller tidal flow than found in the study areas. Most of the trawling fleet fishes offshore for demersal species, although a number of local trawlers remain inshore.
Lobsters and edible crabs are the most important in terms of revenue. Lobsters are exploited nearly all year, and crabs are targeted from March to the start of the peak lobster season in August. Scallops and queen scallops are dredged over sand and gravel bottoms in coastal waters, around the islands and further offshore. The local creel fleet operates along protected areas of Eday but is not likely in the mid-channel area of Fall of Warness. Examination of the landings statistics from 2001 also indicates that the Orkney fishery is dominated by shellfish, with relatively little income from demersal and none from pelagic fisheries.
Shellfish waters and shellfish harvesting areas
Shellfish waters are designated under the EC Shellfish Waters Directive (79/923/EEC), which seeks to protect or improve shellfish waters in order to support shellfish life and growth and thus to contribute to the high quality of shellfish products directly edible by man. The Directive sets physical, chemical and microbiological water quality requirements that designated shellfish waters must either comply with or endeavour to meet.
Shellfish harvesting areas are designated under the Shellfish Hygiene Directive, administered by the food standards agency, and sets conditions for the production and marketing of shellfish intended for human consumption.
There are no designated shellfish waters or shellfish harvesting areas overlapping with the Phase 2 sites under consideration. However, it is likely that fishing for shellfish also takes place outside of the designated shellfish harvesting areas.
There are a large number of salmon farms throughout the Orkney area. Orkney presently has about 33 farms and Shetland over 170. Together, the two areas produce over 30% of Scotland’s farmed salmon. In 2001 Orkney salmon farms produced over 5,500 tonnes of product while Shetland farms produced nearly 40,000 tonnes of fish. Sea trout, halibut, common mussel, native oyster, pacific oyster, scallop and queen scallop are also cultivated in the region. No fish farms are located within the Phase 2 sites.
The location of wave and current energy generating devices in an area used for fishing will cause some impact on the commercial fishery. However, it is likely that the impact will be fairly small, as long as moorings and foundations cover only a small area and trawls (if present) are protected from snagging on cables and other seabed devices.
Oil and gas exploration and related developments
A review of current offshore oil and gas licences has indicated no licensed areas at any of the possible Phase 2 sites.
Cables and pipelines
There are a number of power cables connecting the various islands in the region to each other and to the mainland, and also a number of pipelines. There is an area of submarine cables running through the Yell Sound site connecting Yell to Mainland between the islands of Samphrey and Bigga (see Figure 13). There are also several pipelines running though the north and south of the Yell Sound site. There are no cables or pipelines passing through either of the other Phase 2 sites at Fall of Warness and Pentland Firth. Cables and pipelines are shown on Figures 13-15.
MOD exercise areas
No MOD exercise areas are located in the vicinity of any of the Phase 2 sites.
Wrecks and Archaeology
Wrecks data has not been obtained for this investigation, and the marine archaeology at each of the sites has not been assessed in any detail as it was outside the scope of this study. An initial examination of the Admiralty Charts does not show any known wrecks within the sites at Fall of Warness or Pentland Firth.
Specific Site Requirements:
These sites will be selected based on the following environmental criteria:
Provides a range of depths
Graded in terms of the resource available
A tidal facility to have bi-directional currents along with scale wind and wave conditions
A reasonably sheltered wave facility, giving protection from extreme storm conditions, which gives advantages for device survivability as well as improving access by smaller work boats
Sea bed conditions suitable for moorings
The sites will also meet the following operational requirements:
Good transport links
Access to general engineering services for on-site repairs and design development
A nearby harbour for on-site access with an economical scale of vessel
Ensuring that generic licensing and permitting in place
Well understood and monitored resource regime
Provision of dedicated personnel to facilitate the transfer of knowledge and solutions
No Grid connection required (as this would significantly impact on the cost and complexity)
What Might the Facilities Look Like?
In both cases, the requirement is to allow access to the environment in a safe manner, without the necessity for expensive vessel hire. In many ways the problems facing the tidal developer might be less, as a result of a smaller range of conditions being found in the environment. In principle at least, the wave facilities would have to face extremes of wave height like the full scale berths, whilst the tidal facility, like the tidal full scale sites, need only be exposed to an appropriate range of tidal conditions.
This needs to be in a water depth of approximately 10m. The site needs to be sheltered, to limit wave heights, and should not be exposed to tidal currents. Conditions are subject to Keulegan-Carpenter scaling. Five concrete block moorings could be placed at the corners and centre of a diamond configuration and attached to permanent buoys which are equipped with line loops allowing moorings to be drawn from the surface to the sea bed without diver intervention. The facility should be close to the coast, so that access can be made cheaply by small boat.
As well as being correct in geometric scale, physical models should be exact dynamic replicas of the prototype device they represent. This is achieved by following fixed scaling laws based on Froude similitude criteria and similarity conditions.
A full description of the relationships can be found in Clause 6 of this document but basically all physical properties can be related via inertia scaling such that:
Output power – Wave
At ¼ scale it would be expected that an operational PTO and generator combination is under investigation rather than the secondary conversion system. A scaled design of either a pneumatic air turbine, hydraulic or direct drive generator should be incorporated. This might have to be specifically designed for the large scale application, rather than direct copies of the final full size prototype PTO, but the characteristics will be similar. This ensures results from the trials can be confidently extrapolated. Even at this large scale electricity output levels are small due to the Froude scale factor for power. This limited level makes connection to the grid a moot point which can be decided by convenience, cost or public relations issues.
The calculation below is an example of this power scale issue for a ¼ scale design.
Power (kW) → λ 3.5 → 43.5 = 128
1MW Prototype ≈ 8kW @ ¼ scale
This should be in water depths up to 10m and be exposed to tidal currents up to 1.5m/s. These conditions are subject to Froude scaling.
Ideally, in this case, there could be a rudimentary platform connected to the shoreline by a retractable walkway, as the tidal environment would otherwise preclude easy small boat access. The platform should be equipped with the appropriate safety equipment and low voltage electrical supplies. There needs to be a mechanically driven platform onto which tidal systems could be attached and lowered to the sea bed for testing. The driving mechanism could, if necessary, be manual.
The dimensionless quantity U(gL)−½, where U is a characteristic velocity of flow, g is the acceleration of gravity, and L is a characteristic length. The Froude number can be interpreted as the ratio of the inertial to gravity forces in the flow. This ratio may also be interpreted physically as the ratio between the mean flow velocity and the speed of an elementary gravity (surface or disturbance) wave traveling over the water surface.
When the Froude number is equal to one, the speed of the surface wave and that of the flow is the same. The flow is in the critical state. When the Froude number is less than one, the flow velocity is smaller than the speed of a disturbance wave traveling on the surface. Flow is considered to be subcritical (tranquil flow). Gravitational forces are dominant. The surface wave will propagate upstream and, therefore, flow profiles are calculated in the upstream direction. When the Froude number is greater than one, the flow is supercritical (rapid flow) and inertial forces are dominant. The surface wave will not propagate upstream, and flow profiles are calculated in the downstream direction.
The energy flux available from a device of circular cross section diameter D(m) is proportional to the square of the diameter and the cube of the local flow speed as shown in equation 1.
If two devices are hydrodynamically similar, then the value of fact will be the same!
Similarly if the ratio of gravitational and inertial forces is the same for two devices, they will share a common value of Froude’s number, as defined in equation 2.
L is a characteristic length, which might be the system diameter or, more usually the depth. In the case of a tidal current device, the ratio of depth to diameter would also need to have the same value for two devices if the systems are to be in a state of similarity.
Hence, for a full scale device of diameter DFS, operating at a full scale flow speed of UFS and a “model” device of diameter Dm in a flow speed of Um, the following conditions must be upheld for the devices to be hydrodynamically similar.
p is the power scale factor
Algebraic manipulation reveals that Froude and power similarity will be achieved if:
Hence, consider for example a device rated at 1MW in 3m/s current, which has a nominal diameter of 14m. If this were to be modelled by a device with a power output of 150kW, then the value of p would be 0.15 and the corresponding flow speed would be 2.29m/s. The corresponding system diameter would be 8.14m. If the full scale water depth was 40m, then the model water depth would need to be 23.26m! If these values are not adhered to, it will not be possible to make deductions about the performance of the full size devices from measurements on the model device!
The capital cost and time involved in developing a mechanical testing facility is substantially reduced compared to a grid connected facility. There is a range of options that would affect the accessibility of the proposed sites, from providing piers which would allow land-based access, to relatively simple mooring points on the seabed accessed by boat only.
Depending on the extent and type of mooring, the hardware for a basic facility for the wave site could be installed for in the region of £40,000, and approximately £60,000 for a tidal site. This would exclude the costs of data collection associated with specific site selection and consenting. Note that these costs are indicative only, and would vary depending on the detailed specification of moorings and other options.
A crucial element of the recommended development protocol is that extensive comparisons of the results between the scales are made at the conclusion of each phase change. This procedure serves five main purposes:
• it verifies the smaller, previous model results;
• it validates the application of the similitude law to wave energy device development;
• it highlights if weaknesses exist in physical model testing methodology;
• it checks the importance of various physical properties and processes to device behaviour and performance;
• it reduces the uncertainty of the results.
To date there is little practical evidence to support the accuracy of predictions since only a small number of devices have achieved full/large scale sea trials and of those not all followed a progressive modelling path or designs went through significant changes between phases.
Requirements dictated by a strict scale approach:
The purpose of adopting a scale model is to observe the interaction, development and phenomena of interest at prototype scale (i.e. full scale) using a smaller model scale representation. In order to correctly model two geometrically similar systems, the ratio of inertial force to the individual force components in the prototype system must be the same as the ratio observed in the model system at the corresponding location.
This presents difficulty as the various force balances or ratios are impossible to all be maintained by limitations imposed in the case of interest. For instance it is not possible to substantially vary the force of gravity acting on the system, or the properties of the fluid medium. Hence a hierarchy of importance is necessary to determine which scaling laws must take precedence. In the case of development of a tidal energy capture/converter device, the Froude number becomes the governing parameter as gravitational and inertia forces dominate.
Froude number, Fr =
Where V = representative velocity
g = gravitational acceleration
L = representative length scale
Froude scaling is the de facto approach adopted in similar technology scale testing exercises (e.g. hydraulic structures, hydraulic turbines and pumps, ship resistance).
The Froude scaling parameter of various physical quantities is summarised in table 1
(k is the scaling factor, density in the prototype and model assumed to be equal)
A wide range of data may be gleaned from scale model tests of early-stage tidal stream energy devices. Such information can be valuable to both the device developers and to potential funding bodies that need to make informed decisions regarding which machines have potential for commercial viability and are, therefore, worthy of further development. The methodology, which will be used at any selected sites, is designed to be simple and cost-effective and will provide results in a standardized format, in order to aid the decision making process.
To be given in sufficient detail so they can be traced easily through the library.
For material you do not want in the main body because it is too detailed or peripheral or would interrupt the flow of the report.