Wave and tidal power
Barriers to tidal power: multi basin plants21 April 2004
In part three of a series of articles on tidal power plants, E Van Walsum takes a look at the costs associated with such developments and examines the environmental issues associated with multi-basin plants
During the extensive tidal power studies, carried out in France, the UK, Canada, the US, Russia, Australia and other locations, much attention was paid to double- and multi-basin schemes in the hope that such schemes could economically produce energy on demand or continuous, base-load energy. The nature of a tidal power plant (TPP) however is that of an energy producer, like that of a run-of-the-river hydro plant. To force a TPP to produce energy on demand has proven to be an exercise in frustration. In addition to the above, to produce energy on the demands of a sun-oriented economy, the TPP’s operation would have to be manipulated on a daily basis to overcome the sun-moon timing complications. The environmental dictum that any TPP should be operated in a consistent manner, day in day out, rules out any multi-basin plants operated to produce energy on demand.
From various tidal power schemes, studied in different countries, a consensus developed to the effect that a single basin scheme is most suitable for extracting the maximum amount of energy out of a given site at the lowest unit cost. Thus single basin schemes and combinations thereof, so called ‘paired basins’, appear to be favoured by common wisdom in tidal power circles as of the start of the 21st century: Bernshtein(1997), p. 287, 293, Clark(1993) p. 2653, Baker(1991) p. 33 & 34.
All authors, rejecting double- and multi-basin tidal power schemes in favour of single basin schemes did so on the basis of multi-basin schemes being designed and operated to produce energy on demand or base load power plus fuel-displacement energy. None of the double- or multi-basin schemes considered were to be designed and operated to yield energy at the lowest possible unit cost.
It was subsequently established, as described in Part 2 of this series, that a special double-basin scheme, a ‘linked-basins scheme’, can be designed and operated to produce tidal energy at a lower unit price than single-basin plants at the same location if the geographic setting is favourable. Moreover, energy with such a linked-basins plant would come in four blocks of energy per moon-day of 24h50min rather than two. This means that for future tidal power studies, such linked-basin schemes should be considered where suitable site conditions exist.
Having concluded that for double-basin TPPs an exception exists to the general perception that single-basin TPPs are the way of the future, the question arises if a similar exception might exist for multi basin schemes. A multi-basin plant, operated to produce continuous, base-load energy plus the maximum possible amount of fuel displacement energy could be operated in a consistent manner, day in day out. Such a plant therefore could, at, least from am environmental point of view, be acceptable.
Multi basin TPPs
In France, with its large number of nuclear plants producing energy 24 hours per day with little flexibility to respond to peaks and valleys in demand, tidal power as an additional source of energy seemed to make sense only if it could supply to the utility system the flexibility it needs. As a result, the large tidal power (TP) potential on France’s Atlantic coast has challenged engineers to devise schemes to get energy on demand out of the ocean tides. The Severn Estuary in the UK has similarly inspired the British engineering community.
Consider for example the ‘Caquot-Defour cycle’ as described with thirteen other multi-basin schemes in Bernshtein(1997). This scheme requires a high basin, a low basin, a middle basin and a power house with a forebay and afterbay. The single-effect (SE) power plant would generate energy by flow through the turbines from left to right. At the upstream side of the power house, the forebay would be connected to the middle basin, the upper basin and the sea by means of sluice gates. At the downstream side of the power house, the afterbay would be connected to the middle basin, the low basin and the sea. The dams separating the high and the low basins from the sea would also be fitted with sluice gates by means of which the water levels in these two basins would be kept as high and as low as possible, respectively.
When considering the operation of the Caquot-Defour cycle, it is to be realised that water passing through a sluice causes energy loss due to friction and turbulence, but more relevant in this instance is the fact that, irrespective of a sluice’s hydraulic efficiency, a head across the sluice is required to force the water through. One of the Fundy studies of a single, high-basin scheme under mean tidal conditions shows that for a plant, designed for energy production at the lowest possible cost per kWh, the head loss to be expected across the sluice gates when filling the basin amounts to approximately .22m which in this instance was equal to 5% of the minimum generating head and 2.5% of the maximum generating head.
In this instance there was ample room for sluices so that the optimum number could be readily installed. Thus with water passing through sluices once, we expected to loose about 3.5% of the available head, i.e. 3.5% of available energy. With the Caquot-Defour cycle, we may expect a similar 3.5%loss of head for water passing from sea to high-basin or from low-basin to sea. The situation around the power house with its fore and afterbays is however a bit more crowded. Giving the scheme the benefit of the doubt, let us assume that passage of water from sea to forebay or from after-bay to sea causes also a head loss of 3.5%. The number of sluices between forebay and high-basin, forebay and middle-basin, afterbay and low-basin and afterbay and middle-basin should be 3/7th of the number of sluices between fore-bay and sea and between after-bay and sea.
Hence, if the head loss from sea to forebay is 3.5%, then when forcing the same amount of water through only 3/7th of the number of sluices will cause a loss of total available head of 7/32 x 3.5% = 19.1% as stated by:
If the same amount of water Q has to pass through sluices of which F is reduced by a factor 3/7, then H will of necessity have to be increased by a factor (7/3)2).
In the UK, the Central Energy Generating Board (CEGB) suggested in 1974 a variant of a linked basin scheme, Bernshtein(1997)p.256. A tidal basin was to be created by constructing a dam across the Severn estuary from Watchet on the south shore to a point just west of Barry on the north shore, a distance of approximately 24.5km. This dam would include a power plant with double-effect turbine generators. If the project’s construction were stopped right then and there, the maximum amount of energy would have been captured at or close to the minimum cost per kWh with energy being delivered in four blocks per moon-day.
To improve on the timing of delivery of that energy, it was suggested to create within the initial tidal basin a ‘high’ and a ‘low’ basin. For that, a U-shaped dam was to be constructed in deep water in the centre of the initial basin with a total length of about 96km, including a power plant with turbine generators with pumping capability. While it is easy to put the words ‘high basin’ on a sketch, it would be difficult to get truly high water levels in that basin. There is here a built-in conflict of interests between the DE plant in the initial closure dam and the SE plant with pumping capability between the so-called “high” and low basins. This would result in both plants operating most of the time at less than there design capacity. Here we have not only added dikes and sluices, but even doubled up on the most costly part of a TPP, its power house. And in addition to all that, a pumping plant was to be provided between the newly created low basin and the sea. Such pumps would have to be of a size and price, comparable to that of the turbine generators. Each of these plants would get their turn to produce or use energy at their most appropriate time, resulting in a low load factor for each of the plants.
It is quite evident that the special kind of linked-basins plant as suggested for the Severn had to be abandoned as being uneconomical.
In all of the linked-basins plants considered so far, when proceeding from construction Phase I, i.e. the construction of a single-basin Type X plant to a linked-basins plant, an additional tidal basin was created by building an extra dam with sluices as Phase II. Such an additional basin added appreciably to the potential energy generating capacity of the linked-basins scheme. On top of that, the power plant instead of producing two blocks of energy per moon-day would produce four such blocks, almost doubling its load factor. Under favourable geographic conditions, such linked basin plants were shown to have good economic potential. Proceeding from a simple linked-basins plant to a more elaborate scheme appears to be counter productive.
The cost of TPPs
All aspects of TPPs, including cost, are extremely dependent on local conditions. With the local environmental concerns in mind, it is likely that we would look for suitable sites to build single-basin, DE plants in regions with extremely large tides, i.e. mean tidal ranges of 9m and up. In areas with moderate tides, our first preference would likely be the construction of linked-basin TPPs, designed for maximum energy production. The second choice would then be single-basin TPPs, SE or DE. Taking advantage of local features such as bays and islands would go a long way in keeping the cost of dikes to a minimum.
When designing a TPP, it is to be kept in mind that it will have to be built in an ocean- tidal environment. Rather than fighting that environment, we should aim at taking advantage of that environment while avoiding its punishing aspects. This will in many instances translate in the prefabrication of the power house and sluice structures and , during favourable weather conditions, floating them into place in segments such as concrete or steel caissons. The choice of caisson design would depend to a large extent on the facilities available or to be created locally for their construction.
For single-basin SE TPPs and linked-basin plants, extensive sluice structures will be required. For these types of plants, simple flap gates will be adequate for automatically filling or emptying the TP basins when required. The sizes of flap-gates required will be large and it has been argued that large flap gates can not perform satisfactorily in an ocean environment where wave action would cause uncontrolled slamming of the gates. In addition, the discharge performance of flap gates is often poor and their capacity to resist reverse water pressure can become a limiting factor.
For vertical lift gates, the cost of hoists and controls to operate the gates amounts to approximately 40% of the total cost of those gates. By eliminating the hoists we should not only be able to save in cost but also assist in simplifying the day to day operation of the plant.
Another costly element in the construction of TPPs is armour stone. Due to the substantial tidal ranges to which such dikes are exposed, 75% of the cost of dikes is in the armour stone for dikes in shallow tidal waters. In deep water, this percentage comes down to 62%. This indicates that, instead of placing thousands of pieces of armour stone individually, it might make sense to build the dikes up from prefabricated, ‘blind’ concrete caissons, thus placing sizable sections of dike in one float-in-and-sink operation. In most instances, power house and sluice caissons will already be required so that the support systems like tugboats and other marine plant would be on hand at the construction site. By adding blind caissons for dike construction to the construction programme would tend to standardise all TPP construction operations which would have cost benefits.
The cost of fish-friendly turbine generators
Putting a realistic price on turbine-generator equipment, meeting the sustainability requirements, was not as straightforward as the author had hoped. He requested quotes from two manufacturers for the cost of double-effect machines, fully supported by cantilever action from a central pier without any additional supports, no distributors, runners of 7.5m & #8710; and variable pitch blades or other means of achieving double regulation, and generators with a limiting output of 19MW, operating at slow, variable speed. The rated head was to be 5m, rated discharge 613m3/sec, minimum generating head 2.5m. The quotes received did not quite meet the performance requirements called for. Both suppliers quoted on single effect machines and maintained the distributor as a part of their machines. There were also other minor deviations from what was asked for. As a consequence, the quotes received were increased by 10% to bridge the still remaining gap to environmental sustainability. For double effect machines, the price was increased by another 10%.
Thus the price per single-effect, fish-friendly turbine generator came to $12,375,000 (Can.$’76) A substantial increase over the price, used in BFTPRB (1977) of $8,742,000.
The price for double-effect, fish-friendly machines, came to $13,612,000 (Can.$’76), compared to the price used in BFTPRB (1977) for DE machines of $9,482,000 (Can.$’76).
The cost of sluice caissons with automatic flap gates
The BFTPRB(1977) report listed for the direct cost of sluiceways at Shepody Bay an amount of $4.5M per sluice of 12.2m x 12.2m, including caisson, gate, hoists and controls. The author estimates that of this amount, approximately $1.84M was allocated to the gate proper with its hoists and controls. Of this cost, 30% was for installation leaving for the cost of materials, including hoists and controls, $1.29M. Of this, the cost of gate hoists and controls was 70% or $0.9M. When using a sluice, equipped with automatic flap gates, hoists and controls would no longer be required so that the cost per sluice was estimated to be $4.5M - $0.9M = $3.6M. This cost is therefore based on the assumption that a fully automatic, large-sized flap gate, to operate safely in an ocean tidal environment, will be successfully developed.
For all other construction costs in the Fundy environment, use was made of the well presented cost data of BFTPRB (1977).
In a promoter’s vernacular, the main barriers to be cleared away from the road to TP development are environmental concerns and cost. In order not to waste time and effort on schemes which are doomed to be discarded on environmental grounds, environmental concerns, specific to a given location, should be identified at an early stage to serve as a guide towards achieving the goal of environmentally sustainable tidal power.
In the realm of cost, clever ideas to improve a TPP’s performance should be approached with a cost conscious skepticism.
TP should be accepted for what it is, a source of energy, not of power-on-demand. Its energy comes at the rhythm of the movements of its primary energy source, the moon with its gravitational pull. Forcing a TPP to produce power on demand is contrary to its very nature. It would require the daily manipulation of its operating procedures, resulting in unnatural and erratic tidal regimes within its TP basin(s) which is environmentally unacceptable.
Under most conditions, the maximum amount of energy which can be extracted from a tidal power basin at the lowest possible unit cost of energy will be by means of a single, high-basin plant, separated from the sea by a dam, equipped with a power plant and sluices. The basin would be filled up during high tide through its sluices and its turbines acting as orifices. Energy would be generated during the outgoing tide by single effect (SE) flow from the basin through the turbines into the sea. Such a plant would deliver its energy in two blocks per ‘moon day’ of 24h50min.
In areas with extreme tides, such as at the Bay of Fundy, a single-basin TPP operating in double effect (DE) may be preferred for environmental reasons. In such a plant, the tidal basin would be separated from the sea by means of a dam, equipped with DE turbine generators. The number of such turbines would be much larger than for the comparable single, high-basin plant. All sluices however would be eliminated. Water would flow at the rhythm of the natural tides in and out of the basin through the turbines, generating energy on both the incoming and outgoing tide. This type of plant would deliver its energy in four blocks per moon day.
If the energy produced by a single high-basin plant would prove difficult to absorb in the surrounding utility system(s), a single-basin, DE plant might be preferred for that reason.
Double-basin TPPs - Paired-basins
In the event the development of a certain area’s tidal power resources began with the construction of a single, high-basin plant, a logical next step would be the construction of a single, low-basin plant. Such a plant would be constructed like a single, high-basin plant but with the difference that the turbines would operate in SE by flow from the sea into the basin and the sluices would be used to keep the plant’s basin level as low as possible at all times, releasing water from the basin into the sea. Such a plant would also deliver energy in two blocks per moon day, but during the incoming tide. Thus with such a ‘paired basins’ scheme, energy would be obtained during both incoming and outgoing tide.
Double-basin TPPs - Linked-basins
In certain geographic locations it will be feasible to separate two contiguous tidal basins from each other by means of a dam, equipped with SE turbine generators. One of the two basins would be chosen as the high, the other as the low-basin. Energy would be generated by flow through the SE turbines from the high into the low -basin. The high-basin would at all times be kept as high as possible by flow through filling sluices from the sea into the high-basin. Similarly, the level of the low basin would be kept as low as possible at all times by flow from the low-basin through its dewatering sluices into the sea.
In the past, linked-basins were perceived as producers of energy-on-demand or of continuous, base-load energy. For a linked-basins plant to operate in an environmentally acceptable way, its operation should be consistent from day to day, without exception. To keep the cost of the energy generated down and to comply with environmental requirements, the linked-basin plants of the future should be operated with the objective of producing the maximum amount of energy at the lowest possible cost. Provided favourable natural geographic conditions exist, linked-basin TPPs can produce four blocks of energy per moon-day at a cost below that of a paired-basins scheme at the same location.
Linked-basin plants would be constructed in two Phases, Phase I consisting of the construction of the high basin as a single, high-basin plant. Upon completion of Phase I, this plant would start producing energy on the outgoing tide while construction of the second Phase is carried out. This Phase II would consist of the construction of an additional dam, equipped with dewatering sluices, thus forming the low basin. This second Phase would thus add to the total amount of impounded tidal waters, increasing the plant’s energy generating potential without adding to its power plant. After completion of Phase II, the power plant would do double duty, generating energy during both the incoming and outgoing tides. The amount of energy produced by the linked-basins plant as compared with that produced during Phase I by the single, high-basin plant depends to a large extent on the amount of sluicing capacity provided.
In areas where siltation is an dominant environmental concern, a ‘flow-through layout’ should be considered for the TPP. With such a layout, water flows into a basin at one end and out of the basin at its opposite end. With such a layout, the areas where silt settles down during periods of stagnant flow are swept clean as soon as the next stage of power generation or sluicing starts. This flow through layout can be applied to single high-basin, single low-basin and linked-basin plants. It can not be successfully applied to a single-basin, DE plant since in that case the generating capacity would have to be doubled.
Any TPP with three or more basins can not compete as an energy producer with the plants of simpler layout and should therefore be avoided.
For additional conclusions, phrased in terms of broad policy objectives, see those listed in Parts 1 and 2. The most significant of those, in terms of action aimed at achieving those objectives, is the requirement of a pilot TPP program. The implementation of such a programme should be initiated by the parties in ultimate control of the development of a nation’s TP resources. As for the Bay of Fundy, those parties are the Governments of Canada and the provinces of New Brunswick and Nova Scotia. Interpreting the results of such a pilot plant program would assist in defining the ecological rules under which TP could be developed.
For the engineering and environmental aspects of this paper, the author was indebted to engineers and environmental scientists alike who, each in their own way, communicated their views to others. Some of those are listed, directly or indirectly, under References.
For parts 1 and 2 of this article, please see the September 2003 and October 2003 issues.
In memory of Ewout Van Walsum, consulting civil engineer, who sadly died on Monday 18 August 2003. Memorial donations can be made to: The Saku Koivu Foundation, c/o Montreal General Hospital, 1650 Cedar Ave, Room E6129, Montreal, Quebec, H3G 1A4<