A new solution may double the worldwide potential of tidal energy and half its cost

Printable version Printable version

Published in: Hydropower and Dams, Issue 1, 2014

by F. Lempérière

Hydropower and Tidal Energy have about the same theoretical potential. Hydropower supplies 3 500 TWh/year, Tidal Energy 1 TWh/year. The reason of this gap may be that the technical solutions used successfully for hydropower and chosen for most studies of tidal energy are poorly adapted to most tidal sites. A new specific solution better adapted to the specific tidal energy is presented below; it may apply to 1 500 TWh/year at a cost close to the hydropower cost, and with a better environmental impact.

– The worldwide Hydropower generation is 3 500 TWh/year and will double.
The tidal power generation is 1 TWh/year.

This difference in success is surprising because:

– The theoretical potential is the same, about 20 000 TWh/year.

– The density of energy is 10 GWh/year/km² for hydropower (3 600 TWh/year for 350 000 km² of reservoirs); for tidal energy the feasible energy supply is 12 GWh/year/km² for a tide range of 4 m and 40 GWh/year/km² for a tide range of 7 m.

– Many hydropower schemes in large rivers operate successfully with a 4 or 5 m head, the same as for the usual operation of the two main tidal plants (La Rance and Shiwah).

– The extra cost for operating in salt water is low.

– Environmental impacts are lower for tidal energy.

We may thus wonder if the technical solutions used successfully for hydropower and studied for tidal energy are well adapted to the very specific conditions of tides which are analysed below. Then a solution which seems much better adapted to these very specific conditions is presented and evaluated.

1. Specific data of tidal energy

– As most world tidal energy is where the tides are semi diurnal, slightly longer than 12 hours and quite as high in a same day, the data below are for such tides but the new solution proposed may apply to all cases.
The range of tides H (in m) varies along 14 days, with some days of Spring Tides when H is 30 or 40% higher than the average value Hm and may reach exceptionally 1,5 Hm and some days of Neap Tides when H may be 70% of Hm (exceptionally 50%). The available energy is thus much different between two weeks. But the yearly and monthly energies are quite constant and linked only with the value of Hm and of the reservoirs area.

– Past studies since 60 years have been made essentially where Hm is over 5 m. The corresponding world potential is a rather small part of the total potential because the corresponding area is limited to some dozens thousands km² and the technically feasible potential under 2 000 TWh/year. And the global potential of few famous places where Hm is over 7 m is few hundreds TWh/year. The potential for Hm between 3 m and 5 m is much higher: the potential per km² is lower but the area is hundreds of thousands km²; the feasible potential is over 5 000 TWh/year and may apply to 20 countries.

– During an half tide of six hours, the level in a tidal reservoir is during some time quite the same as the sea level. It may then be impossible to produce much power along one or two hours.

– For a same low head, and for a same power the cost of the civil engineering of a tidal plant is much more important than in a river because the plant height has to take in account the waves height and the total range of Spring Tides.

– In areas where there are significant tides the conditions of foundation of dykes or plants are usually favourable because the depth is 10 to 20 m under the sea level and the soil is rock, sand or gravel. But the waves may be significant and reach well over 5 m. The foundation and the floods are the key problem of design and construction for dams, the waves a key problem for tidal energy.

– The environmental impacts from tidal plants are very different from the impacts of hydropower. Tidal impacts may be better but the possible impacts on natural conditions and especially on biodiversity may prevent now the utilization of some tidal sites (such as estuaries) or of some operation methods. This point which was forgotten 50 years ago is essential for future choices.
A key point is thus that the conditions of tidal energy are very different from the conditions of traditional hydropower: the solutions may not be the same.

2. Present solutions for tidal plants

All studies have been based upon the same principle as for Hydropower in rivers: to create a reservoir by dams or dykes and to use the created head through turbines placed in a concrete structure.

2.1. The reservoirs (basins)

Various solutions have been studied associating hydraulically several basins; this may improve the utilization of turbines; but these solutions may be unacceptable now for environmental reasons because the relevant conditions of tides along the shore are far from the natural ones.
Simple basins may be:

– Estuary basins such as La Rance. They avoid the cost of dykes for closing the basin but environmental problems are more difficult, especially with salt content. Anyhow there are rather few large estuary sites worldwide.

– Artificial islands which avoid impacts along the coastline but the cost of long dykes increases too much the cost per kWh except for very large islands which should be far enough from shore for avoiding sedimentation problems. It is difficult to find such cost effective sites.

– The main potential is thus essentially for large single reservoirs along shore but their impact should be acceptable and thus the tidal range and levels in the reservoirs as close as possible to the natural ones.

2.2. Usual turbines

A specific solution of turbines with horizontal axis, the bulb unit, has been developed 60 years ago for the tidal plant of La Rance in France. It may operate both ways and may also pump. The power of a turbine may reach 30 or 40 MW. And this solution has been used successfully in rivers with most often heads of 5 to 10 m; but the possible power supply is much reduced when the turbine is used with a lower head.
In La Rance the power supplied for a 3 m head is 30% of the rated power and is very low for a 2 m head. It is possible to design plants for operating with 2 m head but the power per m of plant is reduced to few hundred kW when the concrete structure remains important and costly; and the corresponding cost per kW of the civil engineering is then very high.

2.3. Operation of existing solutions

– A reservoir (basin) may be operated one way or both ways.
– The both way operation of tidal energy is as per fig.1 and has 4 advantages:
. Power is supplied 8 hours from 12.
. The operating head is about 0,35 H but the volume through turbines in these 8 hours is 2 x 0,9 x H x S and the energy available

. The yearly energy is in the range of 3 500 hours of the rated power.

. The tides within the basin are very similar to the natural ones, which is very important for the biodiversity. Tides are simply shifted by 2 hours.

But the operating head of 0,35 H is for mean tides 2,5 m for best sites and 1,5 m for most sites and the power supplied by a bulb turbine rather low and costly

– The one way operation is as per fig. 2 (Shiwah); the turbined volume is about 0,7 SH under 0,6 H head, the possible energy about

much less than the possible energy in two ways operation.

The advantage is an higher head for turbines operation.

But this solution has 3 drawbacks:

. Operating 4 hours from 12 is not well adapted to the power needs.

. Limiting accordingly the turbine utilization provides yearly an energy equal to 2 000 hours of the rated value.

. Modifying the tides levels and range in the basin may not be acceptable for the biodiversity.

Bulb units are thus poorly adapted to the most acceptable and largest utilization of tidal energy and their choice favours the one way operation.

It may be the main reason of the lack of progress of tidal energy since 60 years. Another reason was the low cost of electricity from fossil fuels but this cost does not prevent the present progress of hydropower.

The utilization of tidal plants with bulb units may however be of interest if a part of the investment is paid for by favourable impacts beyond power supply, as it is in Shiwah for environment and if it associated with much thermal power.

– A new turbine design with vertical axis, the orthogonal turbine, has been studied and tested in Russia (fig.3). It may operate both ways with an output of 0,75. The turbine design is quite simple. The power is however limited to about 400 kW per m for an operating head of 2,5 and to 150 kW for 1,50 m. The cost of the civil engineering per kW is thus high for Hm in the range of 6 or 7 m and very high for Hm values under 4 or 5 m.

This solution which deserves optimization and further experiences appears more promising than the bulb turbine and may well operate both ways. It seems however more expensive for most schemes than the solution proposed below, at least for tidal ranges Hm under 5.

3.Present solutions for in-stream turbines

Wind farms are successful onshore and offshore because there are many places with enough wind speed for using cost effective power units of 1 to 5 MW onshore, 3 to 10 MW offshore. This success has favoured the study and experimentation of the same principle applying to water streams for which the water speed is significant i.e. in areas with high tides. The power supplied by an in-stream turbine is about (in kW): 0,2 sV3, s being the turbine area in m² and V the water speed in m/s.

The diameter of a turbine may be 12 to 20 m, its area 100 to 300 m². For a huge diameter close to 20 m and thus an area of 300 m², the power, in kW, is 60 V3, i.e. 0,5 MW for 2 m/s and 1,5 MW for 3 m/s. There are rather few places where the water speed is over 3 m/s during 1 000 hours per year and over 2m/s during 3 000 hours; there is thus a rather small potential for units over 1 MW supplying over 2 GWh/year (when an offshore wind unit may supply 15 GWh/year).

The cost per kWh for placing, connecting to grid, operating and maintaining such turbine in open sea is usually much more than the cost corresponding to manufacturing it. And the yearly energy will be only 2 GWh/year. There is thus little world potential at an acceptable cost. Another drawback is that, along 14 days, most power supply is within 4 days of spring tides and there is little power supply during one week from two. The future of this solution in natural conditions is limited to some exceptional places; the feasible world potential may be 500 TWh/year but the cost effective potential seems less than 100 TWh/year.

Actually In-stream turbines should be very cost effective if they could operate most time with a permanent speed of about 4 m/s in favourable marine conditions. There are no such places naturally. The principle of a new solution is to create such places artificially.

4.Principle of a new solution

– For well meeting power requirements and environmental care the best operation of a reservoir is both ways with thus an average water head under 40% of the mean tidal range Hm, i.e. 2 or 3 m for best sites and 1 or 1,5 m for most sites. Bulb units are hardly cost efficient under 3 or 4 m and orthogonal turbines under 2 m.

– A line of in-stream turbines operating with a water speed of 4 m/s uses a water head of about 0,10 m. As example, turbines of 16 m diameter spaced 25 m between axes, in a place 25 m deep, under a flow of 4 m/s supply 0,2 sV3 i.e. about 0,2 x p/4 16² x 43 # 2 500 kW and use about using 3 000 kW if the output is 0,8.

The corresponding flow is 25 x 25 x 4 m/s = 2 500 m3/s and the water head used for a line of in-stream turbines is:

– 20 lines of In-stream Turbines would use a total head of 2,4 m corresponding to the advisable head for best sites. If the lines are spaced by a length of 5 diameters, i.e. 5 x 16 = 80 m, the total length will be 1 600 m.

The principle is thus to catch the tidal energy through in-stream turbines in artificial long channels where a speed chosen initially is kept most time. A value of 3,5 or 4 m/s seems advisable.

Such channels may be obtained through creating by a long dyke a large reservoir along shore and opening the reservoir to the sea by channels including 10 to 20 lines of in-stream turbines (fig. 4 and 5).

– The length of the channel may be 1 600 m for a tidal range of 7 m and an operating head of 2,5 m. It would be reduced to 1 000 m and 10 or 12 lines of turbines for a tidal range of 4 m and a head of 1,5 m.

5.Data of the new solution

Instead of few in-stream turbines in a natural place, many turbines are associated in an artificial place for a better production, this specific solution may justify a specific name such as Tidal Gardens (or T.G.).

– A site for Tidal Gardens (fig.4) is a large basin open to sea by 1 to 2 km long channels in which are placed 10 or 20 lines of in-stream turbines (Green Plants!). The area of the basin may be hundreds km² or possibly thousands of km² with about one channel per 100 km². Smaller basins may be used with one channel. Most future sites will be along shore. A typical basin may then be an half circle along shore.

– A channel (T.G.) linking the basin to the sea is represented per fig.5.

Its length will be according to the mean tide range and turbines data. Its width may be in the range of 500 m for very large basins, 100 or 200 m for small ones.

– The depth may be 15 to 20 m under the low sea level; this may require some dredging or filling. For accepting a significant water speed, the bottom will be lined, for instance by 0,50 m of concrete placed in calm water.

– The channel sides will be limited by dykes 25 m high with low differential head and very reduced waves impact.

They may be as per fig.6.

– The channel is separated from sea by gates to be opened about 4 hours within a 6 hours half tide. The differential head on gates is rather low but the waves impact may be high. Solutions similar to gates for dams spillways may be used but the specific conditions may favour specific solutions for the construction method. Specific designs are also possible.

– For the dyke closing the basin recent progresses in breakwaters designs and dredging efficacy will favour a solution as per fig.7 well adapted to an optimal program of large schemes.

6. Operation of the new solution (T.G.)(fig.1)

It includes 3 phases along a 6 hours half tide.

– After the time when the basin is at the sea level, the gates of the channel are closed during 1 or 2 hours for creating a gap of 1 or 2 m between sea and basin; no power is supplied.

– Along 3 or 4 hours of main power supply, the number of open channels and of operating in-stream turbines is fixed according to 2 targets: to use at best the available power and to keep in the channels the optimum speed for the best turbines efficiency. This speed which is chosen in the design may be obtained permanently by adjusting the number of turbines in operation according to the differential head between sea and basin. When a channel is fully open, and if no turbine is operating the water speed is close to 6 or 8 m/s; according to the number of turbines operating and thus using the energy through the channel the water speed is reduced to the optimum value for turbines. The power supplied by a channel is thus roughly proportional to the differential head and the total energy supplied is according to the width and number of channels fully open. It is thus possible for any differential head to use the energy and the turbines at best with the fixed value of water speed optimal for turbines efficiency.

– Along one hour, the head is lowering from about 2 m to nil, the number of operating turbines is reduced progressively and the water speed in the channel is kept close to 4 m/s till few minutes before an equal level in basin and sea.

7. Impacts of tidal schemes

Many studies for tidal energy were made fifty years ago when environmental impacts were overlooked; most corresponding proposals should thus be unacceptable now.

The impacts should be studied for modern designs which take in account environmental problems. They are analysed below for large schemes operating both ways; a large basin is open to sea by long and wide deep channels where are placed the in-stream turbines. The water speed in the channels is close to 4 m/s.

There are three impacts: visual impact, environmental impact, socio-economical impact.

They should be compared, for a same energy, with impacts of other Renewable Energies.

7.1. Visual impacts

The tidal plants cannot be seen because they are underwater in-stream turbines.

The dykes are about 10 m above the sea level. Most are 10 to 20 km from shore and hardly seen. The links with shore may be used for tourism and fishing harbours (fig.8).

For a same energy, the visual impact is much less than for dams, for onshore and offshore wind farms, for solar energy.

7.2. Environmental impacts

It is a key point to be studied very carefully:

– With the two ways operation, the tides in the basin are quite the same as natural tides. The range of tide is reduced by 10%, but may be even kept the same if some pumping facilities are added at little extra cost (see chapter 9 below).

– The waves along shore are much reduced.

– The movements of sediments generated along shore are much reduced.

– The salt content of water is unchanged.

The environmental impacts on shore and close to shore may thus be more favourable than unfavourable. Reducing the natural sedimentation of gulfs may be very useful.

– The impacts close to the dyke and to the channel are unfavourable because the sea bottom will be modified by the construction: moreover sand brought through the channels during tempests will stay within a few km from channels and will require dredging. However these impacts refer to 10 or 20% of the basin area and most of the basin area has less impact from sedimentation than in natural conditions.

– Fishes will cross the channels at a speed of 4 m/s in conditions similar to the places with turbines placed offshore in natural streams. The impact of noise and vibrations of turbines should be checked and turbines possibly optimized accordingly.

All impacts may be checked in preliminary schemes of few dozens km² before undertaking very large schemes.

7.3. Direct socio-economical impacts

– No population will be displaced for such huge investments.

Much employment will be created locally and on sites prefabricating turbines and dykes caissons.

– With proposed construction methods (see chapter 12 below), there will be little disturbance onshore during works.

– In case of accidents to the main dyke by extraordinary storms, there will be no human risks (similar to risks from dams failures) and quite easy repairs.

7.4. Undirect socio-economical impacts

Creating along shore very large calm areas paid by power supply gives extraordinary opportunities of economic development in many countries.

Onshore reduction of waves and sedimentation and of exceptionally high sea levels will favour along shore various developments, including small harbours and sand beaches.

– Calm basins favour fish farming and shipping.

– It is possible to create by dredging large islands along the dyke. These islands may be used for tourism but also in some places for large industrial schemes such as thermal plants, oil refineries, chemical industry, and corresponding large harbours along the dyke.

– The dyke which is paid for by supplied power may save cities along shore from exceptional storms. It may be adapted accordingly.

– It is possible to operate the schemes for reducing by 1 or 2 m the maximum level reached along shore, even where there are very large rivers; this may solve the huge problem of various countries resulting from the general increase of sea level such as in China, Vietnam and perhaps Bangladesh.

7.5. Shipping

– Locks may be created between sea and basins as far as necessary.

– Small harbours may be built along shore.

– Very large harbours in deep water may be created at low cost along the main dyke, using the same caissons and separating them from the dyke (fig.9).

8. Data and costs for an example

Most cost effective tidal sites will have basin areas S between 100 and 2 000 km² and an average tidal range Hm between 3 and 7 m. Rough evaluations of power and cost are made below for S = 500 km² and Hm = 5 m.

8.1. Physical data

During an half tide of about 6 hours, it is possible to keep the gates closed along 2 hours and to discharge in 4 hours most of the tidal range H. It is possible to discharge 75% of H under an average head of 0,45 H or 90% of H under an average head of 0,35 H. This last solution is possible with the chosen technical solution and it keeps the basin with a tidal range close to the natural one. Calculations are made accordingly.

The discharged volume is 0,9 x 5 m x 500 x 106 and the flow may be quite constant along 4 hours with a flow of

In order to get the best utilization of turbines, the water speed in the channel is kept quite constant, for instance 4 m/s with a channel depth close to 20 m and a total width of channels of , such as 5 channels 400 m wide.

The channel flow will remain quite constant but the differential head between basin and sea will be about 2 m along 2 hours and lower along two hours. The number of lines of turbines in operation will be adjusted to this head.

During Spring Tides, the gates will be opened 5 hours instead of 4, the average head will be slightly higher but the water speed will be kept close to 4 m/s.

During Neap Tides, a channel may remain closed, the operation may be limited to 3 hours and the water speed in channels kept close to 4 m/s.

8.2. Power evaluation

Along an half tide, a volume of 0,9 x 5 x 500 x 106 will be used under an average head of 0,35 x 5 m; for an output of turbines of 0,85 and an hydraulic loss of 25% in the channel, the energy is :

and, for 705 tides per year, an annual energy of 7 x 2 x 705 # 10 000 i.e. 10 TWh/year.

Along an operation of 4 hours the flow is quite constant but the head, i.e. the power supplied is lower along 2 hours and the necessary capacity is about 7 GWh/3 hours = 2,3 GW. This capacity should be increased by about 20% for flexibility and a better utilization of Spring Tides, i.e. 2,3 x 1,2 # 2,75 GW.

An in-stream turbine supplies (in kW) 0,2 x s x V3, s being the turbine area and V the water speed. For a diameter of 16 m (s close to 200 m²) and a water speed of 4 m/s, the capacity is 0,2 x 200 x 43 # 2 500, i.e. 2,5 MW. The site requires 1 100 turbines of 2,5 MW, i.e. 220 turbines per channel 400 m wide.

These turbines may for instance be in lines of 16 turbines (25 m between axes) spaced by 5 x 16 m = 80 m, i.e. 14 lines and a length of channel of 14 x 80 # 1 100 m.

8.3. Costs

The cost includes 3 parts: turbines, channels and main dyke.

8.3.1. Turbines

For a large number of units placed in calm water close to an electric substation, the cost should be close to the cost of a wind plant of similar design and same power. A figure of 1 200 €/kW, i.e. 3 millions € per turbine is used below. This cost should be increased by 15% for financial costs during construction, i.e. increased up to 1 380 €/kW. With 7% per year for depreciation and interest and 3% for operation and maintenance, the cost per kWh is

This value does not vary with tidal range Hm or with basin area S.

8.3.2. Cost of the channels

The total area of the channels is 2 000 x 1 100 = 2,2 millions m² for 2,75 millions kW, i.e. 0,8 m² per kW. The total length of channels is 5 x 1 100 = 5 500 m.

The channel cost includes 3 parts:

– The cost of bottom concrete, 0,50 m thick at a cost of 200 €/m3, i.e. 0,8 m² x 0,5 x 200 = 80 €/kW

– The cost of dykes, i.e. 2 dykes (as per fig.6) using 40 m3 of reinforced concrete per m at a cost of 700 €/m3 for a total dykes length of 2 x 5 500 = 11 000 m, i.e. 11 000 x 40 x 700 # 310 M€ / 2,75 x 106 # 110 €/kW to be increased to 120 €/kW for taking in account some extra dykes length in basin.

– The cost of the gates closing the channel with a 22 m height and 5 x 400 = 2 000 m length, i.e. 44 000 m² at a cost of 10 000 €/m² or 440 M€/2,75 # 160 €/kW.

The total cost for the channels is thus 80 + 120 + 160 = 360 €/kW to be increased by 25% for unforeseen and miscellaneous, and by 20% for financial costs during construction i.e. an investment of 360 x 1,25 x 1,20 # 530 €/kW.

For 7% per year of this value for investment and 1% for operation and maintenance, the yearly cost per kW is 43 € for 3 600 hours per year, i.e. 12 €/MWh.

The cost does not vary with S. It may vary (gates) from 10 € for Hm = 7 m to 15 € for Hm = 3 m.

8.3.3. Cost of the main dyke

For a 500 km² area, it may be evaluated as the cost of an half circle dyke along shore. The diameter is 35 km and the length of dyke 35 x p/2 i.e. 55 km.

The cross section may be as per fig.7 which associates a proofed breakwater (such as for the recent Tanger Harbour) with a wide dyke built by dredging in calm water supporting the rather low differential head between sea and basin.

The cost per m may be

Increased by 20% for miscellaneous, unforeseen and studies and by 20% for financial costs during construction, i.e. for 55 km 55 000 x 44 500 x 1,2 x 1,2 = 3,6 billions €.

The yearly cost may be 6% for capital costs and 1% for operation and maintenance, i.e. about 250 millions for 10 TWh or 25 €/MWh.

The evaluation of 25 €/MWh of the main dyke cost is for Hm = 5 m and S = 500 km².

– It varies as (5/Hm)1,5 because for a same area the power varies as Hm² and the cost per m of dyke very roughly as ÖHm.

– Accordingly, for S = 500 km², the cost of the dyke is 25 €/MWh for Hm = 5 m, 20 for Hm = 7 m, 35 for Hm = 4 m.

It varies as (500/S)0,5 because, for a same shape of basin, the power is multiplied by 4 when the dyke length is multiplied by 2.

The costs above are for a straight shore, they could be much reduced for a more favourable shore topography such as a gulf.

8.3.4. Cost comparison with other solutions

The cost as evaluated above includes 3 parts:

– The cost of 38 €/MWh for turbines, based on a cost of turbines of 1 200 €/kW. This cost of 1 200 €/kW for hundreds 2,5 MW turbines placed easily in calm water with short electric links should not be much different from the cost of onshore wind mills of similar unit power.

– The cost of 12 €/MWh for channels varies between 10 and 15 according to the tidal range. The cost per kW of turbines + channels of 1 200 € + 450 € = 1 650 € is slightly higher than the cost per kW of onshore wind mills, but the yearly power supply is about 2 000 hours of the rated power for wind mills and 3 600 for in-stream turbines.

The total cost per MWh of turbines and channels may thus be lower than the cost of onshore wind farms. It is true as well for tidal ranges of 3 or 4 m or 7 m.

The cost of bulb units operating both ways is much higher because the operating head is under 3 m for best sites, under 2 m for tidal ranges of 4 m, the power supplied per m of structure is limited to some hundreds kW when the cost of the civil works of the structure remains high. The cost per MWh may thus be high even for best sites and very expensive for tidal ranges of 4 m.

The cost per kW of a bulb plant operating one way will be lower than for both ways because the operating head is higher but the yearly supply is much less (2 000 hours of the rated power in Shiwah and La Rance).

The orthogonal turbines studied in Russia appear more attractive, designed for operating both ways and with rather simple turbines. Their power supply per m of structure remains low and the cost of civil works per kW rather high for a tidal range of 6 m and very high for a tidal range of 4 m.

– The cost of the main dyke has been evaluated above as 25 €/MWh for a site of 500 km², 55 km of main dyke and a tidal range of 5 m.

For best sites such as the Severn (U.K.), Chausey (France), Fundy (Canada)) where the dyke length is lower compared with the basin area, the cost per MWh may be 10 € and the total power cost 60 €/MWh.

But the new solution has a larger potential with sites of 3 or 4 m tidal range because there are worldwide many sites of such tides range with areas of hundred km² with a sea depth under 20 m. For a tidal range as low as 3,5 m and a 500 km² area the cost of the main dyke may be 25 x 5/3,5)1,5 # 40 € and the total cost 50 + 40 = 90 €/MWh. For such large schemes, the tidal plant operation has a favourable impact on shore where the waves, the exceptional high water level and the sedimentation are reduced and a part of the dykes cost may be paid by these advantages.

The worldwide cost effective tidal energy potential is thus much higher than it has been estimated in the past.

9. Pumping facility may increase the power supply

In the example above, for a mean tide of 5 m, the tide range in the basin is 4,5 m. For getting the same range as the natural one, it should be necessary to pump 0,25 m x 500 x 106 m3 in 1,5 hours i.e. 23 000 m3/s under 1 m head as average, with an output of 0,75 it requires a power capacity of 300 MW.

The cost of relevant plants which may be specific bulb units may be rather high such as 2 000 €/kW i.e. 600 millions € increasing the total investment by 8%. But for an increase of tidal range by 10% the energy supplied is increased by 15 or 20%. This additional investment may thus be very cost effective and keep in the basin exactly the same tide as the natural tide, which reduces possible environmental criticizism. The utilization of these bulb units for supplying power gives also flexibility for the overall power supply.

It may also be possible to use specific units only for pumping at a lower cost.

10. Utilization in the electric grid and energy storage

Tidal power has the advantage of a reliable and easily foreseen energy but has two variations of power supply:

– Along a half tide of 6 hours

– Along 14 days (Spring Tides and Neap Tides)


Within 6 hours,a two ways operation supplies power along 4 hours; it is thus advisable to store 2 hours of average power supply. For the example above with 10 TWh/year, the average supply is 10 000 GWh/ 8 640 hours # 1,15 GW to be stored for at least two hours i.e. 2,3 GWh.

It is possible for most tidal sites to use few per cent of the tidal basin for a Pumping Storage (.P.S.P.) for instance with 2 basins, one operated between 10 and 20 m under the low sea level, the other one between 10 and 20 m above. The stored energy for 2 basins of 1 km² each, is

Storing 2,3 GWh requires 2 x 2,3/0,6 # 8 km².

Such PSP may be built along the main dyke, in the dry between cofferdams, requiring 15 km of dykes.

These works can be made in calm water after closure of the main dyke, i.e. at a cost per km of dyke about 20 millions/km, i.e. 20 x 15 = 300 millions for 1,15 GW, about 300 €/kW.

The total investment of PSP may then be about 1 000 €/kW, i.e. 1,150 billion and a yearly cost of about 120 millions for 10 TWh, i.e. 12 €/MWh. Should be added the 20% loss of power through storage, applying to one third of the power supply, i.e. 20% x 1/3 # 7% x 75 # 5 €/MWh.

The total cost of storage will then be 15 + 5 = 20 €/MWh but the power supplied may be used when required and especially during the peaks of need.

If there are several tidal schemes in a country with different tides timetable, the need of storage will be reduced.

The need of energy storage may be reduced or nil as far as much thermal power is used in a country. The investment of the PSP may then be postponed by decades.

10.2. Power variation along 14 days

Along 14 days, the power supply will be about 60% of the mean supply along 3 or 4 days of Neap Tide i.e. a lack of power of 40% x 80 hours. Corresponding storage for 30 hours may be possible but requires very large areas and huge extra costs. It will be probably most often less expensive to use during Neap Tides some power from hydropower or biomass power or gas or coal power.


The optimum storage time may however be between 2 and 30 hours, such as 10 hours using 8% of the area of the main basin. The cost of storage will be close to 25 €/MWh instead of 20.

This extra cost for storage has huge advantages:

– Power may be shifted from night to day and to peaks of needs.

– The PSP will be used quite permanently through pumps or turbines and may adjust in seconds the power to requirements.

– The PSP may also be used for other energies such as wind. It may be the most cost effective way in a country for storing energy and it may thus be advisable to increase also the PSP capacity beyond the value above up to 1,5 or 2 GW along 10 hours.

11. Association with wind energy

In many countries, it will be cost effective to use the tidal basins for wind energy, placing wind turbines over the dykes and within the basin.

For the example above of 500 km² and 55 + 11 = 66 km of dykes, it may be possible to place over 300 km² 600 turbines of 5 MW (of which 100 over the dykes and 500 in the basin placed and maintained in calm water). Electric links will be short. The investment per kW will not be much higher than for onshore wind farms and the yearly production more important because wind at sea is more important. The yearly supply may be 600 x 5 MW = 3 GW x 2 500 hours = 7,5 TWh/year at a direct cost close to the cost of onshore wind energy.

And the PSP foreseen for the tidal energy storage may be used for the wind energy storage.

In many tidal sites the wind energy may thus be quite as important as tidal energy and very cost effective.

12. Construction methods and schedule of works

For schemes of reduced area such as 50 km², the best solution may be to build in the dry using cofferdams as it was in La Rance and Shiwah.

For schemes of hundreds or thousands km² and dykes of dozens km, it seems preferable to use marine construction methods; it should be underlined that works at sea are difficult and expensive if exposed to waves but are efficient and cost effective in calm water.

The main dyke will be mainly by prefabricated caissons placed when there are no significant waves and by dykes in sand and gravel placed by large marine dredges operating in calm water behind caissons.

The channels works include dykes to be built as above and bottom lining and gates to be realized in calm water: a calm place may be realized by the channel dykes and by some caissons of the main dyke to be used during few years in front of the gates during the gates construction.

The schedule of works may then be:

1. Two years of preliminary works including a small harbour for works which may be used later for tourism or fishing (fig.8).

2. Four years of main works, each channel may be built in two years and several channels may be built simultaneously. Turbines may be brought and placed by marine equipment in calm water with a very flexible schedule. Most of the main dyke is also built during this phasis but a part remains open for keeping the water speed between sea and basin under 2 or 3 m/s.

3. Six months for closing the main dyke, with all channels open and for starting the power supply.

4. The pumping facilities, the PSP and possible wind farms are built after the closure of the main dyke within cofferdams built in calm water. The basis of the wind turbines, to be placed in depth of about 20 m, may be prefabricated within one of these cofferdams at low cost. All theses works may be realize immediately or postponed by decades

Prefabrication of caissons and turbines may be at a constant rate along 4 or 5 years, which is very cost effective.

13. Impact of large tidal plants on the natural tidal range

Tides are a very complex phenomenon and local high tidal range is most often linked to resonance effects. Using in large tidal schemes a significant part of the energy may modify significantly the natural tidal range close to the basin. This variation varies with the generated power and the operation method. It may be over 10% of the natural range and reduce by 20 or 30% the available power. This applies mainly to sites with high tidal ranges and much less where tidal ranges is 3 to 5 m.

The impact of this problem will be thus rather limited with the solution proposed:

– The reduction of power supply may be significant for sites with mean tidal range over 5 m but a reduction in power by 20% will increase the cost per MWh by 10% and these sites will remain cost effective.

– The impact on sites where the average tidal range is 3 to 5 m will probably be low or nil and such sites represent most of the potential.

 14. World potential and possible implementation

The possibility of using large areas where the tidal range is 3 to 5 m doubles at least the world potential and favours the utilization of tidal energy in countries which have not yet studied this possibility.

The turbines, channels and dykes designs may be quite similar in most countries. Experience of some preliminary sites 50 or 100 km² may be completed in 2025 and large sites may be then implemented worldwide.

A realistic potential of 1 500 TWh/year of tidal energy and 500 TWh/year of associated wind energy is likely. It is over half of the present hydropower or the present nuclear energy, for about same cost with more favourable impacts.

15. Conclusion

For a same potential, Hydropower supplies 3 600 TWh/year and Tidal Energy 1 TWh/year.

For environmental reasons the main utilization of tidal energy should be with large basins along shore operating alternatively along a tide with a water level about 2 m above or under the sea level.

The power plants similar to hydropower plants studied since 50 years are too costly for such low water heads.

A new solution opens the basins to the sea by channels 1 or 2 km long in which are placed in-stream turbines operating in optimal conditions. It has 3 huge advantages:

– The cost for sites of high tidal range is much lower than with traditional solutions.

– The cost efficiency applies also to the huge potential of sites with a tidal range of 3 to 5 m.

– The environmental impacts are better than for hydropower; natural tides are kept in the basins but high waves, storms and exceptional high water levels are avoided along shore.

This solution may have a cost effective potential well over 1 000 TWh/year and apply to 15 or 20 countries with very favourable economic impacts.

Corresponding potential and impact in these countries may be analysed in a next issue of Hydropower & Dams. They may be very different from the past studies.


– Scientific aspects of the use of tidal energy – Gibrat 1975

– Sept années d’exploitation de l’usine de La Rance – Cotillon 1974

– New bulb unit technologies for tidal powerplants (Andritz) – Hydropower & Dams 3-2007

– An overview of tidal power potential and prospects – Hydropower & Dams supplement 2009

– The orthogonal turbines (USACHEV – N.I.E.S.- RusHydro) – Hydropower & Dams supplement, 2009

– E.D.F. studies: GEDEM : 1975-1981

. Mareol : 2007

– F. Lempérière: Usines marémotrices et environnement – Symposium EMR Brest 2013

– F. Lempérière : Quelles usines marémotrices pour le 21ème siècle – Techniques de l’Ingénieur 2013


F. Lempérière has been involved in the construction and/or design of twenty very large hydraulic schemes in rivers or at sea.

He has studied large tidal designs in France, Russia and India and has presented various papers about tidal energy.

He has been Vice-Chairman or Chairman of ICOLD Technical Committees for construction methods, cost analysis and cost savings.

He is Chairman of HydroCoop, a not profit association for advices on hydropower energy and tidal energy.

fig. 1

fig.3: Orthogonal turbine (Russia)


Fig 6: Channel dyke

Fig.7: Main dyke

Fig.8: Main dyke at shore

Fig.9: Possible future additional large harbour

Printable version Printable version

Back to top