There is a lot of power in waves, but one thing to keep in mind is that the amount of extractable energy is proportional to the head (height difference):
https://en.wikipedia.org/wiki/Hydroelectricity
http://www.reuk.co.uk/Calculation-of-Hydro-Power.htm
https://en.wikipedia.org/wiki/Low_head_hydro_power
https://en.wikipedia.org/wiki/Tides#Example_calculation
http://www.infoplease.com/encyclopedia/science/tide-the-magn...
So a dam uses a high head/low flow rate, and tidal or wave power uses a low head/high flow rate.
I’ve seen a few dams built around coves that work like a lock, filling up at high tide and draining during low tide, running a turbine to generate electricity. But if there is only a 1 or 2 meter head, it takes a huge water current to make more than a few MW.
I’ve also seen these snake-like creations that are big floating drums with hydraulics between them, that bend with the waves. So the head is about the same, but the period is a few seconds instead of a day, so a lower current is needed. However, the amount of current they capture is so small compared to a dam that they only generate on the order of a few kW.
The idea of using a large float is certainly compelling. Off the top of my head, maybe it could work similarly to having a bag at the bottom of the ocean with a pipe connected to an artificial floating island on the surface. It would pull on the bag as the tide rises to inhale air into the bag, and then let it exhale as the tide falls. A turbine in the tube would generate electricity in both directions. Water weighs 1000 kg per cubic meter (62.5 pounds per cubic foot), so if we use the dam formula for 1 cubic meter and +/- 1 meter of tide, the float would travel 1 m up, 1 m back down, then 1 m down and 1 m back up, for a total distance of 4 meters every 24 hours. So I think we can use the dam formula:
power P = phrgk
density p = 1000
height h = 4
flow rate = volume/time = r = 1/(60 * 60 * 24) = 1.16e-5
gravity g = 9.8
efficiency k = .6
so:
P = 1000 * 4 * 1.16e-5 * 9.8 * .6 = 0.27 W/m3
As a sanity check, the amount of energy required to raise 1000 kg of water 4 meters is:
E = mgh = 1000 * 9.8 * 4 = 39200 J
So power is energy over time, so:
P = E/t = 39200/(60 * 60 * 24) = 0.45 W/m3
So at 60% efficiency, power = 0.27 W/m3
This is one of the reasons why the head is so important, because it takes remarkably little energy to raise a ton of water, literally.
So if we want a 1 MW plant, we’d need a float about 3.7 million cubic meters, which would be a sphere with a radius of about 100 meters by V = (4 * pi * r^3)/3, or about 600 feet across. It would also need a bag of similar volume anchored to the bottom of the ocean.
Since the pressure is reversed for the bottom balloon, it would take on the shape of a tetrahedron, so would need strong anchors holding 3 corners to the ocean floor, with one of the corners connected to the float by a tube. The 3 edges touching the floor would need to be sealed against water leaking in. A bit of an engineering feat..
To make it worth our while, we’d probably want a float 1000 times bigger, or 6000 feet across to generate 1 GW. I have a little trouble picturing a mile wide float, its surface area would be:
Area = 4 * pi * r^2 = 4 * pi * 1000^2 = 12.6 million square meters
Although this doesn’t seem that bad, even at a cost of $10 per square meter.
Now that I write this, maybe a better alternative to the bottom bag would be a steel cable connected by a pulley to the ocean floor, that comes up on land and moves a meter or two forward and back each day. It could either be connected to a hydraulic system with a thin (high flow rate) tube, or perhaps a high gear reduction transmission that converts the motion to say, 60 Hz or something.
Unfortunately we’re talking 4 trillion kg of tension (4 billion m3 * 1000 kg per m) for a 1 GW plant, and a steel cable to carry that tension is infeasible, at 7500 kg per square cm (75 million kg per square m) of cross sectional area, which I calculated from the 11.5 mm cable which works out to a 1 cm2 cross section at 16540 lb (7500 kg):
http://www.engineeringtoolbox.com/wire-rope-strength-d_1518....
So when it’s all said and done, it’s possible to build the float but I don’t know how the power could be extracted practically.
Why?
Because the energy is derived from the tidal rise, not the amount of water you're displacing. The relevant volume is the area x the tidal displacement, which gives you the amount of water and how far it's been lifted (average: 1/2 the total displacement).
Or, if you prefer to have someone do the math for you, the City of San Francisco and EPRI did an analysis of how much total energy could be derived from tidal flows in and out of the San Francisco Bay, which I became interested in after an Internet crank suggested that the rise and fall of 30 typical ships could power all of San Francisco.
Not hardly.
http://oceanenergy.epri.com/attachments/streamenergy/reports...
The narrow passage under the Golden Gate Bridge, which connects the San Francisco bay to the Pacific Ocean is home to some of the most energetic currents in North America. In average 237MW of power is embodied in the tidal stream, of which about 35MW could be extracted without any negative impact on the environment. A plant of that scale could reach an electrical output of about 100MW at peak.
http://www.sfenvironment.org/article/hydro/tidal-energy
The study conservatively estimated that the Golden Gate site has 35.5 megawatts of total extractable average annual power, and that 15 to 17 average megawatts (MW) of this power could realistically be extracted by technologies currently in development. The cost of electricity generated, assuming incentives similar to those provided to other renewable resources, is estimated at six to nine cents per kilowatt-hour (kWh)—a cost competitive with current wind and natural gas generation, and about one-third the cost per megawatt of solar power.
As I recall, this works out to maybe 10% of the average power utilization of San Francisco proper, and a minuscule fraction of the energy used by the greater Bay Area. You'd do better to plaster the Bay with solar panels.
More: http://redd.it/1wefio
Tidal and wave energy are prodigious on a global scale, but they're simply too dispersed and too expensive to collect for present industrial-scale levels of energy use.
The minimal requirement is that the float, well, float. Other than that, more material is simply excess cost to consider.
The EPRI study I mention is based on, IIRC, turbines installed at the gate. Maximum capture would require building a barrier dam, which has been done at a few limited sites, though it would be impractical in SF due to ship traffic requirements.
https://en.wikipedia.org/wiki/File:Pulley3a.svg
A 200 m (600 ft) diameter, 4 million m3 float, generating 1 MW, would have a tension of 4 billion kg. That would need 533000 pulleys with 1 cm2 steel cable at 7500 kg tensile strength. So 266000 pulleys on each end, which is a 2D array of about 512 pulleys on a side. This doesn't actually seem that difficult to build on an industrial scale. The cable would travel about 2100 km a day, at 25 m/sec (about 55 mph). The cable would have a total volume of 2100000/10000 = 210 m3, or 6 m on side (20 ft), with the ends connected to the opposite sides of the pulley system. So that's quite a spool at the generator. It should be slightly mobile, either on wheels or a spring so that it can adjust to rogue waves. Also it should probably be made from a polymer that's resistant to seawater instead of steel.
Unfortunately we'd need 1000 of these plants to make 1 GW. Just goes to show how mindbogglingly much power humanity uses. Maybe we could at least try to conserve haha.
Can you do any maths on that concept and determine the size of the rafts which would produce usable energy consistently?