Although the optimal renewable energy portfolio consists of combining different renewables from different locations, it can be said with confidence that wind power will form a “staple” of electrical power generation worldwide as it is the most widely distributed, and readily accessible renewable resource.
So, for the sake of argument one can ask the question: How many wind turbines would a country need to satisfy its demand for electricity?
As an example, let’s consider the United States, the world’s largest consumer of energy.
The current generating capacity of the United States is roughly 1000 GW (ref: http://www.eia.doe.gov/cneaf/electricity/epa/epaxlfilees1.pdf).
The United States has abundant wind resources, with a high potential for wind power production. The capacity factor for wind power generation in the United States is around 30%. The Capacity Factor is the percentage of the maximum power that is actually generated, taken as an average. So, if the maximum capacity of a wind farm is 1 GW, and the capacity factor is 30%, then the actual generated power is 1×0.30 = 300 MW. The actual generated power is always much less than maximum due to intermittent wind conditions.
If we were to install large 5 MW wind turbines they would each, on average produce 5×0.3 = 1.5 MW.
Therefore, the number of 5 MW wind turbines needed for the U.S. is 1,000,000/1.5 = 666,667.
We would need approximately 670,000 5 MW wind turbines for the entire U.S. demand.
Another point to mention is that, although the average power will be 1000 GW, there will be significant power fluctuations due to wind variability. However, a baseload of power will be required to meet minimum year round demand.
There was a study by Stanford University indicating that 33-47% of yearly averaged wind power could be used as reliable year round baseload electric power if wind farms from different geographic areas are networked together (ref: http://www.stanford.edu/group/efmh/winds/aj07_jamc.pdf). Average wind fluctuations tend to decrease over large areas, which makes baseload supply a more attractive option.
This means that for an installed capacity of 1000 GW, 330-470 GW can be used as reliable baseload power for the U.S. The remaining electricity that exceeds this baseload level can be used to satisfy peak and intermediate loads on the electrical grid. One way to achieve this is by storing surplus energy when the electrical generation “overshoots” demand.
A necessary criterion is that the stored energy has to be “recovered” quickly, such as during periods of peak demand, or to a lesser extent, during periods of intermediate demand. Intermediate demand is more stable than peak demand but more intermittent than baseload demand.
A quick aside on power generation:
Electricity is produced by three general classes of power plants:
Peak power plants, called Peaker plants, are able to come online quickly when power is needed, such as when air conditioners come on during warm summer days. Peaker plants typically operate for a few hours at a time. They generally use gas turbines, which can be turned on and off quickly, as required.
Intermediate power plants, also called Load following power plants, usually run during the day and early evening. They generally include hydroelectric power plants or steam turbine power plants.
Baseload power plants are generally operated at full power continuously, except during periods of scheduled maintenance. They provide the minimum amount of power that must be delivered to the grid. For instance, power demand during night time is generally minimal and represents baseload power requirement. Baseload power plants are the most efficient operating and the least expensive to run, so it makes economic sense for them to operate continuously. Also, they cannot start and stop quickly. They require a long time to “warm up”; sometimes as much as a day or several days. So it makes sense that they run continuously.
The figure below shows a basic representation of power demand during the winter and summer.
End of aside.
From the above calculation, there is 330-470 GW of baseload power available by “pooling” wind resources. This amount of power is comparable to the amount of baseload power currently generated in the U.S.
However, the intermediate and peak power demands must also be met. As mentioned, the surplus energy can be stored in some way. One way is to use an electrolyzer system that creates hydrogen from water (electrolysis), and then the hydrogen can be used as fuel, such as in a combined gas-turbine/steam generator, with efficiency around 60%. Or, the hydrogen can be used as fuel in a combined Solid Oxide Fuel Cell (SOFC)/gas-turbine generator, with efficiency around 70%. However, these two options result in a round-trip electrical efficiency of less than 50% – the efficiency of the hydrogen electrolyzer is around 70%, and combining this with the efficiency of the combined-cycle generators we have an efficiency of 0.7×0.6 = 42%, and 0.7×0.7 = 49%. This is clearly a large loss in efficiency that results from using electricity to convert water to hydrogen and then using the hydrogen as fuel to produce electricity.
Another option is to use flow batteries such as Zinc Bromine Batteries or Vanadium Redox Batteries. These types of batteries have a high round-trip electrical efficiency of 76 and 78%, respectively (ref: http://www.mtpc.org/rebates/public_policy/dg/resources/2005-09-Wind-Storage-CEC-500-2005-136.pdf). They are able to discharge quickly, are robust, and have a very long service life. However, their energy density is low, and to provide large scale energy output they would have to be extremely large and heavy.
A viable option is hydro-electric pumped storage.
Currently, pumped storage is used during peak periods. During off-peak periods, surplus electricity is used to pump water “uphill” into a large reservoir. The water in the reservoir is then released during periods of high demand. This can balance out the electrical load by providing “extra” power when it’s needed.
It’s my belief that such storage will become very important for renewables such as wind and solar power, which are intermittent in nature.
Hydro-electric pumped storage is a technology that is proven and in place and uses a very abundant resource, water. It can be scaled up just like renewables can. The figure below shows an illustration.
The round trip efficiency is 70-85%, comparable to flow batteries, but relatively speaking it’s much more practical to scale this up.
Hydro-electric pumped storage facilities can be built near large bodies of water, such as on the east coast, west coast, gulf of Mexico, beside lakes, and especially, the Great Lakes.
This set up goes hand in hand with building large offshore wind farms, as they can be located near pumped storage facilities. Their surplus electricity can be used to store power for peak and intermediate demand. An added benefit is that power from pumped storage can easily be dispatched. Water can be made to flow down from the reservoirs, through the turbines, generating power, and the flow can be shut off just as quickly when the demand has subsided.
Ideally, the reservoirs are built on elevated land such as hills and mountains, located several hundred meters above ground.
I worked out that 1 GW worth of pumped storage lasting for 12 hours (12 GWh), would require a reservoir that is 1.5 km wide x 1.5 km long x 20 meters deep, and is 100 meters above ground.
It would take several hundred of these pumped storage facilities to be able to store the surplus energy from 1000 GW worth of wind power. No doubt it is best to distribute these facilities across the country, as close as possible to major wind farms. In the event that limited land is available for pumped storage it may be best to increase the size of the facilities on a smaller number of sites. Or, it may be best to construct elevated reservoirs near large bodies of water, where no natural land elevation exists or is off-limits due to environmental reasons.
Surplus electricity from other renewables, such as solar, geothermal, wave, and tidal power could also be stored in this way. However, in the southwestern U.S. where solar thermal power generation has the greatest potential, it is less likely that pumped storage can be used as this part of the country is arid, and lacking of abundant water resources. Large transmission distance may be the price to pay to deliver surplus power from solar thermal to pumped storage facilities. However, this necessity may be minimized during summer days, as power output from solar thermal plants tends to coincide with peak power demand, such as for air conditioner use. This can mean that little surplus electricity will be generated, if any.
Thermal energy storage for overnight electricity generation is possible with Solar Power Tower plants. These plants store heat from the sun in molten salt, which allows for electricity generation even after the sun stops shining.
As wind power and other renewables continue to grow, their contributions will gradually shift the power burden away from Peaker plants, followed by intermediate, and lastly baseload power plants. Eventually, fossil fuel power generation will no longer be needed except as emergency backup.