I recently reviewed an EPRI document that discussed storage and by far the largest size storage systems were pumped storage plants. I wondered why they did not include hydro (non-pumped) storage, as this form of storage is far larger than any other form of storage that is available on the grid now.
Parts of North America, but sadly not all of it, are blessed with mountainous territory that has many rivers and streams that run downhill, and many of these have been harnessed for electricity production. While not specifically intended as storage plants when built, the value of their storage may well turn out to be larger than the value of the electricity that they may produce.
Consider a hydro dam that is 35 M in height with a reservoir that is 10 km2. Discharging the top 1 M of water through a generating station (90% efficient) would release almost 840 MWh of stored energy.
This is a small hydro plant, with a small reservoir behind it, yet the storage is almost 840 MWh/M of depth that is drawn from the forebay. That is in addition to the electrical energy generated for use.
So how does a utility that has no pumps manage to store and return energy? The process is both simple and efficient. Hydro plants generally have the capability to be started and stopped quickly. They can change load rapidly and are generally a flexible resource.
A utility that supplies power to many customers, that has hydro capacity potentially has capability to take power from others that is actually not stored but is used to power their own loads. At the same time, the capacity at the hydro plant is reduced by the amount of the import, and the lake behind the dam slowly fills. In the example above, an import of about 840 MWh would result in an increase in the lake level of about 1 M.
When the time comes to return the storage, the utility simply generates more power than they need for their own load, and the 840 MWh that was stored by the plant is returned. One real benefit of this process is the efficiency; rather than pumping water uphill and generating when it comes down, at an efficiency that is generally less than 70%, this process is simply “moving” the time at which the energy is generated. It is essentially 100% efficient, with a small increase in line loss as a result of the need to carry more power at peak periods.
This example plant is very small. Several Canadian utilities have very large plants, some that are more than 200 M in height and have reservoirs that cover up to more than 1,500 km2. The BC Hydro dam at Hudson’s Hope has a forebay area of 1,761 km2, and that lake can be drawn down by more than 35 M, resulting in storage capacity of more than 20,000,000 MWh.
By comparison, batteries currently cost more than $200/kWh and are expected to fall significantly in the next 10 years. A battery cost of $60/kWh to replace the Hudson’s Hope storage would cost more than $1400 Billion. These facilities were created to generate electrical energy; the storage is a bonus.
If renewables are to work efficiently, we are going to need a significant level of cooperation between utilities that have large storage capacity and the renewable energy sources. I understand that at present, California is paying rooftop solar generators a high price for their energy, only to find that they have a surplus that they are paying companies with the flexibility to take it, and then after dark, they well may need to be able to purchase the energy back to meet a growing peak demand. They are paying three times for the same kWh. Surely this is not optimal