What's the average efficiency of pumped hydroelectric energy storage facilities?

Question

Pumped hydroelectric storage has, to date, been the most scalable way to store electricity (in this case, in the form of gravitational potential energy), allowing us to spread the time between when the electricity is first generated, and when it is finally consumed.

There are inevitably losses - the pumping of water uphill is less than 100% efficient, the conversions of the potential energy to kinetic energy to electricity are too.

What is the typical efficiency of pumped hydroelectric storage?

On top of the losses described above, the upper reservoir may experience some evaporation, depleting the energy stored; and it might get topped up from precipitation - is there any data on how that affects the overall real-world performance?

Answer 1

In this question about rooftop hydro I covered the efficiency question almost as an aside. I can't find an actual plant with efficiency over 80%, only claims that that might be possible. The average efficiency will be much lower than the peak, as a lot of plants are old and have efficiencies around 60% (although some have been refitted to boost efficiency). Since we're dealing with big civil engineering projects changing the design is usually a big, expensive project. Published efficiency figures often fail to separate natural inflow from pumped, although thanksfully the engineers seem to dominate these discussions so they generally say "{year} input x GWh, output y GWh (including natural inflows)". Otherwise a few plants would have efficiencies over 100% some years :)

I started with the link William also found to the Energy Storage Association who appear to be a trade group promoting the idea.

these plants are typically highly efficient (round-trip efficiencies reaching greater than 80%) and can prove very beneficial in terms of balancing load within the overall power system. Pumped-storage facilities can be very economical due to peak tand off-peak price differentials and their potential to provide critical ancillary grid services.

Wikipedia make a similar claim:

the round-trip energy efficiency of PSH varies in practice between 70% and 80%, with some claiming up to 87%.

But they give references. Unfortunately the Hawaiian Electric Company claim of 87% round-trip is no longer on their web page and I can't find an archived version. They don't appear to have built an actual plant with that efficiency, so I suspect they were repeating advertising claims.

The people promoting trains for energy storage say that 70-75% or more realistic (via The Economist):

Its system uses modified railway cars on a specially built track ... delivers more power for the same height differential. He also says it is more efficient, with a round-trip efficiency—the ratio of energy out to energy in—of more than 85%, compared with 70-75% for Pumped Storage Hydro.

Storing Energy: with Special Reference to Renewable Energy Sources By Trevor M. Letcher says 60% for older plants and up to 80% for newer ones.

Answer 2

Energy Storage Association cites round trip efficiencies above 80%

http://energystorage.org/energy-storage/technologies/pumped-hydroelectric-storage

As for the evaporation question. I don't know of any specific data on it, but would expect it to be minimal if you charge the reservoir during off-peak (cheaper) hours and discharge it during the morning ramp. Evaporation can also be eliminated through underground pumped hydro... one example: http://www.greentechmedia.com/articles/read/Gravity-Power

Answer 3

As a Mechanical Engineer, I know the efficiency of the pump(s) and generator will be a function of the head or the height difference between the upper and lower reservoirs. Also, the line losses will increase with distance. Finally, an uncovered reservoir in an arid climate will suffer evaporation losses. Therefore, there isn't a single answer to this question, once a system is designed the efficiency can be estimated, but the efficiency will vary from one system to another.