Thursday, August 15, 2013

Hawai`i Energy Storage 4: Running the battery numbers--prepare to be confused

There are all kinds of ways to store energy.

First thing to come to mind is a chemical battery, and there are all kinds of chemical batteries. We'll look at some of those in detail in the next installment of this series on energy storage.

You can also store energy in the form of heat. If we can make electricity out of hot rocks using natural geothermal resources, we can make electricity out of hot materials that we heat—perhaps using mirrors to focus the sun’s energy.

If we pump water to a high place, and then run it through a turbine as it flows to a low place, we have what’s called pumped storage. It has been built across the world for most of the last century.

People have pumped air into deep underground chambers, and then collected energy when it is released, a technology called Compressed Air Energy Storage or CAES.

Some folks are fascinated by big fast-spinning flywheels as a storage medium.

If you use energy to make a liquid fuel, like hydrogen, which later can be used to generate power—that’s a form of energy storage, too.

So with all these alternatives, how can we distinguish between them? That’s a complex problem. Some of the factors to be considered:

How much does it cost to build the system, and how much power will it store? This might be called installation cost—how many dollars per megawatt stored.

How much energy is lost in charging and then discharging? In a calculation called round-trip efficiency, if you need to deliver two megawatts to the battery but there’s only one megawatt that can be drawn from it, that’s a 50% round-trip efficiency. There’s some loss in every system, and one standard chemical battery folks shoot for is 80%.

How fast will it deliver its charge, and how fast will it recharge?

How many such charge/discharge cycles will it take before the power storage system--or part of it--needs to be replaced?

One calculation that seeks to answer a lot of those questions is levelized cost of energy, or LCOE. The National Renewable Energy Laboratory defines it here.

It is a calculation that combines the “capital costs, operations and maintenance, performance, and fuel costs.” It is often delivered as units of money divided by units of energy, such as dollars per megawatt-hour.

As an example, the 2013 Electricity Storage Handbook estimates the LCOE at 2012 prices of pumped hydro at $150 to $220 per megawatt hour. Lead-acid battery storage can come in anywhere from $250 to $1,700 per megawatt hour.  Flywheels turn in at about $375 per megawatt hour.

But even this calculation isn’t the whole story. The numbers can change dramatically depending on how you use the storage medium.

A lithium-ion battery used in frequency regulation—meaning it is responding quickly to sharp jumps and drops on the grid—can come in at $100 to $350 per megawatt hour. Use such a technology to help out in transmission and distribution applications and it can run from $700 to nearly $1,200 per megawatt hour. In some commercial and industrial applications, the Energy Storage Handbook puts the LCOE at from $700 to more than $2,500 per megawatt hour.

All those numbers are useful for one type of comparison between different storage types.
But sometimes when you’re looking at references, you’ll see similar dollar numbers attached to similar energy numbers, but this time energy is listed as kilowatt hours.That's because you're looking at a different thing entirely.

Example:  Last year Scientific American reported on a new kind of lithium-ion battery that could be produced for $125 per kilowatt hour. That’s what it costs to buy the system divided by how much energy it can hold at full charge.

The number doesn’t tell you a lot about the cost to operate that battery system. If it is cheap and will operate satisfactorily for 20 years, that’s one thing. If it’s cheap but needs to be replaced after 3 years, then maybe it’s not cheap.

To look at this a different way, consider that electricity in Hawai`i sells to consumers for roughly 35 to 40 cents per kilowatt hour, and maybe 20 cents of that is the cost to generate the electricity. If the LCOE for your storage system is $300 per megawatt hour, that’s 30 cents per kilowatt hour.

So if your solar photovoltaic system in the daytime delivers energy at 20 cents, and you store it for 30 cents—your energy cost is already 50 cents and you haven’t yet paid the cost to deliver it to a customer.

Thirty-cent storage may make a lot of sense for one application, such as stabilizing the grid, where a little storage goes a long way. But it does not make so much sense for bulk storage of daytime solar energy for delivery at night. For that, you might need a better, far cheaper alternative.

Selecting a storage system, then, is a little like buying a car. How much money do you have to spend and what are you going to use the car for? Is your limit $15,000 or $50,000? A little convertible roadster might get you a date, but won’t take six kids to soccer practice, and a passenger van won’t get you down a muddy road to a hunting spot like an SUV or four-wheel pickup might. A golf-cart looking neighborhood electric car does your local shopping duties, but might not get you to a North Shore beach and back for a weekend outing.

Next, we’ll look at some specific non-chemical batteries.

© Jan TenBruggencate 2013

1 comment:

jonathan jay said...

interesting series, Jan. Might you take a closer look at pumping storage like what they are doing in the Canary Islands? That could be appropriate technology for us to consider adapting our legacy plantation hydro systems to.