Energy vs. Power: Where battery storage reporting comes up short.

These two recent articles, which in my opinion missed some key details, inspired this post:

1. New Jersey Awards 9MW of Behind-the-Meter Energy Storage (GreenTech Media)
2. California’s push for clean energy has a problem: no place to store it (LA Times)

If I asked how large the hot water tank in your house is, would you answer “it can dispense [xx] gallons per minute”?

Me neither. Most people understand the difference between gallons of hot water and gallons per minute of hot water. Gallons describe the amount of water a tank can hold, while gallons per minute tell us how quickly those gallons can be dispensed.

Then why do journalists tend to write about batteries using watts — units that describe the maximum rate at which the energy stored in batteries can be dispensed? While important, doesn’t that miss half (or more) of the story?

It’s probably because a lot of people don’t fully understand the difference between power and energy. And you know what, it’s a pretty reasonable thing to be confused about. Here’s why:

1. The units we use are weird (welcome to physics)
2. Power plants and batteries operate under different constraints

So without further delay, here’s my guide to understanding energy vs. power:

Power is the rate at which energy is used or generated.

Think about the hot water tank example. Gallons of hot water (thermal energy) describe the amount the tank holds, and gallons per minute (power) describe the speed at which the energy can be dispensed. Batteries work the same way. They can store a certain amount of electricity and they can dispense it at a maximum rate.

energy / time = power

Even though the units used for batteries and storage tanks are different, the concept is the same. But let’s talk about units, because that’s where things get strange. Just like how gallons of hot water represent energy and gallons per minute represent power, shouldn’t watts represent energy stored in a battery and watt-hours represent power? Unfortunately it’s not that simple, and that’s because watts are a weird unit that are hiding a secret.

A watt is equal to a joule per second. Joules are units of energy (used more frequently in physics than when talking about electricity) and seconds are units of time, so watts are actually already units of energy per time — a.k.a. power. They describe electricity in the same way gallons per minute describe hot water.

watt = joule / second = energy / time = power

This also means that a watt-hour (a watt times an hour) is a measure of energy. This is because when you take a unit that represents power and multiply it by time, you’re left with energy.

watt-hour = watt x hour = power x time = (energy / time) x time = energy

So now that we’ve cleared up that watt-hours are energy while watts are power, and that energy (watt-hours) represent how much electricity a battery can store while watts (power) describe how quickly it can be dispensed, it’s time to understand why journalists tend to focus on watts rather than watt-hours. After all, we want to know how long a facility can run on a battery, not just how much electricity we can squeeze out of it at a given moment.

The confusion arises from old habits developed before people were talking about battery storage.

Centralized power plants typically use a fuel (coal, gas, nuclear rods, etc.) to generate electricity. The amount of electricity that is generated can be controlled by the amount of fuel that is used (and the plant’s ability to modulate), and it is assumed that the fuel will always be available at a given moment in time.

When the above conditions exist, you only need to mention a power plant’s power output to fully describe it. Simply multiply the plant’s power output by the number of hours it will operate, and you have its energy output. For example, Veolia Energy owns and operates a 163 MW natural gas power plant in Philadelphia. In one year the plant is capable of producing 1,427,880 MWh of electricity, assuming they operate it at all times.

power x time = energy

163 MW * (365 days x 24 hours) = 1,427,880 MWh

But how do you treat energy resources that don’t have always present fuel sources? Solar PV and wind turbines generate electricity from energy sources which are outside their operator’s control. The sun isn’t available at night or when it’s cloudy, so to simply multiply a solar panel’s rated power output by a number of hours does not genuinely represent the energy which would be produced during that time period.

The term capacity factor is used to deal with this issue. It is the ratio of an energy generating resource’s actual output over a period of time, to its potential output if it were possible for it to operate at full power rating continuously over the same period of time.

actual energy / theoretical energy = capacity factor

actual energy / (time x power rating) = capacity factor

For example, a 20 MW wind farm in Northamptonshire, England might have an expected capacity factor of around 25%, primarily due to the availability of wind in the region. This means over a period of one year the wind farm will likely produce around 43,800 MWh rather than the 175,200 MWh its power rating suggests.

power rating x time x capacity factor = expected actual energy

20 MW x (365 days x 24 hours) x 25% = 43,800 MWh

To be clear, even dispatchable generating assets have a capacity factor. Turbines have discs, rotors, shafts, bearings, blade rings, shells, and diaphragms which all need periodic maintenance. Reciprocating engines need oil/filter changes, valve adjustment, and overhauls. For these types of plants, capacity factors can reach up to 95%, depending on their maintenance requirements and intended run-hours.

Interestingly, the only generating technology I know of with a 100% capacity factor are radioisotope thermoelectric generators, which have been used in satellites, space probes, and remote Soviet lighthouses.

But how does this relate to batteries? Well, because up until recently we have been able to (with relative accuracy) describe an electricity source’s output using only its power rating. A 1 GW coal plant can reasonably be expected to produce 1 GWh over a period of one hour, 2 GWh over two hours, and so on. Once renewables started becoming more popular, some journalists even began citing capacity factors or anticipated annual outputs in addition to power ratings (although not frequently enough) to account for the availability of their energy sources. But old habits die hard, and even though reporting on battery storage should require citing both power rating and energy storage capacity, journalistic practices just haven’t caught up yet.

Neglecting to cite energy storage capacity when writing about batteries can be incredibly misleading.

Remember those two articles I mentioned at the beginning of this post? They completely mess this up. The first one, an article about battery storage incentives being dished out in New Jersey, shows a table describing each battery project, the storage capacity, the incentive request, and the project cost. Here it is:

Of course, they describe storage capacity as power rather than energy (kW instead of kWh) in the header of their table. Why is this such a bad thing? Well apart from putting the phrase “storage capacity” next to “kW”, by not including kWh ratings the article does nothing to describe what operational value these batteries will provide. While power rating is important for brief load balancing operations and frequency regulation, these are supposedly resiliency projects. Without a meaningful value for storage capacity, we have no idea how long these schools and municipal buildings (which often serve as shelters during emergencies) will be able to operate without the grid.

A second issue is that they placed dollar values next to kW ratings. A huge part of the conversation surrounding energy technologies is cost. We want a cleaner environment, but we have to figure out a way to do it which people can actually afford. This is why, for example, it is so exciting that the $/kW cost of solar PV has dropped significantlyover the past 10 years.

While using the $/kW metric for a generating asset makes sense, using it for storage assets is really misleading. The table above suggests battery storage capacity costs between $1,300/kW and $1,600/kW. Wow, that’s cheaper than commercial solar (~$2,300/kW) and right in line with commercial CHP (~$1,400-$2,000/kW)! But as we’ve discussed, the relative size of a battery system has a lot more to do with kWh than kW. Even though this table makes battery storage appear extremely competitive, we really have no idea if these projects stand-up economically. My guess is they don’t. This does nothing for helping readers understand where the energy storage market is going and if its a viable solution for behind-the-meter capacity.

At this point I think I’ve probably spent enough time on this topic, but I hope I’ve made the case for talking about batteries using both energy and power ratings. If there is anything I missed here be sure to let me know.