Energy storage brings a renewable energy future one step closer

by Julia Rosen
Wednesday, September 4, 2013

The 20-megawatt Gemasolar concentrated solar power plant in Spain uses a "power tower" and is capable of storing 15 hours of solar power. Copyright Markel Redondo, Greenpeace

Renewable energy sources promise to address many of the energy challenges facing society: They derive power from inexhaustible supplies of sunlight and wind and have the capability to meet a substantial portion of global electricity demands without adding greenhouse gases to the atmosphere. However, renewable power supplies must first overcome one inherent drawback: variability. Humans can’t control when the sun shines or how much the wind blows — nor can society adjust its daily rhythms to use energy only when nature provides it — and that’s a fundamental problem for utilities seeking to incorporate renewable power sources into the existing energy infrastructure.

The commercial power grid delivers energy produced by power plants to consumers, but providing a stable supply of electricity poses a range of challenges even without renewables. Imbalances between the amount of energy scheduled to be available and the amount of energy drawn from the grid crop up on different timescales for different reasons. For instance, short-term variations in demand, like when a factory turns on its machinery, can lead to temporary imbalances of seconds to minutes. Daily imbalances occur during demand peaks, when many consumers require large amounts of energy at the same time, like at the end of the day when work, dinner and sunset coincide. Finally, weekly to monthly imbalances stem from changes in energy demand caused by weather or seasonal variations in industrial or domestic activities.

Such imbalances have always plagued the electrical grid, and utilities have traditionally met them by simply ramping up power production at coal and gas plants to increase supply (or, in some cases, by forcibly decreasing demand through brownouts and rolling blackouts). However, renewable energy sources cannot be harnessed the same way that fossil fuels can, because the forces that power them lie outside a utility’s control.

In fact, renewables make the stability problem worse: They introduce variability from the supply side as well. A passing cloud can decrease solar power production by up to 80 percent for minutes at a time, magnifying the need to regulate short-term power variations. In some places, the wind blows strongest at night, when there is little electricity being drawn from the grid, meaning it can’t help meet the daily peak in energy demand. And snowmelt often produces the most hydroelectric power in the spring, despite consumption typically being higher in the summer due to greater air conditioning demand.

Energy storage, however, holds the potential to overcome all of these issues: Storing energy during times of excess production allows it to be released later, when demand is high. Although batteries and capacitors — traditional ways of storing electrical energy — have continued to improve, researchers have also pursued other innovative storage technologies to bring renewable energy sources one step closer to large-scale energy production.

The strengths of individual storage techniques often cater to one particular kind of energy imbalance. Taken together, however, these technologies allow traditional power plants to operate more efficiently and help alleviate the problem of renewable energy’s variability, making it more economical and streamlining its integration into the grid.

Ironing out the Kinks: Flywheels and Short-Term Frequency Regulation

Buffering the grid against short-term variations in electrical supply and demand serves a vital role in energy delivery. “The grid is operated within a very tight tolerance at 60 Hertz [the standard operating frequency in the U.S.],” says Georgianne Huff, project manager of energy storage at Sandia National Laboratories in New Mexico. “When it goes out of that range, that’s when we start having problems.” So, as renewable energy plants expand, bringing increased variability, a new class of service providers — separate from power plants — are stepping up to use energy storage to smooth out hiccups in power supply, a process known as frequency regulation. In addition to batteries, one promising technology involves modern adaptations of the flywheel.

Flywheels, such as those produced by Beacon Power, a company funded in part by U.S. Department of Energy’s (DOE) renewable energy initiative,  store excess electricity as rotational energy by accelerating a 2-meter-high cylinder up to tens of thousands of rotations per minute. The flywheel’s rotor (often made of ultra-light carbon fiber) spins on a shaft inside a vacuum chamber to minimize friction. When energy is needed, the cylinder decelerates, converting its stored energy back to electricity with an efficiency of about 85 percent.

Flywheels lose little energy during storage, and when used in vast arrays, can buffer the grid with up to 15 minutes of electricity with each discharge, enough to smooth out most problematic short-term fluctuations in voltage, Huff says. “One of the benefits is that flywheels can respond immediately,” she says. They react so quickly to grid imbalances — they require less than 1 second to reach full power — that the Federal Energy Regulatory Commission recently issued an order allowing utilities to pay faster-responding providers like flywheel companies higher rates for their services, which easily outperform fossil fuel power plants at frequency regulation.

While batteries are equally capable of rapid response, flywheels have some significant advantages over other forms of short-term storage and frequency regulation, Huff notes. They have a small geographic footprint for the amount of storage they offer, and once assembled, can operate for decades with minor routine maintenance.

In the last five years, Beacon Power has built a string of fully operational grid-scale flywheel frequency regulation facilities on the East Coast, each capable of absorbing or releasing 20 megawatts (MW) of power, or between 5 and 20 percent of a typical power plant’s output.

Waste Not Want Not: Concentrated Solar Power and Daily Peak Demand

Unlike the short-term energy imbalances that flywheels mitigate — which appear suddenly and without warning — the daily peak in energy demand is as reliable as the setting of the sun. Although the shape of the daily demand curve varies with the season as winter heating demands give way to cooling needs in the summer, peak demand usually occurs about 6 p.m. As many different energy needs collide, from cooking to lighting homes, the grid must cough up more power than at any other time of day. This persistent problem has led to the construction of numerous so-called “peaking” power plants over the years, the sole purpose of which is to provide extra electricity for peak demand. However, an alternative solution is to store renewable energy during the lower demand hours of midday and dispatch it later, obviating the need for additional plants.

The most obvious candidate for this kind of time-shifting storage is solar, exclusively available during the daytime, but capable of providing more energy than could possibly be used at any given time. One form of solar storage already in use at a handful of concentrated solar power (CSP) plants across Europe and the Middle East is called molten salt thermal storage. Molten salt storage will also be included in two new CSP plants under construction in the American Southwest, the 110 MW Crescent Dunes “power tower” plant in Tonopah, Nev., and the 280 MW Solana parabolic trough plant in Gila Bend, Ariz.

CSP plants employ mirrors to concentrate sunlight to tens or even hundreds of times the levels received at the surface, storing it as heat in a thermal fluid. “It’s similar to what you see when playing around with a magnifying glass — the more you concentrate that sunlight on a small spot, the higher temperatures you will get,” says Mark Mehos, program manager of the Concentrating Solar Power division at the National Renewable Energy Laboratory in Golden, Colo. Some of the heat stored in the thermal fluid is used to generate steam that drives a conventional turbine power block, but at plants with storage, some heat is siphoned off to wait in hulking tanks filled with molten salts.

These solutions of alkali nitrates can reach temperatures up to 560 degrees Celsius, possess an extremely high heat capacity and lose almost no energy overnight. Therefore, molten salt storage allows CSP plants to reserve and later release between 6 and 15 hours of their typical energy production during peak demand, competing with conventional peaking plants that use nonrenewable fuels on demand.

This also allows CSP plants to sell their power at a higher cost in the energy marketplace, making it a more valuable resource. In addition to increasing the value of the energy delivered, energy storage lends CSP what Mehos calls “capacity value,” or “the ability for the plant to eliminate the need for the construction of a new gas plant.”

Go Big or Go Home:  Long-Term Storage Using Pumped Hydropower and Compressed Air Storage

[CSP allows energy to be stored overnight, but storing energy for longer periods of time — for instance, to offset seasonal changes in energy demand — is a substantial technological challenge. “The amount of electrical energy trafficked on the grid at any given time of day is just an absolutely mind-bogglingly large number,” says Ted Brekken, a professor of electrical engineering at Oregon State University who specializes in renewable energy systems. “So when you are talking about storing energy on that scale, it immediately limits a lot of your energy storage options,” he says. To date, Brekken says, the only technologies up to the task include an existing method called pumped storage hydropower and a newer alternative known as compressed air energy storage (CAES).]{style=“font-size: 12px;"}

Pumped storage relies on the simple concept of converting excess electrical energy into stored potential energy (the energy it takes to resist gravity) by pumping water from a lower reservoir to a higher reservoir during times of surplus energy production. Then, when power demands increase, this water runs back downhill through the hydropower turbines to produce electricity. Although pumping the water uphill consumes energy, the process allows power produced during off-peak hours, including renewable power that would otherwise go to waste, to be stored for later use. Pumped storage accounts for 99 percent of global energy storage today (both renewable and nonrenewable together), holding some 130 gigawatts of power, according to a 2013 report on energy storage compiled by Sandia Labs. The first plants were built more than 30 years ago and recover 70 to 80 percent of the energy used to elevate the water.

More recently, CAES has joined pumped storage as a viable grid-scale solution. In the technique, excess available electricity — produced from any type of energy, including wind, solar, hydropower and conventional fossil fuel sources — powers a compressor that drives high-pressure air into storage cavities in bedrock near the power plant. The air remains in these subsurface reservoirs until needed. Then, the air is released, heated to increase output and used to power conventional turbines that generate electricity. This method has been in use at a plant in Germany since 1978 and in Alabama since 1991. In both locations, air is stored in caverns excavated for salt mining.

But greater utilization of CAES could be particularly important for renewables. “The wind often blows at inconvenient times,” says Casie Davidson, a research scientist at the Pacific Northwest National Laboratory (PNNL) in Richland, Wash. Davidson was involved with a recent study exploring the suitability of the regionally ubiquitous Columbia River basalts for compressed air storage in cooperation with the Bonneville Power Administration. “Being able to take that power, store it and use it later would allow us to use that wind resource more effectively,” she says.

CAES can operate on daily timescales, and PNNL even demonstrated its use in short-term load balancing. But its real advantage comes in its ability to store tremendous amounts of energy for very long periods of time, Davidson says. One of the sites investigated in the study, on the banks of the Columbia River in eastern Washington, can continuously take in air for 40 days without surpassing the capacity of the reservoir, she says. That would hold enough energy to power 85,000 homes for an entire month. Another site, near Yakima, Wash., can accommodate a full year of compressed air input. This site uses geothermal energy to reheat the air, resulting in zero greenhouse gas emissions.

Davidson says that while there are minor risks associated with CAES, such as leakage, they can be minimized using the knowledge engineers have gained through managing other kinds of injection sites and researching carbon dioxide sequestration. And even in the case of leakage, Davidson says, “it’s just air.”

The main obstacle in the way of large-scale CAES or pumped hydropower is geography. For pumped storage, Brekken says, “you have to have a reservoir and a dam,” because it is prohibitively expensive to build the infrastructure solely for storage purposes. Similarly, CAES also requires specific geologic assets, although these are likely to be more widespread. Both previous CAES sites used relatively rare dissolution-mining salt caverns, but PNNL’s new study examined the suitability of natural rock formations.

“In this particular project, we were looking at basalt,” Davidson says, “but I think the implications are broader than that.” Any place with a porous and permeable rock formation, capped by an impermeable layer and ideally folded into a geologic structure called an anticline that traps the air, could be suitable, she says.

The Bottom Line: What Is Renewable Worth?

Storage is such an important idea that even conventional power plants are getting on board, says Huff of Sandia Labs. It increases revenues by converting low-cost off-peak power into valuable peak power, eliminating the need for additional plants to meet short-lived demand peaks, and lowering greenhouse gas emissions by helping plants operate consistently at maximum efficiency. “The conventional [economic] planning tools that utilities use do not incorporate storage very well,” Huff says, in part because storage straddles the generation and transmission sides of a utility’s responsibilities. It took the renewable energy revolution to open utilities' eyes to the value of storing energy, she says.

But while storage benefits all forms of energy generation, it may prove uniquely transformative for renewables, many of which cannot currently compete with the economics of fossil fuels — barring the assistance of government subsidies. This inequality stems, in part, from the fact that, historically, renewable power could not be completely consumed during the limited times it was abundantly available. Storage helps solve that problem, making renewables a viable alternative to fossil fuels, Brekken says.

Energy storage also leverages the fact that renewable resources, by definition, never run out. Fossil fuels may have some major advantages — for example, “they are crazy energy-dense and that’s why we love them,” Brekken says. But they also possess a fatal flaw, he says. They will run out. We are using fossil fuel energy at a rate much faster than we are accumulating it, he says. In other words, we “are spending down our bank account.” Energy storage offers the opportunity to refill that bank account whenever there is an excess supply of renewable power, he adds, and that may very well be worth paying for.


© 2008-2021. All rights reserved. Any copying, redistribution or retransmission of any of the contents of this service without the expressed written permission of the American Geosciences Institute is expressly prohibited. Click here for all copyright requests.