One of the main criticisms of renewable energy facilities is that they are unable to dispatch electricity when it’s needed. The great game-changer is low-cost energy storage, which would enable renewable energy production to be stored and rapidly released when needed. Here are seven promising distributed energy storage technologies that could be commercialized in the near future.
The electricity industry has enabled customers to expect that their lights will turn on with each flip of the switch. Hidden behind that switch is an enormously complex machine of transmission and distribution wires, substations, and a network of power plants that must respond, immediately, with the electricity required to operate those bulbs. Add up the myriad requests by millions of consumers over a typical day and the synchronicity of demand and supply is astounding.
Even less appreciated is the human element that controls the machine’s daily operation. The technology behind dispatching a strictly fossil-fuel and nuclear fleet has been well-honed over many years by utility engineers. Today, modern automatic generation control systems can quickly respond to commands of a single dispatcher responsible for continuously balancing electricity supply and demand for those millions of customers—a necessity, as storing any appreciable quantity of electricity is impractical today.
It’s Complicated
Adding intermittent renewable sources in small quantities to the grid, as has occurred over the past decade, has made the dispatcher’s job much more complex. Accurately predicting the amount of wind or solar generation tomorrow, or even an hour from now, in order to balance electricity supply and demand, is problematic (Figure 1). Ask the typical dispatcher (as I did during a recent tour of a Midwestern utility’s dispatch center) what new tools would make his job easier, and the answer is invariably: “Give me electricity that can be connected to the grid at the touch of a button.” Today, with the exception of the 10 pumped storage plants in the U.S. and a single compressed air energy storage facility, bulk electricity storage remains a fictional technology. But all that is about to change, as the discussion of emerging technologies below shows. (See also “Bulk Storage Could Optimize Renewable Energy,” in POWER’ s September 2010 issue or in the online archives at http://www.powermag.com.)
 |
| 1. Tough balancing act. Public Service Co. of Colorado (PSCo), an Xcel Energy company, has been aggressively adding wind power to its system for the past few years. The data in this figure, captured for July 2, 2008, illustrate how the variable nature of wind-produced electricity can affect the dispatching of fossil plant resources. Early in the morning, coal was providing baseload electricity (yellow) while gas-fired assets (green) were load-following. The blue line shows wind generation. The red line is the NERC-defined Area Control Error (ACE), which represents, in essence, the difference between supply and demand. Note how the morning increasing wind forced PSCo’s coal plants to minimum load. Wind then quickly dropped between 7 and 8 a.m. and almost disappeared by mid-morning, forcing both coal- and gas-fired plants to quickly ramp up to match the lost wind load and meet the rapidly rising summer day demand. Wind was essentially absent during the peak demand period. An effective form of energy storage would eliminate these large and inefficient load swings and the system disturbances they cause. Source: PSCo |
In the interim, dispatchers have learned to manage the unpredictable side of renewable generation, usually by keeping sufficient spinning reserve capacity at the ready to cover any eventuality. Utilities have also upgraded ramp rates (the rate that electricity production can be increased or decreased) on existing fossil plants, added automatic generation controls for remote control of the faster-responding units, and installed fast-acting simple-cycle combustion turbines (see “Flexible Turbine Operation Is Vital for a Robust Grid,” September 2010) or reciprocating engines (see “Top Plants: Goodman Energy Center,” September 2009) in some regions. These technologies provide even faster response during periods when intermittent renewables were predicted to generate electricity but didn’t, or in regions where market conditions place wind first in the dispatch order.
But these technical improvements in existing equipment are merely stopgap measures for many regions that have plans to add even more renewables to the grid. California, for example, is mandated to have 33% renewable use by 2020; Colorado’s renewable portfolio standard goal is 30% by 2020. By 2020, fewer, faster fossil plants will be under the control of the California Independent System Operator, so they will not alleviate grid reliability concerns, especially given the large amount of imported energy the Golden State relies on. And the U.S. nuclear fleet, in general, is not designed for load-following or cycling operation, so that option is off the table.
While dispatchers are grappling with these new grid realities, the grid machine continues to evolve. In the future, dispatchers will be asked to manage a much more diverse power generation system consisting of many more variable electricity producers on the supply side and even more behind-the-meter distributed resources, such as rooftop photovoltaics, on the demand side (see p. 46, “The Smart Grid and Distributed Generation: Better Together”). Bulk quantities of electricity from larger wind farms will continue to increase (although at a slower rate in the future) while the wind turbine technologies they use promise to become more efficient (see p. 38, “Changing Winds: The Evolving Wind Turbine”).
The politics of renewable energy incentives play a large role in determining which plants will be dispatched, yet there’s no accounting for the hidden costs to the fossil fleet. For example, it’s now common practice in the Electric Reliability Council of Texas (ERCOT) for wind generators to sell electricity at negative prices—owners are paying ERCOT to take the energy—in order to qualify for a production tax credit that is based on megawatt-hours produced. During the summer months, off-peak wind is replacing gas-fired generation in ERCOT and disrupting the market, even when natural gas prices are low. The problem was exacerbated by the average wholesale price of electricity dropping as much as 50% in some regions of the U.S. over the past year.
Another complication is that energy storage rules and regulations are in their infancy; only a single state-required energy storage law is on the books. On September 29, 2010, California Governor Arnold Schwarzenegger signed AB 2514, the Energy Storage Bill, into law. According to the California Public Utilities Commission, the next step is to “open a proceeding to determine appropriate targets, if any, for each load-serving entity to procure viable and cost-effective energy storage systems.”
The new California law requires utilities to obtain 2.25% of their peak power from storage systems by 2014 and 5% of their peak power from storage by 2020. Perhaps this renewed interest in energy storage will breathe new life into the moribund 500-MW Lake Elsinore Advanced Pump Storage Project that has been under development in southern California since 1987.
The problem of balancing renewable generation with other environmental demands is also forcing some utilities to make tough decisions. A good example is when very high water flows in the Columbia River, as are being experienced this year, reach Bonneville Power Administration (BPA) hydro plants. Coincident with high hydroelectricity production is the production of enormous amounts of power from wind farms that have been installed in the region. In fact, total hydroelectric and wind generation exceeds grid demand during certain times of the year. When grid demand is low, BPA’s choice is to either bypass dam generators or reduce wind energy purchases. But excessive flows through the dam bypass (required when hydropower is curtailed but the water flows continue) damage salmon and other fish species, violating Endangered Species and Clean Water Act requirements placed on the hydro plants. BPA has proposed a program of “environmental dispatching,” whereby the hydro dams are dispatched first to ensure all environmental operating standards are met and wind generators get the remainder of the load. BPA is expected to release its final plan on April 1.
These energy supply timing problems have other significant consequences for grid management. For example, for planning and dispatching purposes, ERCOT assumes that only 8.7% of all the wind power generated in Texas will be available to meet on-peak power demand. The uncertain and variable output from renewables also makes these energy sources unsuitable for providing contingency spinning reserve, load-leveling arbitrage (moving electricity production from off-peak to on-peak hours), and regulation services (including black start and VAR support), which are vital to reliable grid operation. These basic grid services (called ancillary services in some regions) are critical for reliable grid operation and are costly to provide. Whether those costs should be shared by all electricity users or by all electricity providers remains a topic of heated debate.
Email
Comments
Print
Reuse
Comments (2)
Amy Hunter
Finanicial Analyst, Sr
Xtreme Power