The old shibboleth to some extent is literally true. The electric industry appears different from the natural gas industry in that demand must be matched immediately with production. No viable location comes to mind to put away some of that extra power until it is needed. But literal truth is not necessarily the whole story. Competition has a way of changing perception.
Gas storage acquired a new role after the Federal Energy Regulatory Commission (FERC) issued Order 636. Deregulation will do the same on the electric side. The roles and uses of electricity storage are likely to change in ways similar to changes already seen on the gas side.
Gas storage fields are now used more consistently throughout the year, as reflected by the increasing numbers of gas-injection and
-withdrawal transactions per field.1 Gas consumers and distribution companies now augment traditional seasonal storage with services like daily and monthly supply balancing and price hedging. They want services that help them resolve imbalances quickly and enable them to profit or protect themselves from short-term changes in gas commodity and transmission prices.
Demand for increased working-gas capacity and daily deliverability has altered the development of storage facilities.2 Historically, most storage facilities made use of depleted gas and oil fields. Now, salt-cavern facilities account for 68 percent of proposed additions to withdrawal capacity, as compared to 10 percent of total withdrawal capacity in 1993.3
Salt-cavern peaking facilities are designed to offer greater deliverability than traditional storage fields.4 They give buyers and sellers the operational flexibility they need to respond to quickly changing market conditions, or provide new services such as supply balancing.
With the "commoditization" of natural gas, hubs and market centers have sprung up to facilitate the physical movement and pricing of natural gas. Market centers offer a myriad of administrative, operational, trading, and financial services, including balancing, parking, and risk management.
Another byproduct of note: independently developed, owned, and operated gas storage facilities. Previously, only a gas utility or pipeline would possess such a specialized asset. Now that the values of these assets have become a function of arbitrage opportunities and versatility, entrepreneurs have entered the business.5
As it did in the gas industry, open access in the electric industry will change the analytical framework for evaluating electric storage technologies. These changes will follow from rising demand for services such as load balancing, and by the collapse of average-cost pricing. Electric storage will become a means of reducing costs and providing utilities, suppliers, and customers with vital operational flexibility.
Historically, the most common type of bulk electricity storage developed by utilities has been hydroelectric pumped storage (HPS). HPS provides fuel-cost savings by using power from coal-fired and nuclear generating plants (each boasting low variable costs) to pump water uphill during offpeak hours. The stored water generates hydropower during periods of peak demand.
By adding HPS plant to its generating mix, a utility can increase its use of base-load power plants with low marginal production costs during periods of low demand to provide energy with a low variable cost during periods of high demand. It thus avoids capacity additions and reduces its peak use of expensive fuels like oil and gas.
While an HPS dam or reservoir appears more expensive and capital-intensive to build than a combustion turbine, the additional capital cost can be offset by other HPS system benefits, such as lower average production cost based on arbitrage between low-cost units off peak and high-cost units on peak. In effect, HPS allows utilities to shield customers from the reality of power generation, depending upon the level of coincident demand. End-use pricing has not reflected real-time costs, but new markets will deny utilities that luxury.
As with gas, new storage techniques will emerge to meet the demands of a competitive electric market: compressed air energy storage (CAES), battery energy storage systems (BESS), superconducting magnetic energy storage (SMES), and flywheel energy storage (FES).6 While not actually "storing electricity," these systems store energy to regenerate electricity and to capture the economics and operational benefits that actual electricity storage would provide.
CAES, by far the most well developed, relies on commercially available "off-the-shelf" combustion-turbine technology, modified to allow compression during the time the generator is not operating. Typically, a combustion turbine uses nearly two-thirds of its fuel energy to compress air while it is generating power. The CAES system uses inexpensive offpeak energy from base-load units to compress air in below-ground salt caverns or depleted hydrocarbon fields. During peak-demand periods, the expander turbine takes air from the caverns and gas fields to drive the generator, but uses much less fuel than a combustion turbine does.
"Virtual" electricity storage will arise through the use of financial instruments such as swaps and demand-side processes or products. In a storage swap, two parties would agree to purchase power at a certain time and then sell an equivalent amount at a later date (to each other or a third party). The parties could simply ramp their generation assets up or down at times that differ from those they would normally choose absent the agreement. This strategy allows the parties to exploit efficiency gains from temporal differences in their marginal production costs or supply availability.
Real-time end-use pricing will encourage customers to seek demand-side options to change their patterns of consumption and lower their bills. For example, by melting metal during offpeak periods or turning large industrial freezers off during onpeak periods (those with insulation that can hold temperatures for several hours), businesses can shift their consumption to lower-cost time periods. In effect, such customers can deliver capacity back to utilities when needed.
Electricity hubs and market centers will also evolve to serve the multitude of transactions inherent in a deregulated supply market. CAES and HPS plants might prove ideally situated to function as the physical "swing" for an electricity hub or market center. Hubs geographically situated close to CAES or HPS plants could attract more buyers and sellers than hubs that are unable to offer storage services from such facilities.
To attract customers, electricity need not remain in storage for a long period of time, such as a full calendar season. Today's "just in time" economy and real-time inventory management techniques render longer-term storage unnecessary. Electric customers will more likely desire services such as "instantaneous load balancing." In this sense, the electric industry may stand better prepared for competition than the gas industry, where time of day plays a less critical role in system operations.
Given the instantaneous nature of electric power flows, transmission operators require some direct or indirect control over generation assets to maintain reliability, protect against unscheduled transmission outages, and ensure network coordination. However, with the coming functional (and perhaps structural) separation between generation and transmission, grid operators will exert less direct control over generation resources. Ancillary services will provide grid operators with the means to maintain reliability under open access and vertical disaggregation.7
Historically, the cost of these ancillary services remained hidden and internalized within the vertically integrated utility structure. Now, the FERC's open-access policy for electric transmission will unbundle these services and price them separately.8 In fact, the FERC is applying lessons learned from the gas industry. The six ancillary services identified in the final rule in Order 888 have clear corollaries in the gas industry (see Table).
Transmission providers must provide or at least offer these six services, although the FERC will allow transmission customers to supply some ancillary services from their own generation assets or purchase them from third parties. This policy should spur formation of a competitive market for these generation-based ancillary services.
Storage technologies like CAES and HPS appear well-suited to support ancillary services such as load following, because they are designed to sustain frequent startup-shutdown cycles and variations in loading with a minimum of equipment stress or operational inefficiency. They operate like efficient shock absorbers (em helping to smooth out fluctuations in supply, demand, and price. A CAES plant is designed to cycle on a daily basis and to operate efficiently during partial load conditions. It can also swing quickly from a generation to a compression mode (effectively doubling a unit's swing capability). CAES plants are the most cost-effective generating technology at annual capacity factors ranging from 10 to 40 percent.
An increasingly competitive ancillary services market has developed in the United Kingdom since privatization and disaggregation of its electric industry. The National Grid Company (NGC) owns and operates the transmission system. It also operates an ancillary services business that executes contracts to buy ancillary services from independent providers (generators, the regional electricity companies, large consumers, or external pool members).9,10 For the most part, NGC employs a competitive bidding process to procure the ancillary services it needs to maintain target reliability levels at least cost. It then recoups the cost of these services through a customer surcharge or "uplift fee" on its unbundled transmission services.
Ancillary services contracts in the United States will most likely resemble those from across the pond, as stateside utilities unbundle generation from transmission and turn to an independent system operator (ISO) to manage the grid.11
The emergence of ISOs will likely increase interest in CAES and HPS plants, which could be dispatched to provide ancillary services and balance supply and demand.
* * *
Industry restructuring presents a two-edged sword to developers of new power plants capable of storing electricity. On the one hand, utilities and other potential customers feel logical concern about making long-term capital investments. On the other, electric storage offers strategic benefits under a regime of open access. The challenge lies in assessing these risks and benefits. Some of these storage technologies should make
the cut. t
Philip O'Connor is principal, and Erik Jacobson manager, of the Utilities/ Energy Division at Coopers & Lybrand Consulting. Mr. O'Connor served previously as chair of the Illinois Commerce Commission. Mr. Jacobson has worked as deputy director of the Division of Ratepayer Advocates at the California Public Utilities Commission. The authors prepared this article with funding from TPC Corp. (Tejas Power), Houston, TX.
1. EAI reports that storage use per field increased significantly from 1989 through 1993 compared to 1982 through 1986. The later period represents the industry after participants assimilated the effects of Order 436. Average monthly injections per field increased from 520 to 642 MMcf. Average monthly withdrawals per field increased from 531 to 644 MMcf, an increase of 23 and 21 percent, respectively, for the two periods. Energy information Administration, The Value of Underground Storage in Today's Natural Gas Industry, DOE/EAI-0591, March 1995, p. 26-27.
2. "Deliverability," the measure of the amount of gas that can be withdrawn from a storage facility over a period of time, has increased about 10 percent, from 61,718 MMcf/day in 1990 to 67,729 MMcf/day in 1993. "Working-gas capacity," the maximum amount of gas that can be stored in a reservoir, increased 4 percent, from 3,550 Bcf in 1990 to 3,695 Bcf in 1993, Ibid, p. 32.
3. Of the 20,746 MMcf added to daily deliver-ability from proposed storage projects for the period 1994-99, 68 percent (14,115 MMcf/day) comes from proposed salt-cavern storage projects. In 1993, total deliverability reached 67,729 MMcf/day (em 7,041, or 10 percent, from salt-cavern storage facilities. Ibid, p. 59.
4. Most salt-cavern facilities can cycle their entire working-gas capacity 5 to 10 times per year, as compared to once a year for storage facilities using depleted reservoirs. Ibid, p. 33.
5. Although interstate pipeline companies have historically dominated the underground storage business, over 53 percent of the 20,746 MMcf/day in proposed additions to storage withdrawal capacity are now made by independent developers, Ibid, p. 60-63.
6. CAES, BESS, SMES, and FES differ in the amount of energy they are likely to store, the rate at which they can deliver the energy, and the limits of their delivery durations. However, they all use "mechanical" or chemical conversions to facilitate storage, and they all provide for regeneration of electricity. SMES is the nearest to a purely electric storage system.
7. In its pro forma tariffs proposed on March 29, 1995, as part of its "Mega-NOPR" proceeding, the FERC defined "ancillary services" as "those services necessary to support the transmission of energy from resources to loads while maintaining reliable operation of the Transmission Provider's transmission system in accordance with Good Utility Practice." See, Promoting Wholesale Competition Through Open Access Non-Discriminatory Transmission Services by Public Utilities (Notice of Proposed Rulemaking an supplemental Notice of Proposed Rulemaking, Appendix B, Tariff Sheet No. 1, March 29, 1995, 70 FERC (pp 61,357.
8. As reported by Oak Ridge National Laboratory (ORNL), the cost of ancillary services ranged from .15 to .68 cents/Kwh and from 5 to 25 percent of generation plus transmission costs for a sample of 12 utilities. This large range stems, in part, from differing definitions of ancillary services between utilities. In addition, approximately 30 percent of ORNL's cost estimate of .41 cents/Kwh was for real power losses, a service that the FERC has decided not to require from transmission providers. Brendan Kirby and Eric Hirst, ORNL, Ancillary-Service costs for 12 U.S. Electric Utilities, March 1996.
9. Until recently, NGC also owned 2,088 megawatts of HPS plants, which it called upon to support the reliability of the grid. In January 1996, Southern California Edison Co. affiliate Mission Energy acquired First Hydro co. from the National Grid Co. for $1 billion. First Hydro, which controls 2,088 Mw of HPS assets in Wales, provides rapid generation response and will now sell electricity into the U.K. power pool on a competitive basis as an independent power producer.
10. Paul Davis, Coopers & Lybrand Consulting, London, Mary 10 1996. During 1995-96, revenues for ancillary services reached about $210 million, or .08 cents/Kwh. Strict comparisons between these ancillary services costs and the ORNL cost estimates cited previously are not appropriate because of differences in the definitions and elements of the ancillary services included in the cost figures.
11. For example, in their joint application to the FERC to form the Western Power Exchange (WEPEX), Pacific Gas & Electric Co., San Diego Gas & Electric Col, and Southern California Edison Co. describe how the WEPEX ISO would purchase operating reserves on the market through a daily auction, and ancillary services such as reactive power/voltage control through an annual bid process. Pacific Gas & Electric Co., San Diego Gas & Electric Co., and Southern California Edison Co., Joint Application for Authorization to Convery Operational Control of Designated Jurisdictional Facilities to an Independent System Operator, filed at the Federal Energy Regulatory Commission, April 29, 1996, p. 63.
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.