MEMORANDUM

TO: Interested Persons

FROM: Dick Watson

SUBJECT: Summary of Power Committee's June 6, 2000 Roundtable Discussion on Using End-user Resources to Address Power Supply Adequacy Problems

    On June 6, the Power Committee held a Roundtable Discussion on the use of end-user resources (self-generation, load shifting, load reduction, partial or full curtailment) to address expected power supply adequacy problems. This meeting was motivated by the Council's power supply adequacy study released earlier this year. Key conclusions of the study were:

• The probability of some level of power supply inadequacy is becoming uncomfortably high – reaching 24 percent by 2003.
• The causes of these problems are combinations of extreme cold weather, poor hydro conditions and possible forced outages of generating units. With the exception of unscheduled generation outages, these problems are largely foreseeable, i.e., you can project on at least a day ahead basis that supplies are likely to be extremely tight.
• The characteristics of the problems are mixed – energy deficiencies (about half); capacity deficiencies (about 20 percent), and operating reserve violations (about 30 percent).
• The frequencies and durations of these events are relatively small, although their consequences could be quite costly.
• Because of the limited frequency and duration, it is unlikely that the expected market price of power would support building sufficient capacity to address these kinds of problems.
• As of yet, relatively few end-users see real-time market prices, thereby muting demand response to impending supply inadequacy problems
• Finding means of facilitating voluntary end-user participation in addressing supply inadequacy problems appears most promising. Possible examples include: dispatch of existing end-user generation; load management efforts (primarily aimed at achieving load reduction but some peak shaving/shifting as well); partial or full curtailment.

    The purpose of the roundtable was to begin to identify the opportunities for and barriers to using end-user resources to address adequacy problems. The participants were in three groups: utilities, industrial and municipal end-users, and energy service providers. The participants were:

End-Users: Jim Stromberg; Columbia Falls, Aluminum; Ken Canon, Industrial Customers of Northwest Utilities; Tom Yarborough, Weyerhaeuser; Eric Larson, Wah Chang; Al Worman, Boeing; John Siekas, Oregon Steel; Curt Nichols, City of Portland; Michelle Wynne, consultant to customers providing demand side reserves to the California ISO (by phone)

Utilities: John Pyrch, BPA; Matt Northway, Eugene Water and Electric; Stan Watters; PacifiCorp; Daren Sanders, BC Hydro (by phone)

Energy Service Companies: Alan Kirn; Johnson Controls; John Gibson, Tri-Gen, and David Slifer, Planergy

    Some of the key points raised during the discussion are:

We're not in the electricity business. From the industrial end-user's perspective, electricity is an input to their business and they want to focus on that business. It's somebody else's business to provide them with a reliable supply of electricity.

But we could be interested. That said, however, there was recognition that as more of them are exposed to market prices of electricity, they will have to take a more active approach to managing their electricity purchases. There was also recognition that there is an electricity price at which taking some actions to reduce their loads makes sense.

Everybody is different. Each industry is different and, within an industry, every plant is different. The opportunities for and costs of reducing loads are inherent in the industrial processes used and the design and operating protocols of particular plants.

Everybody (probably) has some potential. Everyone probably has some potential for reducing loads when the price is right. Some industries might have the ability to totally suspend operations for a period. (examples: air products; rock crushing -- both of these are more-or-less continuous production processes, but easily interrupted because the feedstock and intermediate products hold their state under ambient conditions or can be quickly removed from the processing equipment) Some industries might have the ability to suspend part of their operations temporarily (example: paper plants with mechanical pulping could build of an inventory of pulp and temporarily suspend pulping operations if necessary). Industries such as steel or aluminum may be able to curtail loads by cycling production.

Self generation? It was estimated that there is something like 400 to 450 MW of installed self-generation capacity in the region's industries. [Council staff estimates a somewhat larger number]. This does not include backup generation in hospitals, large office buildings, and so on. Nor does it include generation like that at Portland's sewage treatment plant and other similar facilities ( ~ 10MW regionwide). It is not clear how much of that generation could actually be dispatched to address power supply adequacy needs. Much of it is currently being used internal to the host plant. Dispatching that power into the grid would require curtailing the load in the host plant. Some of the systems use backpressure turbine-generators that require steam load (or venting) to operate. On the other had, some systems may not currently be in use or not used to full capacity and could be dispatched into the grid. However, operation of inactive on-site generation may be subject to air or water permit limits. The amount of self-generation installed in the region that could be dispatched and the barriers to using that generation need to be better understood. [Bonneville has a survey underway which may start to answer these questions for service areas of its utility customers]

Hedgerow by hedgerow. The implication of the above points is that mobilizing end-user participation will not yield to cookie cutter approaches. This will take an individualized approach to major customers. This is time and people-intensive on the front end.

Procedures and protocols need to be in place, personnel trained, metering installed. Load reduction cannot be counted on if implemented on an ad hoc basis. Procedures and communications protocols have to be in place and, ideally, tested and benchmarked. Key personnel have to be trained. Metering is necessary to determine how much load reduction is achieved. Fifteen minute metering was cited as adequate. However, aggregated load curtailment programs have been put together in three to six months. Periodic testing is needed to ensure that curtailment will be available if needed. For example, Planergy tests its programs annually.

Season matters. Some processes are sensitive to ambient temperature or other seasonal factors. Things that you might be able to do in the summer you might not be able to do in the winter or vice versa.

The state of the economy may matter: Industries operating at or near full capacity may be unwilling to curtail load that affects production. Lost production cannot be recaptured under full capacity conditions.

Lead time matters. Lead-time is important for both the end-user and the utility. For the most part, we are not talking about using load reduction for operational reserves for system stability purposes. (The exception is the ability to interrupt the DSIs for short periods for system stability) These situations require a very rapid response. The kinds of situations we are looking at are ones in which we could anticipate a day or two in advance the likely need/opportunity for load reduction. For example, it's been a poor water year and an "arctic express" is forecast for the Northwest. Or, very high temperatures are forecast for California and a major generator is out of service.

    The end-user would like as much time as possible to prepare for and implement a load reduction. The utility, on the other hand, would like to minimize lead-time. It bears the risk that the anticipated need/opportunity (prices) may not materialize while it has committed to purchase load reduction for a particular price. The longer the lead-time, the greater the risk. Experience was cited where it is possible to implement load reduction with only a few hours notice. Some period of time is necessary to establish a baseline from which the load reduction can be measured. [Outstanding question: Is the lack of a Northwest hourly market a problem? Are there significant discontinuities between California markets and the Northwest such that basing payment for load reduction on California market prices constitutes a significant risk for utilities or end-users in the Northwest?]

Frequency matters. It has been observed in California thatthe responsiveness of end-users to calls for demand reduction diminishes with the frequency of calls within a period. This suggests the need for a large portfolio of participants so that frequency with which any one participant is called upon can be kept to acceptable levels.

Aggregation matters and is possible. Control area operators do not have the time to call up small increments of load reduction. Utilities would like to be able to call on significant "chunks" of load reduction – tens or hundreds of megawatts at a time. Load reduction opportunities typically come in smaller chunks. However, the examples were given of third party aggregators who aggregate load reduction and sell that to the utility (Planergy). The contractual arrangements are between the utility and the aggregator and the aggregator and the end-users. The aggregator takes responsibility for achieving and verifying the load reduction. For example, see: http://www.planergy.com/Energy%20Mgmt.htm  Similarly, BC Hydro also cited their experience with curtailable contracts with 10 customers for 200 MW of demand reduction. In their programs, customers nominate a "trigger price" at which they are willing to reduce load. The savings are split between Hydro and the customer.

New Technology Might Help. New information technology could help. For example, the "Demand Exchange" is an internet based exchange in which utilities can indicate how much demand reduction that are willing to buy at what price during what hours.  Participants must have demonstrated their ability to a level of demand reduction. Both Bonneville and PGE have pilots planned for this year. Software is available which provides an automated system by which utilities or third party energy managers can manage load reductions.

What about new generation? Some expressed the opinion that these problems cannot be solved solely through reliance on end-users and that new generation would be needed. It was also pointed out that the Council's analysis and that of the California Energy Commission found that the electricity market prices would not support the development of new capacity to address relatively short duration, relatively infrequent power supply problems. Some suggested that the old system of paying for little-used capacity and spreading the costs to all customers might not be so bad. The question was also raised as to whether Regional Transmission Organizations (RTOs) might be the appropriate vehicles for providing a market for such capacity.

What about investment in efficiency? One participant pointed out that the discussion had focused on curtailment and, while that was understandable given the relatively low capital cost of curtailment, efficiency improvements could also play a role. That role was acknowledged. However, like generation, these investments must meet high investment hurdle rates on the basis of the average energy prices over the period that savings are produced. The question was asked whether capital constraints were an issue. The answer was yes. Reference was made to innovative tariffs in which money for efficiency improvements is earmarked for the company or the industry. (Puget Sound Energy) The same is true of Oregon's system benefits charge. These approaches lessen capital barriers. It was further asked whether third party financing could play a role. The answer was yes, sometimes.

What about outsourcing your energy supply management? The question was also asked whether outsourcing your energy management was potentially attractive, i.e., turning over the decisions about energy supply, demand management, risk management etc to a third party who is in a position to trade off between energy prices, investments in efficiency, demand management and so on. Enron, for example, offers such services. There was some but limited experience with outsourcing. On the face of it, it would appear that outsourcing could have potential for helping engage the end-user in responding to supply adequacy issues.

This world is very new. The Northwest has never seen the (electricity) prices we’re seeing today. As economics change, opportunities change, and we haven’t really explored those opportunities yet. We’re in a new phase. Utilities are beginning to talk to their customers about those opportunities. At the end of the day, reliability comes at a cost as does the lack of reliability. The solution lies in finding the appropriate balance. As of yet, the potential for demand reduction is relatively undeveloped. At some point the supply curve for demand reduction will cross that of generation.

Please direct questions or comments to Dick Watson at the Northwest Power Planning Council.

^ top