UNITED STATES PATENT AND TRADEMARK OFFICE
`_____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________
`
`HONEYWELL INTERNATIONAL, INC.
`
`Petitioner
`
`v.
`
`ALLURE ENERGY, INC.
`
`Patent Owner
`_____________
`
`Case No. IPR2016-___
`Patent No. 8,509,954
`
`PETITIONER’S EXHIBIT NO. 1009
`
`

`
`LBNL-48374
`
`
`
`
`
`Demand Responsive Programs -
`An Emerging Resource for Competitive
`Electricity Markets?
`
`
`
`Dr. Grayson C. Heffner and Charles A. Goldman
`
`
`
`Environmental Energy Technologies Division
`Ernest Orlando Lawrence Berkeley National Laboratory
`University of California
`Berkeley, California 94720
`
`
`
`August 2001
`
`
`
`Download from: http://eetd.lbl.gov/EA/EMP/
`
`
`
`
`
`In the Proceedings of the International Energy Program Evaluation Conference (2001 IEPEC)
`
`The work described in this study was funded by the Assistant Secretary of Energy Efficiency and Renewable Energy, Office
`of Power Technologies of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098.
`
`
`
`Honeywell Exhibit 1009, Page 1
`
`

`
`Demand Responsive Programs -
`An Emerging Resource for Competitive Electricity Markets?
`
`Dr. Grayson C. Heffner, Lawrence Berkeley National Laboratory, Berkeley, CA
`Charles A. Goldman, Lawrence Berkeley National Laboratory, Berkeley, CA
`
`
`
`
`
`ABSTRACT
`
`
`The restructuring of regional electricity markets in the US has been accompanied by numerous
`problems, including generation capacity shortages, transmission congestion, wholesale price volatility,
`and reduced system reliability. These problems have created significant new opportunities for
`technologies and business approaches that allow load serving entities and other aggregators, to control
`and manage the load patterns of their wholesale or retail end-users. These technologies and business
`approaches for manipulating end-user load shapes are known as Load Management or, more recently,
`Demand Responsive programs.
`Lawrence Berkeley National Laboratory (LBNL) is conducting case studies on innovative
`demand responsive programs and presents preliminary results for five case studies in this paper. These
`case studies illustrate the diversity of market participants and range of technologies and business
`approaches and focus on key program elements such as target markets, market segmentation and
`participation results; pricing scheme; dispatch and coordination; measurement, verification, and
`settlement; and operational results where available.
`
`Introduction
`
`
`Demand Responsive Programs, once called Load Management, have recently re-emerged as an
`important element in the fine-tuning of newly restructured regional electricity markets. These programs,
`which can include everything from direct control of small customer end uses to voluntary load shedding
`by large commercial/industrial customers, have experienced explosive growth in number and
`participation over the past two years (Hirst 2001). The resurgence of demand responsive programs
`stems directly from their rediscovered value as a dual hedge against reliability risks such as generation
`shortfalls and transmission congestion as well as financial risks such as wholesale price spikes. The
`versatility of these programs has attracted the interest of many market participants in both traditional and
`newly competitive electricity supply markets.
`The research described here is part of a larger LBNL effort, funded through the US DOE
`Restructuring and Electricity Reliability Program to examine the potential role of customer demand
`management in an overall program of power system reliability improvement. The objective of this
`particular project was to increase understanding of how new technology and new business approaches
`are being combined to create rapid innovation and explosive growth in the load management industry.
`For reasons of both cost and time, the authors chose a “case study” approach to this initial survey
`of demand responsive programs. Three criteria were used to select the cases - innovation in technology
`
`
`
`Honeywell Exhibit 1009, Page 2
`
`

`
`or business approach, diversity of market players in traditional and competitive electricity markets, and
`type of demand response program (see Table 1).
`
`
`
`Honeywell Exhibit 1009, Page 3
`
`

`
`
`
`
`
`Case Study
`BPA
`
`Cinergy
`
`PJM
`
`Puget Sound Energy
`
`Table 1. Overview of Selected Demand Response Programs
`
`Type of Organization
`Federal Power
`Marketing Agency
`Investor-owned G, T, &
`D Utility
`Independent system
`Operator
`
`Investor-owned T&D
`Utility
`
`Rural Electric
`Cooperative
`
`Type of Program
`• Large Customer “Demand Exchange”
`
`• Voluntary Interruptible programs for
`medium & large C/I customers
`• Voluntary interruptible programs –
`Emergency and Economic Options – for
`large customers
`• AMR-based Energy Mgmt Advisory
`Program for all customers
`• Thermostat control for residential customers
`•
`Interruptible Program for large customers
`• Residential Load Control
`
`Wabash Valley
`Power Association
`
`
`We developed a standard survey protocol and conducted in-person interviews with program managers.
`Program characteristics of particular interest included:
`• Motivation of the LSE or ISO for offering the program
`• Exact nature of the pricing scheme
`• Program operational details and results
`• Customer participation and results, including customer retention
`• Program costs, including hardware and communications requirements
`•
`Industry issues and technology development needs
`
`
`Overview of Selected Programs
`
`Bonneville Power Administration
`
`Utility Motivation: Bonneville Power Administration’s (BPA) Power Business Line is transitioning its
`1999-2000 Demand Exchange Pilot Program into a full-scale, system-wide offering. The program
`provides needed flexibility as BPA operators seek to optimize four operating objectives – least-cost,
`reliability, fish & wildlife management, and energy savings.
`Program Design: The Demand Exchange, or DEMX, is an internet-based auction site where
`participants are alerted to real-time, day ahead, and two-day-ahead pricing and then can post their
`willingness to voluntarily curtail loads at a given price (Gilbert, 2000). The BPA version of the DEMX
`offering has two parts – a Voluntary Curtailment Option and a Pre-Purchase Option. The Voluntary
`Curtailment Option is a classic “Quote Scheme”, while the Pre-Purchase option is very similar to a “Call
`Scheme1” (Hairston, 2001a, 2001b).
`
`
`1 Many interruptible and curtailable programs have separate ”quote” and “call” options. A “quote” program allows the
`customer to specify when and at what price they are willing to reduce load. A “call” option requires the customer to reduce
`load when called upon or face penalties.
`
`
`
`Honeywell Exhibit 1009, Page 4
`
`

`
`Program Operations and Results: Both the Voluntary Curtailment Option and the Pre-Purchase
`Option are operated on a day-ahead basis. BPA seeks out end-users whose electricity costs make up a
`large portion of operating costs and who have some flexibility to change their pattern of electricity use.
`Participants include municipal and investor-owned utilities serving as aggregators, direct-serve
`customers, and several large retail end-use customers of the resellers. Pulp and paper and
`aluminum/basic metals were the two most represented SIC codes.
`As of April 2001, BPA has signed up 14 customers with ~525 MW of potential demand
`reduction potential. BPA has received over 6500 MWh of load reduction over the life of the program.
`Over 90% of these load reductions have occurred since December 2000 in response to high wholesale
`prices. BPA is currently averaging nine MW per hour of load reductions and has a target of ~25 MW
`per hour in 2001.
`
`Conclusions: The most common use of these programs is as a hedge against energy shortages from the
`hydro system when wholesale purchase prices are high. When this is the case, usually during the winter
`peak months, BPA will initiate both day-ahead and two-day-ahead load curtailment requests covering
`many hours or even entire days. Such extensive use of the curtailment option for days or even weeks
`may result in customers reducing their overall manufacturing output. BPA is tracking closely any labor
`or regional GDP impacts that could be traceable to this use of the program.
`
`Cinergy
`
`
`Cinergy has consolidated its demand management programs into one umbrella offering - the
`PowerShare Pricing Program. PowerShare provides customers with a menu of choices and is a
`market-based program of financial incentives that encourages Cinergy customers to reduce summer peak
`load (Darnell, 2001, 2000).
`
`Utility Motivation: Cinergy has arguably the most aggressive demand management program in the
`U.S., driven in part because of their experience during the disastrous summer of 1999 when the
`wholesale electricity market experienced extreme price volatility in the Midwest. High temperatures,
`low rainfall, extended heat waves, and generation outages all combined to expose Cinergy and its
`customers to extreme price spikes and severe reliability problems. After Cinergy’s well-publicized
`contract default and subsequent exposure to liquidated damages, the Company made a concerted effort
`to rapidly grow their demand management programs.
`
`Program Design: To participate in the CallOption, the customer must have a potential load reduction of
`at least 500 kW. The customer will select a Strike Price (10, 30, 60, or 90 cents per kWh) based upon
`their own estimate of the costs of complying with curtailments. When the day-ahead market prices are
`projected to be greater than the Strike Prices, Cinergy can call the option by notifying customers by 4
`pm the day ahead. In exchange for participation, customers receive a guaranteed premium plus an
`additional Energy Credit whenever they are called. The penalty for non-compliance is payment of
`market prices (assuming energy is available) or a penalty cap of $10 per kWh. Customers can specify a
`Firm Load Level; identify a generator (2 MW minimum) to operate; pledge a specific end use or process
`to shut down; or pledge a fixed reduction in their pro forma load. . Customers may also select among
`three levels of curtailment frequency and duration as specified in Table 2 below.
`
`
`
`Honeywell Exhibit 1009, Page 5
`
`

`
`
`
`
`
`Table 2: CallOption Choices Offered by Cinergy’s PowerShare Pricing Program
`
`Option
`
`Max Duration
`
`Period (time)
`
`Calls per year
`
`Consecutive days per week
`
`CallOption A
`
`8 hours
`
`CallOption B
`
`4 hours
`
`CallOption C
`
`8 hours
`
`12-8 pm
`
`2-6 pm
`
`12-8 pm
`
`12
`
`12
`
`4
`
`3
`
`3
`
`4
`
`
`
`The QuoteOption is much less complex. It is designed to be a no-risk proposition for the same
`group of customers eligible for the CallOption but who are reluctant to commit to a Firm Load Level or
`be subject to non-compliance penalties. Customers pre-specify only the type of load block (load
`reduction from a pro forma load shape or generator of minimum 2 MW size to be switched on) and a
`Strike Price below which they are not interested in participating. The QuoteOption is a day-of program.
`Cinergy provides price quotes for the same day and interested customers must respond with an estimate
`of voluntary load reduction within one hour.
`Energy credits are calculated as the product of kW reduction, hours of curtailment, and the
`selected Strike Price. A single customer with 500 kW of curtailable load choosing CallOption and a
`$.15/kWh strike price would thus receive $14,000 of annual premium and $600 in energy credits each
`time they successfully curtailed their load. In the Shared Energy Credits scheme, customers can save up
`to 50% of the difference between their Strike Price and the projected hourly wholesale price.
`
`Program Operations and Results: The Cinergy PowerShare website is a key vehicle for program
`operations. The web site for each customer contains their pro forma load shape plus price quotes for
`both the day-off and day-ahead program options. The customer accesses the web site and for the
`QuoteOption they can then nominate their loads for that day. The CallOption customers are also given
`notification by e-mail, cell phone, pager, or fax of an impending curtailment the next day.
`Cinergy currently offers the PowerShare program to Cinergy customers with peak demands
`greater than 500 kW. Beginning in 2001, Cinergy will begin offering two new variants of the basic
`PowerShare program – PowerShare Basic and PowerShare Lite – to 750 additional customers with
`summer peak demands that range from 200 to 500 kW. These new programs target customers that don’t
`yet have hourly interval metering or a dedicated phone line. The strategy is to offer a way for customers
`to put smaller blocks of load on offer for either the day-ahead Call or day-of Quote options. Readily-
`identifiable and measurable blocks of load, such as retail lighting or individual motors or drives, are
`likely to be the most-common load blocks that are targeted.
`As of late 2000, over 90% of Cinergy’s 312 large customers were participating in one or more of
`the PowerShare options. These customers together comprise 2500 MW of subscribed load. Cinergy
`estimates that as much as 600 MW would be available from these customers on a summer peak day with
`high wholesale prices. The per-customer load reduction expected averages about 10-15%, with some big
`customers (such as steel producers) able to drop as much as 50% of their pro forma loads.
`In Summer 2000 there were over 300 customers participating with an estimated curtailable load
`of 440 MW. However, the weather was so mild that the program was not operated at all during the year.
`In 1999, with prices as high as $850/MWh, Cinergy received as much as 200 MW from the pilot
`program participants.
`
`
`
`Honeywell Exhibit 1009, Page 6
`
`

`
`
`Conclusions: The Cinergy program illustrates that high market penetration can be achieved among large
`C/I customers with targeted and customized program offerings. However, the restricted annual number
`of hours (only 96 for the most-severe CallOption offering) makes them more suitable for emergency as
`opposed to economic operations. The extension of these offerings to medium-size C/I customers without
`interval metering is a key development to watch in 2001.
`
`PJM Interconnection, L.L.C.
`
`
`PJM Interconnection, L.L.C. became the first operational Independent System Operator in the
`U.S. on January 1, 1998, managing the PJM Open Access Transmission Tariff and facilitating the Mid-
`Atlantic Spot Market.
`
`ISO Motivation: In response to a May 2000 Federal Energy Regulatory Commission (FERC) order
`that directed ISOs to expedite procedures for including distributed and demand-side resources in bulk
`power markets, PJM formed the Distributed Generation User Group (DGUG). The PJM DGUG focused
`on developing pilot demand response projects that would help identify key issues and requirements for
`future system-wide programs (Bressler, 2001).
`
`Summer 2000 Pilot: The Summer 2000 temporary pilot provided compensation for end-use customers
`willing to reduce energy consumption from the PJM system during emergency conditions. Upon
`declaration of a Maximum Generation Emergency Event, PJM would request participants to reduce
`load, and all participants that did so would be paid the higher of $500/MWh or their zonal Locational
`Marginal Price (LMP) for the reduced consumption.
`The Summer 2000 Customer Pilot Program was operated on a strictly emergency basis. Since
`PJM has a $1,000 per MWH price cap on all in-system purchases, the criteria for operation was a
`Maximum Generation Emergency Event, which immediately precedes going out of system to purchase
`Emergency Energy at any cost.
`
`Summer 2001 Pilot: Greater efficiency would exist in the PJM marketplace, and indeed the existing
`$1,000/MWh cap on generator bids might not even be necessary, if the load in PJM could respond to
`high prices and reduce demand during times of short supply. The main obstacle to tapping the potential
`of price-responsive load in the PJM system is the fact that most end-use customers are not exposed to
`real time prices. In addition to end-users not being exposed to real-time price signals, Load Serving
`Entities (LSEs) may pay more for energy in the wholesale market than they collect from their retail
`customers during times when the wholesale energy price in the PJM market rises above the applicable
`retail rate. The Summer 2001 Load Response Pilot Program includes two options (Emergency and
`Economic) and will test whether the ISO, working directly with both LSEs and customers, can begin to
`tap these potential savings.
`PJM expects that the Emergency Option will attract principally large industrial customers and is
`similar to the program offered in 2000. The Economic Option is based on the economic decisions of the
`PJM market participants in response to market conditions. Participants in this Pilot Program are
`responsible for determining when load reductions will take place and implementing the reductions
`should pre-determined conditions arise. The prime indicator will be the Locational Marginal Price
`(LMP) of energy on the PJM system. In order to maintain system control, PJM operators will be know
`the amounts of load expected to be reduced at different price levels. (These amounts may change on a
`
`
`
`Honeywell Exhibit 1009, Page 7
`
`

`
`daily basis.) Each PJM market participant is therefore responsible for informing PJM daily of the
`amount of load reduction for which they have contracted in each PJM zone, and the price at which that
`load may be reduced. A web page will be created through which market participants may submit the
`expected amount of load reduction, together with the LMP values at which the load may be reduced.
`PJM will then compile daily aggregate load reductions on a zonal basis for use in operations.
`
`Conclusions: The PJM philosophy is that economic programs will lead to a shift away from Active
`Load Management (ALM)2 and emergency-type programs and towards real-time pricing and interval
`metering for all customers. This will lead to power systems and markets that are decentralized and
`market-driven, with less of a need for highly centralized ISO-style command and control procedures.
`Roles and responsibilities among organizations participating in demand response programs
`offered by newly formed ISOs are not fully resolved. Some Transmission Owners and LSEs have raised
`concerns that they will lose distribution revenues as a result of these new programs. Distribution
`companies are uncertain as to whether they want to get involved and, if so, in what role – aggregator?
`Some LSEs – notably PP&L – are already very active in the area of price-responsive demand
`management and see the ISO as a competitor.
`
`Puget Sound Energy (PSE)
`
`
`Puget Sound Energy is a $2 billion investor-owned utility providing electricity, natural gas, and
`energy related services to 1.2 million homes and businesses in Washington. PSE has been steadily
`implementing integrated technologies including Automated Meter Reading (AMR) and advanced
`Customer Information Systems (CIS).
`PSE has two distinctive programs now underway:
`• Personal Energy Management Program. This is essentially an information program. Customers
`are provided an informational bill showing usage by time of day and are given advice on how to
`shift usage away from on-peak periods. Customers can then track their progress in shifting their
`usage from bill to bill.
`• Home Comfort Control Program is a cooperative pilot project between Carrier, Silicon Energy,
`Schlumberger CellNet and PSE. About 110 residences were fitted with controllable thermostats
`that could be remotely accessed to adjust the set point. Customers agreed to allow the utility to
`adjust the thermostat by up to 4 o F. A key feature of the program was a PSE-managed web site
`from which set point adjustment commands could be sent and thermostat status monitored using
`the AMR communications medium. Customers were given the option of over-riding the utility
`control of the set point but were charged a penalty when they did so.
`
`
`Motivation for PSE: The main drivers for PSE are the ability to offer value-added services to
`customers, and the flexibility gained with a customer communications system that can provide dynamic
`pricing.
`
`
`2 Active Load Management programs are defined as those that can be directly dispatched by the ISO or the LSE. The ALM
`programs are a legacy resource embedded in the operations of PJM’s LSE members, and include three types of programs:
`Direct load control (DLC), where load management is initiated directly by the LSE’s control center using a communications
`signal to control equipment such as air conditioners and water heaters; Firm Service Level (FSL), where load management is
`achieved by a customer reducing its load to a pre-determined level upon notification for the LSE’s control center; and
`Guaranteed Load Drop (GLD), where load management is achieved by a customer reducing its load by a predetermined
`amount upon notification from the LSE’s control center.
`
`
`
`Honeywell Exhibit 1009, Page 8
`
`

`
`Motivation for Customers: Customers have embraced the concept. PEM was originally contemplated
`as a four-month pilot, but it has been so successful that it has been continued indefinitely. A business
`version – “Business Energy Management” – is also available. Once time-of-day rates are introduced, bill
`savings will become a motivator for customer participation.
`The Comfort Home program was also very popular – especially the value-added capability to
`remotely access and control their thermostat for pre-heating or other purposes. Customers appreciate
`being able to log on to the internet, see the temperature and temperature setting in their house, make
`their own adjustments to the set point, and have the power to over-ride the utility if they want to (this
`happened only a few times).
`
`Current Customer Participation and Results: As of late 2000 there were 400,000 customers with
`AMR who were eligible for the PEM program. The early load impact results (based on November,
`December, and January) indicate a 3-4 % shift in usage from the “expensive” to the “economy” and
`“bargain” periods.
`The Home Comfort Control program pilot was conducted between February and April 2000.
`During this period the 105 voluntary participants experienced 41 two-hour “setback episodes”, during
`which their thermostat control was adjusted either 2o F or 4o F lower than the usual set point. Customers
`over-rode the utility control on only three instances and in most cases the customers did not even notice
`that utility control was exercised. A load impact regression model (Puget Sound Energy 2001) was used
`to analyze the demand impacts of morning, mid-day, and evening “set-back episodes” of both 2o F and
`4o F. Although the results varied between these cases, reductions of 1.2 to 1.6 kW were observed for
`morning and evening setbacks of 4°F. Energy savings were modest – about 1-3 kWh for the 4°F setback
`and negligible for a 2°F setback.
`
`Conclusions: Both types of programs can work wherever prices are volatile and the regulatory
`environment is supportive. However, both programs require significant investment in a backbone
`communications system such as AMR. Since this experiment was for residential space heating only, the
`potential for summer air conditioning demand reduction was not measured.
`
`Wabash Valley Power Association (WVPA)
`
`WVPA distributes electricity to member co-ops who in turn serve 200,000 residential,
`commercial and industrial customers located throughout Indiana, southern Michigan, and northwestern
`Ohio. WVPA operates two load management programs: a commercial-industrial voluntary interruption
`program (Customer Payback Plan); and the “It Pays to be COOL” residential air conditioner and
`electric water heater control program (Mizelle 2000, 2001).
`
`Motivation for WVPA: Wabash offers these programs as part of their efforts to keep power costs
`down by providing a hedge against wholesale price volatility. In 1999, which was a very bad summer in
`the Midwest, WVPA saved millions of dollars in avoided high-price wholesale purchases as a result of
`its load control program. They calculate that in just one week during June 1999 the water heater program
`paid for itself by saving over $500,000 in expensive power purchases.
`
`Motivation for Customers: The customer participates to save money and support the utility’s efforts.
`Participation in the COOL (Conserve Our Overall AC Load) Program earns residential customers an
`annual payment of $25.00. Customer Payback Plan participants can save $250/MWh for all reduced
`usage or generated energy when an “energy management period” is declared. Program literature uses the
`
`
`
`Honeywell Exhibit 1009, Page 9
`
`

`
`example of a customer with a 250 kW back-up generator that operates for 100 hours at utility request.
`The customer would make over $6,000 in “payback rewards” ($250*100*250/1000) plus the savings in
`avoided power purchases.
`
`Program Design and Operation: Both programs are focused on the summer season and are designed
`specifically as a hedge against wholesale price volatility. Customer Payback Plan participants must be
`over 50 kW to participate but do not have to have hourly interval meters. Interested customers work
`with customer reps to conduct a Facility Review and work out ahead of time very specific load
`curtailment strategies that would yield the pledged load reduction during summer afternoon peak
`periods. Customers with auxiliary or emergency generators are especially sought-after. The utility
`notifies the customer by 4 pm the day before and energy management period is expected; frequency is
`no more than 10 days per summer. Following an energy management episode, utility staff will use a
`variety of methods to estimate the amount of reduced energy or the output of the generator. These
`methods might include, depending on the customer, direct metering, analysis of demand and energy
`components of bills, timers or data loggers on specific end uses, and examination/comparison with
`baseline or benchmark data collected during the pre-season Facility Review. Obviously, these non-
`interval-metered methods require a high degree of collaboration and trust between the customer and the
`utility representative.
`
`
`Results: The residential appliance control program currently delivers an estimated 30 MW of load relief
`in summer and 20 MW of load relief in winter. As of 2001, the Customer Payback program included
`250 customers capable of delivering as much as 30 MW of load curtailment. The agricultural irrigation
`pump load control program has 130 participants and about 30 MW of controllable load. There was no
`need to operate any of these programs during Summer 2000, as it was unusually cool and wholesale
`prices stayed low.
`
`
`Conclusions: Co-ops have a unique niche in terms of serving rural loads. Over 31,000 MW of load –
`especially in the Midwest – is served by member-owned G&T utilities and distribution co-ops. These
`co-ops enjoy usually close relationships with their customers, where cooperation and trust can be real
`factors in the design and implementation of demand management programs. Wabash Valley Power
`substitutes simplicity in program design, a good multi-channel communications program, and what is
`basically an “honor system” for the large amounts of interval metering and other M&V hardware
`observed in other programs.
`
`Comparison of Key Program Design Features
`
`
`Table 3 summarizes key program design features of our selected demand response programs.
`• Target Markets: “Mass market” (Violette 2000) programs for small customers vs. large
`customer demand responsive programs
`• Dispatchability: Utility-controlled vs. customer-controlled loads
`• Resource Firmness: Call (participation is pre-paid) vs. quote (participation is fully voluntary)
`programs
`• Operational threshold: Emergency vs. Economic programs
`• Exposure to & assignment of forecast risk: Day-ahead vs. day-of vs. real-time demand and
`prices
`• Role of aggregators/Third Parties: How many relationships are in play?
`
`
`
`Honeywell Exhibit 1009, Page 10
`
`

`
`• Shared savings scheme: Flat rate or variable according to market conditions.
`
`
`
`Based on these cases, there is considerably more emphasis on large customer programs than on
`small customer or “mass market” programs. However, this may change as new technologies such as
`dispatchable thermostats and AMR advance in performance and affordability. Customer-controlled
`demand response programs continue to be much more common than utility-controlled alternatives, at
`least for the large-customer offerings. This too may change as “hybrid” approaches such as “permission-
`based control”3 and dispatchable back-up generators become more popular. The degree of firmness of a
`curtailable load block is generally thought to be higher with “call” programs, where the customer is
`provided a reservation payment, then with fully voluntary “quote” programs. Another issue is the
`relative cost per MWH of call vs. quote programs and the optimal split between “reservation” and
`“performance” payments.
`An important change over previous load management programs is the emergence of Economic
`vs. Emergency approaches to demand management. Most of the programs reviewed here are primarily
`economic in nature, as they seek to minimize the amount of expensive energy that must be purchased
`during periods of wholesale price volatility. In our five cases, only one – PJM Interconnection – was
`based on Emergency or Reliability needs. For some program administrators, the underlying philosophy
`is that demand response programs may provide a sufficient hedge of price-driven demand reduction in
`the long term that will minimize the need to implement demand management for reliability reasons.
`Forecast risk is a not-well-understood issue that is revisited in the Research Avenues discussion
`below. Many of these programs are based on day-ahead or two-day-ahead price projection or strike
`prices which customers pledge load reductions in response to. Any risk that the spot price will be lower
`than the projected or strike price currently devolves onto the program sponsor.
`Aggregators and “DMSCOS/CURTAILCOS” are new organizations that can potentially play a
`vital role in realizing the potential of demand response programs. Of these five cases only Cinergy has
`carved out a specific role for aggregators and other third parties to play. Marketing and recruiting
`participants for demand response programs may be an important niche for third parties.
`We can observe numerous approaches to pricing and shared savings in even the few case studies
`reviewed here. Across programs, payment levels and the design of the payment schemes are
`complicated, non-uniform, and seemingly ad hoc.
`
`Promising Research Avenues
`
`
`Based on this initial review of selected demand response programs, we describe several
`promising areas where additional analysis and/or research would be useful to policymakers and program
`designers.
`
`Analysis of Variations in Participant Payment and ”Shared Savings” Schemes
`
`
`
`Table 3 indicates a remarkable range in potential payment levels for participants in commercial
`and industrial load curtailment programs. Flat payments range from $100/MWh for Cinergy’s lowest
`Strike Price up to $500/MWH for PJM’s Emergency Pilot Program. Some programs offer a variable
`credit based on a share of the difference between a Strike Price or flat rate and the day-ahead forecast,
`day-of forecast, or actual hourly spot price. A quick calculation reveals that for any of the three days
`
`3 “Permission Based Control” or “Positive Control” requires that customers provide permission before a end-use load control
`switch is activated. The exchange of utility request and customer permission typically takes place via the inter-net or pager.
`
`
`
`Honeywell Exhibit 1009, Page 11
`
`

`
`during the summer of 1999, when PJM prices approached the $1000/MWH cap (June 7, July 6, or July
`29. See PJM 2000), the payment for the exact same customer participating in the different programs
`described here could be as low as $600 per MW for a six-hour curtailment to as high as $5000 per MW
`for the same period. Such an order-of-magn