Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
Railbelt Intertie Feasibility Study Interim Report 1989
DECISHON FOCUS El Camino Rea Los Altos, California 94022 415 960 3450 RAILBELT INTERTIE FEASIBILITY STUDY Interim Report Prepared for Alaska Power Authority P.O. Box 190869 Anchorage, Alaska 99519-0869 Project Manager: Richard Emerman Prepared by Salim J. Jabbour, Principal Investigator Richard B. Fancher Michael S. Gordon Jennie S. Rice Decision Focus Incorporated 4984 El] Camino Real Los Altos, California 94022 January 30, 1989 Decision ‘Focus Incorporated Section TABLE OF CONTENTS INTRODUCTION 1.1 Objective 1.2. Background 1.3 Feasibility Study Overview 1.4 1.5 Summary of Preliminary Findings Report Organization OVERVIEW OF STUDY ELEMENTS 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 Preliminary Design and Cost Estimate of New Line Between Anchorage and the Kenai Peninsula Preliminary Design and Cost Estimate for Upgrade of the Existing Anchorage—Fairbanks Intertie Preliminary Design and Cost Estimate for New Line From Palmer Through Glennallen to Delta Junction “Northeast Intertie" Estimated Costs and Environmental Impacts of Coal-Fired Power Plants in the Railbelt Estimated Costs and Environmental Impacts of a Natural Gas Pipeline Linking Fairbanks With the Cook Inlet Area Fuel Price Forecasts—Oil and Cook Inlet Natural Gas Fuel Price Forecasts—Railbelt Coal North Slope Natural Gas Sensitivity Assumptions Railbelt Population and Employment Forecasts Railbelt Electricity Demand Forecasts Additional Loads Served by the Northeast Intertie Costs and Impacts of Electric End Use Conservation Programs li Page 2-2 2-3 2-3 2-5 2-6 2-12 2-13 2-15 2-17 2-21 2-22 Decision Focus Incorporated TABLE OF CONTENTS (Continued) Section Page 2.13 Transmission System Stability With Bradley Lake at Full Capacity 2-25 3 SCREENING ANALYSIS OF POWER SUPPLY ALTERNATIVES AND END-USE PROGRAMS 3-1 3.1 Overview 3-1 3.2 Introduction to the Technology Screening Method 3-1 3.3 Technologies Evaluated 3-3 3.4 Evaluating Costs for Each Technology 3-5 3.5 Fixed Costs 3-5 3.6 Variable Costs 3-8 3.7 Economic Assumptions 3-10 3.8 Base Case Results 3-10 3.9 Sensitivity Analysis 3-12 3.10 Comparative Evaluation of Supply-Side Technologies and End-Use Conservation Programs 3-18 3.11 Conclusions 3-20 3.12 References 3-20 4 IMPACT OF PROPOSED INTERTIES ON SYSTEM RELIABILITY 4-1 4.1 Overview 4-1 4.2 Analysis of Historical Customer Outages 4-2 4.3 Potential Changes in Customer Outages 4-9 4.4 Costs of Customer Outages 4-13 4.5 Value of Improved System Reliability 4-18 4.6 Useful Readings 4-22 4.7 References 4-22 iii Decision Focus Incorporated TABLE OF CONTENTS (Continued) Section Page 5 ANALYSIS OF COGENERATION POTENTIAL 5-1 5.1 Overview 5-1 5.2 Characterization of Cogeneration Technical 5-2 Potential 5.3 Cost and Performance of Potential Cogeneration Technologies 5-6 5.4 Cogeneration Market Potential 5-7 5.5 Other Potential Railbelt Cogeneration 5-11 5.6 Results and Conclusions 5-11 5.7 References 5-14 6 INPUT DATA AND MODELING ASSUMPTIONS 6-1 6.1 Objective/Overview 6-1 6.2 System Loads 6-1 6.3 Fuel Price Forecasts 6-5 6.4 Power Generation Plants 6-6 6.5 Transmission System 6-7 6.6 References 6-9 Appendix A SUMMARY OF UTILITIES DATA ON CUSTOMER OUTAGES A-1 Appendix B CHARACTERIZATION OF COGENERATION TECHNICAL POTENTIAL B-1 Appendix C SUMMARY OF ISER END USE SURVEY C-1 Appendix D CHARACTERIZATION OF POTENTIAL COGENERATION TECHNOLOGIES D-1 Appendix E SUMMARY OF OTHER POTENTIAL COGENERATION PROJECTS E-1 Appendix F POWER PLANTS DATA F-1 iv Decision Focus Incorporated 3-1 3-2 3-3 3-4 3-5 3-6 3-7 3-9 4-1 4-2 4-4 4-5 4-7 5-1 6-1 6-3 LIST OF FIGURES A Screening Diagram Technology Screening (Base Case—All Technologies) Technology Screening (Base Case—Selected Technologies) Technology Screening (Middle Gas Price) Technology Screening (High Gas Price) Technology Screening (Low Coal Plant Capital Costs) Technology Screening (Low Coal Plant Capital Costs and High Fraction of Waste Coal) Supply-Side Technologies and End-Use Conservation Programs (Base Case) Supply-Side Technologies and End-Use Conservation Programs (Middle Gas Prices) Outage Hours Per Customer of Railbelt Utilities Unserved Energy Per Customer of Railbelt Utilities Distribution of Railbelt Unserved Energy by Utility Outage Hours of Railbelt Areas (1986/87 Average) Unserved Energy of Railbelt Areas (1986/87 Average) Distribution of Unserved Energy by Area (1986/87 Average) Railbelt Areas Unserved Energy by Outage Duration (1986/87 Average) Railbelt Utilities Unserved Energy by Outage Duration Probability Tree Load Duration Curves for the Kenai Peninsula Load Duration Curves for Anchorage Load Duration Curves for Fairbanks Page 3-2 3-11 3-11 3-13 8-15 3-16 3-18 3-19 3-20 4-5 4-5 4-7 4-8 4-8 4-10 4-10 5-11 6-4 6-5 Table 2-1 2-2 2-3 2-4 2-6 2-7 2-8 2-9 2-10 2-11 2-12 2-13 2-14 2-15 2-16 2-17 2-18 3-1 3-2 3-3 3-4 3-6 3-7 3-9 3-10 4-1 4-2 4-3 4-4 4-5 Decision Focus Incorporated LIST OF TABLES Capital Cost Estimates Annual Operations and Maintenance Cost Estimates Crude Oil Price Scenarios Cook Inlet Natural Gas Prices Marginal Cost of Producing New Reserves Cook Inlet Natural Gas Prices Price of No. 4 Fuel Oil to Golden Valley Price of No. 2 Fuel Oil in Fairbanks Sensitivity Test—North Slope Gas Railbelt Population Forecasts "Middle" Case—Railbelt Population Forecast 1987 Uses of Utility-Supplied Electricity End-Use Breakdown of Residential and Commercial Sales Railbelt Electric Demand Forecast Mid Case Electric Demand Forecast Railbelt Industrial Demand Forecast Potential Purchases of Civilian Electricity by the Military in the Fairbanks Area Northeast Intertie Load Forecasts Total Plant Investment Costs Fixed O&M Costs 1995 Fuel Costs Variable O&M Costs Base Case Sensitivity Case #1: Middle Gas Prices Sensitivity Case #2: High Gas Prices Sensitivity Case #3: Low Coal Plant Capital Costs Sensitivity Case #4: Low Coal Plant Capital Costs Combined With High Fraction of Waste Coal End-Use Programs Form of Utility Data on Customer Outages Demand by Utility and Residential-Industrial/Commercial Split Summary of Outage Hours and Unserved Energy for the Railbelt Summary of Unserved Energy Reduction by Intertie Distribution of Customer Classes in the Railbelt vi Page 2-4 2-4 2-6 2-8 2-9 2-10 2-11 2-11 2-14 2-15 2-16 2-18 2-18 2-19 2-19 2-20 2-21 2-22 3-6 3-7 3-8 3-9 3-10 3-13 3-14 3-16 3-17 3-19 4-3 4-4 4-7 4-14 4-17 Table 4-6 4-7 4-8 4-9 4-10 4-11 5-2 5-3 5-4 5-5 5-6 5-7 5-8 5-10 5-11 5-12 6-1 6-2 6-3 6-4 6-6 6-7 Decision Focus Incorporated LIST OF TABLES (Continued) Summary of Customer Outage Costs Unserved Energy Saved by the Interties Value of Unserved Energy Saved by the Interties Present Value of Reliability Benefit Amount and Value of Unserved Energy Saved by New Kenai-Anchorage Line: Sensitivity Case #1 Amount and Value of Unserved Energy Saved by New Kenai-Anchorage Intertie: Sensitivity Case #2 Commercial Sector Building Types Commercial Sector Cogeneration Technical Potential Industrial Sector Electricity Demand Electricity Demand by Type of Industry Cost and Performance of Selected Cogeneration Units Railbelt Commercial and Industrial Electricity Prices Gas Prices Oil Prices Customer Nominal Hurdle Rates Market Potential Results Detailed Market Potential Results by Region (MW) Market Potential Sensitivity Analysis Results Forecase of Load Growth Rates Service Areas of Railbelt Utilities Energy and Demand Breakdown for CEA Energy Breakdown by Area and by Utility Forecast of Fuel Prices Energy of Railbelt Hydro Power Plants Intertie Efficiencies Page 4-17 4-19 4-19 4-19 4-21 4-21 5-4 5-57 5-5 5-7 5-9 5-9 5-10 5-12 5-12 5-13 6-2 6-2 6-3 6-3 6-8 6-8 Decision Focus Incorporated Section 1 INTRODUCTION 1.1 OBJECTIVE The purpose of this interim report is to provide a brief overview of the Railbelt intertie studies undertaken by the Alaska Power Authority (APA), including a status report on the various elements of the study and preliminary findings that can be presented in advance of the study’s completion. The primary purpose of these studies is to assess the economic feasibility of various intertie proposals that have been suggested for the Railbelt, as well as the feasibility of coal-fired power plants, electric end-use conservation programs, and a natural gas pipeline between Anchorage and Fairbanks. The final report is scheduled for distribution in draft form at the end of March 1989. Conclusions regarding project feasibility will be presented at that time. 1.22 BACKGROUND Statutory direction to undertake these studies was provided in the capital budget passed during the special legislative session in July 1987: The sum of $2,500,000 is appropriated from the Railbelt energy fund in the general fund to the Alaska Power Authority for preparing studies required under AS 44.83.177-44.83.185 for electric interties between the Kenai Peninsula and Fairbanks. This language directs the Alaska Power Authority to perform a feasibility study of Railbelt intertie alternatives. Further action in the 1988 legislative session resulted in a reduction of the appropriation to $2,250,000 and added legislative intent that the feasibility of a proposed natural gas pipeline between Cook Inlet and Fairbanks also be assessed as part of the overall study. The electric intertie projects that were initially identified for review were 1. A new transmission line between Anchorage and the Kenai Peninsula. RI776A 1-1 INTERIM REPORT Decision Focus Incorporated 2: Upgrade of the existing intertie between Anchorage and Fairbanks to substantially higher transfer capability. A third intertie project was later added for consideration as an alternative to the proposed upgrade of the Anchorage-Fairbanks line. 3. A new transmission line from Palmer through Glennallen to Delta Junction, where it would connect with the Golden Valley system in the Fairbanks area. (This project has been referred to as the "Northeast Intertie.") Further, although the statutory direction makes it clear that the intertie projects are intended as the main focus of the study, it was decided that the feasibility of several other Railbelt energy proposals would also be assessed within the study’s overall framework, specifically a A natural gas pipeline from Cook Inlet to Fairbanks. 2: Coal-fired power plants in the Railbelt. 3. Electric end-use conservation programs (i.e., programs designed to induce higher levels of efficiency among electric energy consumers). 1.3 FEASIBILITY STUDY OVERVIEW The feasibility assessment of these selected projects and programs will be focused on a comparison of their expected economic costs and benefits. A comparison of their expected environmental consequences will accompany the feasibility assessment and will appear in the final report. APA assigned the primary task of performing these economic assessments to Decision Focus Incorporated (DFI). The costs of each proposal, as well as other important inputs to the economic analysis such as fuel price and electric demand forecasts, were established in advance of the overall economic assessment through a series of studies undertaken by APA in conjunction with other contractors. DFI has established most of the remaining inputs and assumptions needed for the analysis, and is currently estimating the economic benefits that would be expected from implementation of each proposal. There are several categories of possible benefit that should be quantified for evaluating the intertie proposals. These primary benefit categories include: RIT76A 1-2 INTERIM REPORT Decision Focus Incorporated 1 Reliability. Intertie projects can affect system reliability and a value can be attached to estimated improvements. Reliability can be measured by the number, duration, and magnitude of customer outages. Reliability benefits are explored in this interim report in Section 4. 2: System stability. An intertie project may enhance the stability of an electrical system following certain transmission disturbances and, as a result, may either allow greater operating flexibility or the avoidance of other costs that would be necessary to provide comparable stability conditions.’ 3. Economy energy transfer. Savings are realized when an intertie project allows more displacement of higher cost energy in one area with lower cost energy imported from another area. 4. Capacity reserve sharing. An intertie project could allow two or more areas to share capacity reserves and, as a result, an increment of future investment in plant capacity could be deferred or avoided. 5. Operating reserve sharing. Operating (or “spinning") reserves are typically maintained to help avoid customer outages. An intertie project could allow two or more areas to share operating reserves and therefore reduce operating costs. 6. Transmission efficiency. Improved interties can produce savings to the extent that transmission losses are reduced. Benefit categories 3 through 6 are currently under evaluation, and the results will be presented in the final report. The economic feasibility of a coal-fired power plant can be assessed by comparing total system costs over the long term for scenarios that include the coal plant with scenarios that do not. To gain insight into coal plant feasibility, we compared its total costs of generation over the long term with the total costs of competing technologies under a range of assumptions. This analysis, which we refer to as a "screening analysis," has been performed and the results are presented in Section 3. In the case of the proposed new intertie between Anchorage and the Kenai Peninsula, the new line would allow the avoidance of certain costs that might otherwise be incurred for stability considerations. The status of this element of the analysis is presented at the end of Section 2. RIT76A 1-3 INTERIM REPORT Decision Focus Incorporated The economic feasibility of end use programs can be similarly assessed. The results of the screening analysis with respect to these programs is also presented in Section 3, as well as our conclusions with respect to two other power supply technologies: generators fueled by wood or by refuse-derived fuel. These two latter technologies were included in the screening analysis to help ascertain whether other power supply options not included in our original scope would be likely to affect the results if they were given consideration. The economic feasibility of the proposed gas pipeline linking Fairbanks with the Cook Inlet area will be assessed by estimating its impacts both within the electric power sector and also within the residential and commercial heating markets. The upcoming system modeling effort will be the vehicle for exploring impacts within the power sector. The impact in heating markets will be assessed by the Institute of Social and Economic Research (ISER), and the results from these two perspectives will be combined and presented in the final report. Finally, Section 5 of this interim report includes an analysis on the technical and market potential for cogeneration in the Railbelt. The main purpose of this analysis is to estimate the extent to which the forecast of electricity demand should be reduced as a result of commercial and industrial cogeneration in the future. Note that the demand forecast is for utility-supplied electricity. If cogeneration is expected to make substantial inroads into the market for utility-supplied power, then demand forecasts should be adjusted accordingly. 1.4 SUMMARY OF PRELIMINARY FINDINGS This summary is limited to those preliminary findings developed by DFI and presented in Sections 3, 4, and 5 of this interim report. There are many other findings presented in Section 2 that contribute to the overall feasibility assessment. 1. Coal-fired power plants constructed in the mid-1990s are unlikely to be economic investments compared with natural gas-fired generation. Gas-fired generation maintained its economic advantage over coal in all of the cases that we examined. Generators fueled by wood or refuse-derived fuel are less economic than both coal and gas-fired generation. 2. Three end use conservation programs are likely to be economic investments. They include two commercial lighting programs and one program to encourage residential water heating conversions from electric to natural gas. Five other programs appeared RIT76A 1-4 INTERIM REPORT Decision Focus Incorporated economic in a sensitivity test that assumed higher gas price forecasts. 3. The estimated reliability benefit for the proposed new intertie between Anchorage and the Kenai Peninsula is $2.4 million. This is the estimated value of customer outages avoided as a result of the new line, extended over 35 years and then discounted to its present value. However, a _ sensitivity analysis based on substantially different assumptions produced an estimate of $13.7 million. 4. The estimated reliability benefit for the proposed upgrade of the Anchorage-Fairbanks intertie is $1.4 million, calculated in the same manner. 5. The estimated reliability benefit for the proposed new Northeast intertie is $10.3 million, again calculated in the same manner. 6. The present market potential for commercial and industrial cogeneration in the Railbelt is estimated at 21 MW. However, this result is highly sensitive to the investor “hurdle rate" assumed in the analysis. Using a low hurdle rate, the market potential increases to 76 MW; using a high hurdle rate, the market potential is only 2 MW. 1.5 REPORT ORGANIZATION Section 2 provides brief summaries of the inputs and assumptions developed by contributors other than DFI. Section 3 presents the screening analysis of power supply alternatives and end use programs. Section 4 presents the analysis of the impact of the proposed interties on system reliability. Section 5 presents the analysis of cogeneration potential. Section 6 presents a compilation of data and certain key assumptions that will be used in the upcoming system modeling task. RIT76A, 1-5 INTERIM REPORT Decision Focus Incorporated Section 2 OVERVIEW OF STUDY ELEMENTS: (Inputs and Assumptions Developed by APA and Contractors Other than Decision Focus Incorporated) 2.1 PRELIMINARY DESIGN AND COST ESTIMATE OF NEW LINE BETWEEN ANCHORAGE AND THE KENAI PENINSULA (Contractor: Power Engineers, Inc.) Status: Complete Two routes have been identified: i "Enstar" route, which follows an existing natural gas pipeline through the Kenai National Wildlife Refuge followed by a submarine crossing of Turnagain Arm into Anchorage. The capital cost is estimated at $79.0 million (in 1987 dollars). Annual operations and maintenance cost is estimated at 1.5 percent of capital cost, or $1.2 million per year. 2. "Tesoro" route, which follows an existing oil products pipeline along the west coast of the Kenai Peninsula followed by a submarine crossing of Turnagain Arm into Anchorage. The capital cost is estimated at $99.4 million (in 1987 dollars). Annual operations and maintenance cost is again estimated at 1.5 percent of capital cost, or $1.5 million per year. Either line would be constructed at 230 KV and have a transfer capacity of 250 MW. Because the Enstar route crosses land within the Wildlife Refuge that had been proposed (though not yet designated) as "wilderness," it was anticipated that both Congressional and Presidential approval would be required to obtain the necessary right of way. Though cost considerations clearly favor the Enstar route, the Tesoro route was developed in case the proposed wilderness designation forced abandonment of the less expensive alternative. However, the Department of Interior has now acted ‘This section was prepared by the Alaska Power Authority. RI776A 2-1 INTERIM REPORT Decision Focus Incorporated favorably on a request by the State to exclude from wilderness designation a corridor adjacent to the Enstar pipeline for possible future construction of the proposed intertie. If Congress agrees to exclude the intertie corridor from wilderness designation, the two proposed routes would then be roughly equivalent in terms of permitting difficulty. Preliminary schedules for permitting and construction suggest that completion of the intertie should not be expected prior to 1994, regardless of the route, assuming the project were approved by the 1989 Legislature. 2.2 PRELIMINARY DESIGN AND COST ESTIMATE FOR UPGRADE OF THE EXISTING ANCHORAGE—FAIRBANKS INTERTIE (Contractor: Harza Engineering Co.) Status: Complete Presently, the transmission link between the Wasilla area and Fairbanks consists of three segments: 1. Wasilla to Willow—138 KV line owned by Matanuska Electric Association. 2. Willow to Healy—345 KV line owned by Alaska Power Authority. The line is presently operated at 138 KV, consistent with voltages at either end. 3. Healy to Fairbanks—138 KV line owned by Golden Valley Electric Association. The upgrade proposal consists primarily of new 345 KV line construction between Willow and the Chugach Electric transmission system south of Wasilla, and between Healy and Fairbanks. (Existing segments would be supplemented, not replaced, by the new line construction.) This revised link between Anchorage and Fairbanks would initially be operated at 230 KV, raising the transfer capability from the present level of 70 MW to a revised level of 225 MW. The capital cost of this upgrade is estimated at $118.2 million in 1987 dollars. The additional operations and maintenance cost of the intertie following this upgrade is estimated at $900,000 per year, again in 1987 dollars. The main issue with respect to land use involves the new segment from Healy to Fairbanks. The proposed route crosses federal land south of the Tanana River near Fairbanks. Agreement would have to be worked out with the military at Fort Wainwright. RIT76A 2-2 INTERIM REPORT Decision Focus Incorporated Again, preliminary schedules for permitting and construction suggest that completion of the upgrade should not be expected prior to 1994 assuming the project were approved by the 1989 Legislature. 2.3 PRELIMINARY DESIGN AND COST ESTIMATE FOR NEW LINE FROM PALMER THROUGH GLENNALLEN TO DELTA JUNCTION “NORTHEAST INTERTIE" (Contractor: Power Engineers, Inc.) Status: Draft under development; Final report due March 1989. The proposed line would be constructed at 230 KV but operated initially at 138 KV with a transfer capacity of 150 MW. In combination with the existing Anchorage- Fairbanks intertie, the combined transfer capability would therefore be 220 MW, minus whatever intermediate load along the Northeast intertie route might be served. For illustration, if the intermediate load served by the intertie in the Glennallen-Valdez area were 10 MW, the combined transfer capability between Anchorage and Fairbanks would be 210 MW. The capital cost of the Northeast intertie is estimated at $155 million in 1988 dollars. Annual operations and maintenance cost is estimated at 1.5 percent of capital cost, or $2.3 million per year. Preliminary schedules for permitting and construction suggest that completion of the intertie should not be expected prior to 1994, assuming the project were approved by the 1989 Legislature. 2.4 ESTIMATED COSTS AND ENVIRONMENTAL IMPACTS OF COAL- FIRED POWER PLANTS IN THE RAILBELT (Contractor: Stone & Webster Engineering Corp.) Status: Complete Capital cost as well as operations and maintenance cost estimates were developed for coal-fired power plants in three different sizes (50 MW, 100 MW, and 150 MW ) and four different Railbelt locations (Healy, Nenana, Beluga, and Matanuska Valley). Table 2-1 shows a summary of the capital cost estimates and Table 2-2 shows a summary of the operations and maintenance costs. The combustion technology selected for development of these estimates is atmospheric fluidized bed, based primarily on its expected cost advantage over conventional pulverized coal plants. RI776A 2-3 INTERIM REPORT Decision Focus Incorporated Table 2-1 CAPITAL COST ESTIMATES (1988 dollars) Healy Nenana Beluga Matanuska 50 MW $/kW 3,322 3,378 3,476 3,119 Total ($M) 166.1 168.9 173738 15559 100 MW $/kW 2,499 2,022 2,610 2,340 Total ($M) 249.9 25a. 2 261.0 234.20 150 MW $/kW 2,143 2,158 2n235 1,952 Total ($M) SVAL SG) S23 iu, Soo 292.9 Table 2-2 ANNUAL OPERATIONS AND MAINTENANCE COST ESTIMATES* (millions of 1988 dollars) Healy Nenana Beluga Matanuska 50 MW Ta Nae Ine 7.4 100 MW 10.2 10.2 10.2 L065 150 MW 13.0 13:30 13.0 Si * Excludes first year costs for training and commissioning Organizations proposing to build coal-fired power plants at Healy and at Nenana have thus far maintained that such plants with capacities of approximately 100 MW could be built at an installed cost of about $1,600 per kW, in contrast to the Stone & Webster estimate of about $2,500 per kW. In other words, the Stone & Webster estimate is on the order of 50 percent higher than the estimates suggested by these prospective sponsors. Because comparable detail has not been made available for the lower estimates, the causes of this substantial difference are not precisely known. However, it appears that the major issue is the estimate of cost differential between Alaska and the lower 48, especially in the area of labor cost. The estimates from Stone & Webster are based on their experience with Bradley Lake construction bids and other recent Alaska projects. RIT776A 2-4 INTERIM REPORT Decision Focus Incorporated A further possible source of discrepancy could be in the area of construction technique. Prospective sponsors of a plant at Healy have suggested "modular" plant construction, whereas the Stone & Webster estimates are based on "conventional" construction. Conventional construction in this case means delivery to the jobsite of the largest components possible by rail and preassembly of these components into larger components under controlled conditions at the site. An "all out" modular construction approach could be more cost effective. However, detailed information regarding this approach was not available, nor has a circulating fluidized bed unit been constructed to date utilizing extensive modular construction techniques. Although these are not "minimum" estimates reflecting the best that a project sponsor might be able to accomplish, they are considered by APA to be conservative estimates appropriate for planning. The feasibility assessment will include sensitivity analysis to determine the impact of assuming the lower capital cost estimate. Stone & Webster concludes that coal-fired power plants at any of the four sites, and at any of the three sizes, could meet all environmental standards including air quality standards, and should be able to obtain all necessary permits. 2.5 ESTIMATED COSTS AND ENVIRONMENTAL IMPACTS OF A NATURAL GAS PIPELINE LINKING FAIRBANKS WITH THE COOK INLET AREA (Contractor: Stone & Webster Engineering Corp.) Status: Draft complete; Final report due February 1989. The capital cost of a 16-inch diameter natural gas pipeline linking Fairbanks with the Cook Inlet area is estimated at $190.0 million in 1988 dollars. A 16-inch system could accommodate preliminary projections of residential and commercial consumption in the Fairbanks area over the next 30 years and, if required, its capacity could be expanded with compression to accommodate military consumption as well. (For purposes of comparison, the Stone & Webster capital cost estimate for a 20-inch pipeline—the size initially proposed by Enstar Natural Gas Co.—is $235 million in 1988 dollars.) The probability that North Slope natural gas will be available in Fairbanks for transmission to Anchorage at sustained price levels that undercut Cook Inlet gas during the next 30 years was judged by the Power Authority to be too low to form a basis for pipeline planning at this time. Though possible future levels of Anchorage demand for natural gas were, as a result, not considered in sizing the pipeline proposal, the selected 16-inch system would be capable of carrying nearly enough gas to satisfy current levels of residential and commercial demand in Anchorage. RI776A 2-5 INTERIM REPORT Decision Focus Incorporated The capital cost of the distribution system in Fairbanks is estimated at $33.8 million in 1988 dollars. The annual operations and maintenance expense for the system additions is estimated at $4.0 million ($2.4 million for the distribution system, $1.6 million for the main transmission pipeline). The major environmental issue with respect to pipeline construction would be the potential cumulative effect on fisheries resources of the numerous instream crossings proposed. However, proper construction techniques can reduce these impacts below significant levels. With respect to air quality impacts, it is expected that widespread conversion to natural gas would reduce pollutants, especially sulfur dioxide and particulates, though increased production of water from natural gas combustion compared with coal or oil may produce increased ice fog during cold weather conditions. 2.66 FUEL PRICE FORECASTS—OIL AND COOK INLET NATURAL GAS (Contractor: ICF, Incorporated) Status: Complete 2.6.1 Crude Oil Price Forecasts In early 1988, ICF collected a sample of 17 long-term crude oil price forecasts, most of which were produced during 1987, from a broad selection of firms and agencies in the United States and Europe. Based on the range of opinion expressed in these forecasts, three price scenarios were developed to represent the main "schools of thought.” Table 2-3 displays the three price scenarios. Table 2-3 CRUDE OIL PRICE SCENARIOS Saudi Light Delivered to U.S. Gulf* (1987 dollars per barrel) Year Low Mid High 1990 $14 $18 $20 2000 18 24 30 2010 20 30 40 *North Slope oil delivered to the U.S. Gulf is estimated at about $1 per barrel less than Saudi Light, while West Texas Intermediate is estimated at about $1 per barrel more. An analysis of the main schools of thought, describing the reasoning and evidence put forward in support of each, was provided to the APA Board of Directors. RI776A 2-6 INTERIM REPORT Decision Focus Incorporated The Board then assigned the following probabilities to each price scenario for use in this Railbelt study and for APA planning generally: low = 60%, mid = 30%, and high = 10%. The set of oil price forecasts published by the Alaska Department of Revenue (ADOR) for Fall 1988 was substantially lower than the set established by ICF for the Power Authority. For projects or programs judged feasible according to the ICF "low" price forecasts; additional sensitivity testing will. be performed consistent with the ADOR outlook. 2.6.2 Cook Inlet Natural Gas Price Forecasts Recent contracts provide the best available indication of the current value and long-term price outlook for Cook Inlet natural gas. Two such contracts have recently been negotiated and submitted to the Alaska Public Utilities Commission for review: 1. Contract between Marathon Oil Co. and Enstar Natural Gas Co. covering an initial commitment of 456 Bcf, with options for additional commitments in the future; 2. Contract between Marathon Oil Co. and Chugach Electric Association covering an initial commitment of 215 Bcf, with options for additional commitments in the future. Each contract specifies a base price plus a periodic price adjustment factor. For the Enstar contract, the adjustment factor is based on changes in the price of crude oil. For the Chugach contract, the adjustment factor is based on price changes for crude oil, (refined) fuel oil, and natural gas in the lower 48. Because fuel oil and lower 48 natural gas prices are expected to follow a path roughly similar to crude oil prices over the long term, APA has assumed a simplified adjustment factor for present purposes consisting of crude oil prices only. Table 2-4 shows the price projections that result when the crude oil price scenarios are applied according to the contract terms. Although the prices in Table 2-4 are estimated until the year 2010, the gas delivery commitments made under recent and existing contracts do not actually provide enough gas to supply expected requirements for that long. For the Chugach price forecast, the assumption is that new contracts with Beluga producers (Chevron, ARCO, and Shell) will be negotiated at prices comparable to those in the new contract between Chugach and Marathon, and that total gas commitments to Chugach will then suffice through the year 2010. Although more than the initial 215 Bcf can be provided to Chugach by Marathon under their new contract, Marathon is not required to do so. RIT76A 2-7 INTERIM REPORT Decision Focus Incorporated Table 2-4 COOK INLET NATURAL GAS PRICES* (1987 dollars per MBtu)** Chugach Chugach Chugach Enstar Enstar Enstar Low Mid High Low Mid High 1990 hun? 1.43 1.50 1.43 1S 1.69 1995 1.31 1.69 1.98 1.54 1.98 aaad 2000 1.47 1.93 2.38 1.66 mei? 2.67 2005 1.57 ie aul munt 2.49 2.06 2.63 2010 365) 2.41 Bukt 1.56 2.28 3.01 *Note that the Chugach prices do not account for blending in remaining quantities of lower priced, "old" Beluga gas. The Enstar prices, however, represent the blended acquisition cost to Enstar from its two main contracts (primarily the new Marathon contract), exclusive of any distribution margin to customers such as Anchorage Municipal Light & Power. **Note also that 1 MBtu (i.e., million Btu) is approximately equal to 1 Mcf. Similarly, new and existing contracts for Enstar do not provide gas commitments for expected demand beyond the 2002 to 2004 time frame. Again, more gas can be provided by Marathon to Enstar under terms of their new contract, but Marathon is not presently required to do so. The forecast of Cook Inlet gas prices becomes more uncertain as we approach, and go beyond, the year 2010. Yet prices in that distant time frame could still be significant in the analysis. For example: us Coal Plant Feasibility. The Stone & Webster report suggests it is unlikely that a coal-fired power plant could be brought on line before 1995. A feasibility assessment of such a plant with an assumed economic life of 35 years should therefore extend until the year 2030. With massive coal reserves, and a forecast that domestic coal prices will be based on cost of production and may well stay even with inflation (i.e., constant in real terms) through 2030, the forecast of gas prices between 2010 and 2030 could make a difference. 2. Gas Pipeline Feasibility. The time frames noted above for the coal plant roughly apply to the pipeline proposal as well. In one sense, RITI6A 2-8 INTERIM REPORT Decision Focus Incorporated escalating Cook Inlet gas costs after 2010 could narrow the advantage of gas over oil and reduce the feasibility of the pipeline. On the other hand, escalating Cook Inlet gas costs in that time frame could actually enhance its feasibility if North Slope gas were available at that time in Fairbanks at a lower price and could, via the proposed pipeline, be sent to Anchorage. There is a suggestion of this possibility in the work that ICF performed for the Power Authority before the two recent contracts were released for public review. Based on the "50th percentile" estimates of undiscovered Cook Inlet gas resources issued by the Potential Gas Committee, ICF produced a forecast of the marginal cost of adding new reserves sufficient to meet annual production requirements under (1) a "base case" demand scenario that provided only very gradual growth above the existing demand of approximately 200 Bcf per year, and (2) a higher demand scenario that included an additional 50 Bcf per year for new LNG export starting in 1995 (ie., an additional 750 Bef by the year 2010, equal to about 4 years of "normal" demand). Table 2-5 shows the results. Table 2-5 MARGINAL COST OF PRODUCING NEW RESERVES (1987 dollars per MBtu) "Base" "Higher" Demand Demand 1990 mew ® o--* 1995 0.47 0.73 2000 0.86 1.65 2005 LiciOS 3.54 2010 Sjoerd 5.33) *No reserve additions required until 1991. Again, our gas price forecast in the "low" case (to which we give a 60 percent probability per APA Board direction) is in the range of $1.56 to 1.65 in the year 2010. However, the supply analysis noted above suggests that the marginal cost of reserve additions will be substantially higher by then. Further, the ICF cost estimates under the higher demand case suggest that marginal costs could escalate sharply from the $3.00 range to the $5.00 range perhaps with the 2010 to 2015 period, even without additional LNG export. RI776A 2-9 INTERIM REPORT Decision Focus Incorporated Although our long-term planning forecasts extend only to 2010, the feasibility assessments will require extension periods through 2030. While conservatism with respect to such distant forecasts suggests that all prices and assumptions be held constant between 2011 and 2030 for the base case, sensitivity analysis will be performed with substantially higher Cook Inlet gas prices during that extension period. Table 2-6 displays a range of “extension period" prices that are under consideration for such sensitivity analysis. Table 2-6 COOK INLET NATURAL GAS PRICES (1987 dollars per MBtu) Chugach Chugach Chugach Enstar Enstar Enstar Low Mid High Low Mid High 1990 ee 1.43 1.50 1.43 1.60 1.69 1995 Few 1.69 1.98 1.54 1.98 Aol 2000 1.47 1.93 2.38 1.65 2i7 2.67 2005 TOW z.iL7 2.77 1.49 2.06 2.63 2010 1.83 2.41 3.17 1.56 2.28 3.01 2015 2.50 3.50 4.50 2.50 3 150 4.50 2020 4.00 5.00 6.00 4.00 5.00 6.00 2025 5.50 6.50 71,58 5.50 6.50 7.50 2030 7.00 8.00 9.00 7.00 8.00 9.00 2.6.3 Fuel Oil Price Forecasts There are two sets of fuel oil price forecasts that are of primary interest in this study: is No. 4 distillate fuel oil purchased by Golden Valley Electric Association in Fairbanks for use in power generation; 2. No. 2 distillate fuel oil purchased by residential and commercial customers in Fairbanks for use primarily in space heating. Golden Valley has a contract with the State that extends through 1995 for the purchase of royalty oil from the North Slope. The royalty oil purchased by Golden Valley is assigned to the Mapco refinery in Fairbanks for processing, and the refined product (i.e., No. 4 fuel oil) is sold back to Golden Valley at a reduced margin. It is assumed in this forecast that future prices to Golden Valley will conform generally to RI776A 2-10 INTERIM REPORT Decision Focus Incorporated this price setting mechanism, the main elements of which are the wellhead price of crude oil on the North Slope and the TAPS tariff. The resulting price forecast is shown in Table 2-7. Table 2-7 PRICE OF NO. 4 FUEL OIL TO GOLDEN VALLEY (1987 dollars per MBtu)* Low Mid High 1990 2.50 3.19 3.54 1995 2.84 3.68 4.34 2000 3.19 4.16 5.13 2005 S237 4.65 5.94 2010 3.54 5.13 6.75 *Note that 1 gallon equals approximately 0.144 MBtu The expected price of No. 2 fuel oil to residential and commercial customers is relevant to this analysis primarily in the assessment of the proposed natural gas pipeline from Cook Inlet to Fairbanks. Price forecasts were developed based on the expected costs of crude oil acquisition to the marginal supplier, plus expected refining, transportation, and distribution costs. Table 2-8 shows the results. Table 2-8 PRICE OF NO. 2 FUEL OIL IN FAIRBANKS (1987 dollars per MBtu)* Residential Large Commercial Low Mid High Low Mid High 1990 5.43 6.08 6.44 4.13 4.85 5.14 1995 St2 6.59 7.24 4.49 5.36 5.94 2000 6.08 7.10 8.11 4.85 5.86 6.88 2005 6.26 7.61 8.98 5.00 6.37 ToT 2010 6.44 8.11 9.85 5.14 6.88 8.54 * Note that 1 gallon equals approximately 0.138 MBtu RIT76A 2-11 INTERIM REPORT Decision Focus Incorporated 2.7. FUEL PRICE FORECASTS—RAILBELT COAL (Contractor: Institute of Social and Economic Research) Status: Complete For this analysis, it is assumed that minemouth coal prices will be based on the cost of production and that the cost of production will not increase in real terms over the long run. Constant real price assumptions are as follows: Healy Minemouth $1.30 / MBtu Healy Delivered to Fairbanks 2.52 / MBtu Healy Delivered to Nenana 1.69 / MBtu Beluga Minemouth 1.15 / MBtu Matanuska Minemouth 1.15 / MBtu Healy Waste Coal at Minemouth 0.07 / MBtu Beluga Waste Coal at Minemouth 0.07 / MBtu Differential cost of Healy coal delivered to Fairbanks and Nenana is based on estimates of rail transport costs. The Beluga minemouth price depends on the development of an 8 million ton per year mine at Beluga for export. It is assumed that a mine at Beluga would not be developed solely to supply coal to a power plant in Alaska. The current proprietors of the Matanuska field intend to develop the remaining reserves for export, not for in-state power development. However, in the event that the export plan is not realized, the Matanuska resource could still be available in the 1990s and beyond for alternative purposes. The cost estimate used here is based on production scaled to a power plant of approximately 100 MW for the 28 years that currently estimated reserves would last at that rate of depletion. Waste coal is a mixture of coal and earth that results from the inability of the dragline to cut the edge of the coal seam cleanly. The ratio of waste coal to regular coal at the Healy mine is estimated at about 1:10. While this presently results in far less waste coal than would be necessary to fuel a 100 MW power plant, it could make a significant contribution to fuel requirements if blended with regular coal. The cost of this material at the minemouth is estimated to be very low. Because the seam width at Beluga is estimated to be similar to Healy, comparable production ratios and costs for waste coal are estimated for Beluga. Given an 8 million ton per year export mine, however, the aggregate amount of waste coal available from such an operation would be significantly higher. Very limited resources were available to ISER for developing these estimates. Should the analysis suggest that the feasibility of coal-fired power plants over the next RIT76A 2-12 INTERIM REPORT Decision Focus Incorporated 20 years is sensitive to these coal price assumptions, then further examination of these price assumptions would be warranted. 2.8 NORTH SLOPE NATURAL GAS SENSITIVITY ASSUMPTIONS It is possible that North Slope gas will become available for in-state use in the Fairbanks area during the next 20 years; an event that could influence the feasibility of projects under review in this study. As a result, certain assumptions are being considered for sensitivity analysis to shed light on these possible impacts. The first issue is the probability that North Slope gas will become available for use in the Railbelt. For purposes of the population and employment forecasts, the Power Authority adopted certain probabilities that a natural gas pipeline from the North Slope (e.g., TAGS) would be constructed during the study period. Specifically, it was decided that construction of a North Slope gas pipeline would be assumed during the 1990s under the "high" crude oil price scenario, during the decade 2000 to 2010 under the "mid" price scenario, and not at all prior to 2010 under the “low" price scenario. Given the probabilities attached to the price scenarios, this translates into a 10 percent chance of North Slope gas pipeline construction during the 1990s, a 30 percent chance between 2000 and 2010, and a 60 percent chance of no construction prior to 2010. The duration of any excess capacity in the pipeline is also central to the question of North Slope gas availability for the Railbelt. To the extent that excess capacity exists, the incremental cost of transporting gas from the North Slope to Fairbanks for in-state use could be very low. However, the owners of the pipeline will seek to commit the full capacity of the line throughout its length, and thereby minimize the duration of any excess capacity. As a result, a scenario could develop in which the cost of North Slope gas delivered to Fairbanks is low initially, but escalates sharply at such time as pipeline capacity can be fully committed for export sales. Although a number of variations like this can be imagined, basically three price scenarios need to be considered for North Slope gas given the assumption that such gas is made available for use in the Railbelt: i Price of North Slope gas delivered to Fairbanks is relatively high. If the delivered price is significantly higher than Cook Inlet gas prices, and is not sufficiently attractive relative to oil to induce much switching from oil to North Slope gas, then the impact of North Slope gas availability on the various projects and programs under consideration would be insignificant. RIT76A 2-13 INTERIM REPORT Decision Focus Incorporated 2. Price of North Slope gas delivered to Fairbanks is comparable with the price of Cook Inlet gas. In that event, projects such as the proposed upgrade of the Anchorage-Fairbanks intertie and the proposed natural gas pipeline from Cook Inlet to Fairbanks would lose much (or most) of their usefulness at the moment North Slope gas became available in Fairbanks. 3. Price of North Slope gas delivered to Fairbanks is substantially below the price of Cook Inlet gas. At some point, relatively low delivered gas prices from the North Slope could enhance the feasibility of the proposed upgrade of the Anchorage-Fairbanks intertie or the proposed gas line between Cook Inlet and Fairbanks, by allowing the lower priced gas (or gas-fired electricity) to flow south from Fairbanks to the Anchorage area. On the other hand, that scenario could have a negative impact on the feasibility of coal-fired power plants or conservation programs by reducing the long run cost of the gas-fired generation alternative. Clearly, there are many possible combinations of events and conditions. As noted earlier, Cook Inlet gas prices are assumed constant in real terms between 2011 and 2030 for all base case scenarios. We assume that North Slope gas would be priced substantially higher than Cook Inlet gas throughout the analysis period for all of these cases, and therefore would not compete as a Railbelt energy source. However, sensitivity analysis will be performed assuming Cook Inlet gas prices escalate sharply beginning in 2011 as a result of potentially higher production costs. A further sensitivity test will be examined with the additional assumption that North Slope gas becomes available at a price that substantially undercuts the escalating Cook Inlet gas price between 2010 and 2030 as shown in Table 2-9. Table 2-9 SENSITIVITY TEST—NORTH SLOPE GAS (1987 dollars per MBtu) Cook Inlet Gas North Slope Gas (Sensitivity) (Sensitivity) 2010 2.41* 2.00 2015 3.80 200) 2020 5.00 3.00 2025 6.50 3550 2030 8.00 4.00 *$2.41 reflects the "Mid" case price in 2010 for gas supplied to Chugach Electric Association. RIT76A 2-14 INTERIM REPORT Decision Focus Incorporated Again, although this sensitivity test could be instructive in suggesting a higher level of potential benefits particularly for an Anchorage-Fairbanks gas pipeline, the high level of uncertainty in these distant scenarios should produce caution in drawing conclusions. For example, as noted earlier, the price of North Slope gas delivered to Fairbanks could start out low relative to Cook Inlet gas, then reach equivalence for a while, and then exceed Cook Inlet prices as pipeline capacity from the North Slope is used up. On the other hand, a relatively constant real price of gas from the North Slope over time delivered to Fairbanks (e.g., $4.00 per MBtu) could initially be higher than Cook Inlet prices, then fall into equivalence if Cook Inlet prices rise, and finally undercut Cook Inlet prices based primarily on differential production costs at the wellhead over the long term. It will be difficult to base planning decisions today on speculation over what may happen to North Slope gas prices, and their relationship to Cook Inlet gas prices, 20 years or more into the future. 2.9 RAILBELT POPULATION AND EMPLOYMENT FORECASTS (Contractor: Institute of Social and Economic Research) Status: Complete Electricity demand is dependent on the forecasts of population, households, and employment in the study area. A range of forecasts was developed based on alternative assumptions and varying combinations of assumptions. Using probabilities adopted by the Power Authority Board, a “low,” "middle," and "high" forecast was identified, each of which is judged to be equally probable. Neither the "low" nor the "high" forecast is intended to represent a boundary (i.e., a "worst" or best") case. Table 2-10 shows a summary of the three population forecasts for the Railbelt. Table 2-11 shows a breakdown of the middle case into the three main regions selected for purposes of the intertie analysis. Table 2-10 RAILBELT POPULATION FORECASTS (thousands) Low Middle High 1987 388.0 388.0 388.0 1990 385.6* 383.9* 389.5 1995 399.4 405.4 418.3 2000 416.7 436.5 465.9 2005 445.7 479.7 S27 2010 480.3 538.7 SSO cu *The selection of cases was determined by the number of households and level of employment in the year 2010. As a result, there can be overlap among the selected cases in the initial years. RIT76A 2-15 INTERIM REPORT Decision Focus Incorporated Table 2-11 "MIDDLE" CASE—RAILBELT POPULATION FORECAST (thousands) Anchorage Fairbanks and Kenai North Mat-Su Peninsula Star Boroughs Borough Borough* 1987 268.6 39.6 79.8 1990 262.1 39.7 82.2 1995 277.4 41.3 86.8 2000 300.2 44.4 91.9 2005 333)03 47.7 98.6 2010 a70.6 52.3 107.9 *Includes the Southeast Fairbanks census area. Although numerous combinations of assumptions can produce a roughly comparable forecast of population, the main assumptions underlying the selected middle case are briefly summarized below. The price of oil is assumed to rise from $14 per barrel in 1990 (1987 dollars) to $20 in 2010. Production from existing fields continues, and technological advances combined with cost control allow the West Sak field on the North Slope to come into production after 2000. Production falls off from a peak of 723 million barrels in 1989 to 411 million in 2000 and 265 million in 2010. Frontier areas, including ANWR and the OCS, are not developed because sufficiently large discoveries are not made and the cost of development of small fields cannot be recovered due to the low price. In spite of the decline in production, however, total employment in the industry does not fall because of the increasingly labor-intensive nature of the process of extracting the maximum amount of oil out of currently producing fields. It is assumed that a TAGS gas line is not built within the forecast horizon (i.e., prior to 2010). The federal government role as a basic industry remains constant with the exception of the deployment of the Light Infantry Division in Fairbanks. Tourism expansion continues at a rate of 20,000 additional tourist visitors annually. The mining industry grows in the late 1980s and 1990s at a rapid rate with the development of the Red Dog, Greens Creek, and U.S. Borax projects, a new coal facility for export in the Railbelt, and other unspecified activities projected to increase at three percent annually. RI776A 2-16 INTERIM REPORT Decision Focus Incorporated The timber industry expands into the early 1990s, at which time further growth is constrained by the size of the resource base, except in Southcentral Alaska where a modest industry develops in the 1990s. The traditional commercial fishery is constrained by the size of the resource base, but the bottomfish industry expands over time, centered in the Southwestern part of the state, but with additional activity in the Southern Railbelt and Bristol Bay. State government gradually contracts through the 1990s in spite of revenue augmentation measures, including the use of Permanent Fund earnings beginning in 1992, the reimposition of the personal income tax in 1996, and the elimination of the Permanent Fund dividend in 1999. State petroleum revenues decline in real terms to $1241 million in 2000 and $842 million in 2010. The Permanent Fund real rate of return averages only three percent annually. Government expenditures are concentrated on the operating budget, leaving little for capital expenditures. In spite of wage levels held constant in nominal dollars for several years, government employment levels fall over time due to revenue constraints. The Railbelt economy continues to be the support center for the majority of the state. Its economy grows in response to basic sector growth, which occurs largely outside the boundaries of the Railbelt, and also in response to per capita income growth, which is assumed to resume in the 1990s following the current recession. 2.10 RAILBELT ELECTRICITY DEMAND FORECASTS (Contractor: Institute of Social and Economic Research) Status: Draft complete; Final report due March 1989. Electricity demand forecasts have been developed for each category of electrical use and for each of the main regions of the Railbelt. (Examples of electrical use categories are commercial lighting and residential hot water heating.) The main purpose for developing the demand forecasts at the "end-use" level of detail is to provide a foundation for assessing the impacts of conservation programs. For example, to estimate the cost and effect of implementing a program to encourage more efficient commercial lighting, it is necessary to start with a forecast of electrical demand for commercial lighting based on estimates of efficiency levels that would develop over time in the absence of the program. Only then can the expected incremental effect of the program be estimated and judged. In 1987, the seven electric utilities in the Railbelt required approximately 3400 gigawatt hours (GWh, or millions of kilowatthours) to meet their customers’ needs. Table 2-12 shows the estimated breakdown of this energy requirement. Table 2-13 shows the proportional breakdown of residential and commercial sales to each end-use category in 1987. RIT76A 2-17 INTERIM REPORT Decision Focus Incorporated Table 2-12 1987 USES OF UTILITY-SUPPLIED ELECTRICITY (total Railbelt) GWh Percent Residential 1245 Si Commercial 1516 45 Industrial 256 8 Street Lights & Public Authorities 49 1 Distribution Losses & Office Use 225 U Transmission Losses 82 2 TOTAL 3373 100 Table 2-13 END-USE BREAKDOWN OF RESIDENTIAL AND COMMERCIAL SALES 1987 Residential Sales 1987 Commercial Sales End Use Percent End Use Percent Space Heating 19 Space Heating a Water Heating LS Space Cooling 4 Refrigerators 12 Ventilation z5 Freezers 6 Water Heating 2 Cooking 5 Refrig & Freez 12 Clothes Drying 8 Cooking 3 Lighting 13 Lighting 49 Miscellaneous 22 Miscellaneous 12 TOTAL 100 TOTAL 100 Again, a range of forecasts was developed based on alternative assumptions and varying combinations of assumptions, including the following: 1. Population, households, and employment. 2. Energy prices. 3. Consumer discount rates (for modeling consumer purchase choices). RIT76A 2-18 INTERIM REPORT Decision Focus Incorporated 4. Technological change (e.g., possible change in efficiency options and costs). 5. Southern Railbelt natural gas market penetration (i.e., different expansion scenarios for the natural gas distribution system). Based on probabilities established by ISER and Power Authority staff, "low," "mid," and "high" forecasts were selected from the distribution such that each of the three is judged to be equally probable (i.e., the low represents the bottom third of the distribution, the mid represents the middle third, and the high represents the top third). Table 2-14 shows these three forecasts aggregated for the entire Railbelt. Table 2-15 shows a further breakdown of the mid case for each of three main Railbelt regions. Table 2-14 RAILBELT ELECTRIC DEMAND FORECAST* (total energy, GWh) Low Mid High 1987 3305 3305 3305 1990 3237 3225 3269 1995 3153 3271 3432 2000 3156 3395 3675 2005 3289 3641 4058 2010 3495 4053 4427 *Excludes transmission losses. Weather adjusted. Table 2-15 MID CASE ELECTRIC DEMAND FORECAST (three Railbelt regions, GWh) Anchorage Kenai Fairbanks . and Mat-Su Peninsula North Star Boroughs Borough Borough* 1987 2262 455 588 1990 2189 438 598 1995 2219 430 622 2000 2306 442 646 2005 2493 462 685 2010 2805 497 752 *Includes Southeast Fairbanks census area RI776A 2-19 INTERIM REPORT Decision Focus Incorporated Residential electricity sales are forecast to grow more slowly than the stock of occupied housing due to higher efficiencies for new equipment (which in part reflects implementation of new federal appliance efficiency standards), assumed increase in average electricity prices due largely to the expiration of existing contracts for "old" Beluga gas supplied to Chugach Electric Association, and continued erosion of electric market share to natural gas particularly in the category of space heat. Commercial electricity sales are forecast to grow more slowly than commercial floorstock due primarily to higher efficiencies for new equipment, particularly in the lighting sector. Implementation of new federal standards for fluorescent ballasts contributes to this outlook. The relatively low level of electric space heat is expected to decline further, while miscellaneous equipment per square foot is expected to grow. For the industrial forecast in the mid case, the Tesoro refinery on the Kenai Peninsula is projected to reduce its purchases from Homer Electric from 89.3 GWh in 1987 to 59.5 GWh in all subsequent years as a result of increased self-generation at the refinery. In the low case, Tesoro purchases from Homer Electric decline to zero in 1995 and beyond, consistent with 100 percent self-generation. For the entire Railbelt, Table 2-16 shows the total industrial demand for the three selected cases. Table 2-16 RAILBELT INDUSTRIAL DEMAND FORECAST* (total energy, GWh) Low Mid High 1987 256 256 256 1990 244 244 244 1995 Liz 252 311 2000 174 263 327 2005 176 270 364 2010 178 278 380 *Utility supplied, includes no self-generation Presently, the military bases in the Fairbanks area and the University of Alaska at Fairbanks supply nearly all of their own electric power requirements by operating cogeneration plants that supply both electricity and heat to their respective facilities. Cogeneration facilities efficiently produce electricity and heat in particular proportions. If electricity needs outstrip this balance, the additional electricity is more costly to produce. Particularly for the military bases, there is evidence that electrical needs beyond the balance point could be supplied more efficiently by a local utility, and RI776A 2-20 INTERIM REPORT Decision Focus Incorporated discussions are in progress between the military and Golden Valley Electric Association regarding sale and purchase of this specific increment of power. Forecasts of utility-supplied load will be considered with and without the assumption of power sales beyond the balance point to the military in Fairbanks. (Power sales to the University of Alaska at Fairbanks will also be considered but are expected to be insignificant—i.e., about 2 GWh per year—due to the specific characteristics of the University plant and load). Table 2-17 presents these potential purchases by the military above the cogeneration balance point. Table 2-17 POTENTIAL PURCHASES OF CIVILIAN ELECTRICITY BY THE MILITARY IN THE FAIRBANKS AREA (GWh) Year Eielson Wainwright Greely Total 1990 12.4 14.0 16.2 42.6 2000 15.1 17.0 16.9 48.9 2010 17.2 1923 19.2 55.7 The estimated peak purchase in 1990 consistent with these potentials would be approximately 9.5 MW, and would occur during the summer season. 2.11 ADDITIONAL LOADS SERVED BY THE NORTHEAST INTERTIE (Contractor: Institute of Social and Economic Research) Status: Complete To assess the feasibility of the Northeast intertie proposal, the additional electrical loads that would be connected to the Railbelt grid as a result of the new intertie must be estimated. These loads occur primarily in the Glennallen-Valdez area, presently served by Copper Valley Electric Association (CVEA). Table 2-18 presents the low, mid, and high forecasts of Northeast intertie loads. Peak demand in 1990 is approximately 10.5 MW. In the high case, peak demand increases to 20 MW in the year 2010. The backscatter radar facility will be supplied in whole or in part by on-site generation, and may or may not accept a portion of its electrical requirements from the grid. For the low, mid, and high cases, 0, 2.7, and 3.7 MW average load from the grid were assumed respectively. The mid case assumption is consistent with the bid submitted by CVEA for provision of power to the site. RIT76A 2-21 INTERIM REPORT Decision Focus Incorporated Table 2-18 NORTHEAST INTERTIE LOAD FORECASTS (GWh) Year Low Mid High 1990 51 51 52 1995 51 83 meted 2000 54 85 99 2005 55 87 Za 2010 58 90 La The sharp rise in demand for the high case in 1995 reflects not only the high assumption for the radar site but also assumes construction of a major refinery in Valdez at that time. The second obvious bulge in the high case around 2005 reflects construction of the TAGS line. Of the roughly 50 GWh presently demanded in the CVEA system, roughly 40 GWh is supplied from the Solomon Gulch hydro project in Valdez with the other 10 GWh supplied from diesel generation. Additional energy is available from Solomon Gulch in the summer season, but presently must be wasted in the form of spilled water because: (1) CVEA summer loads are not large enough to absorb it, and (2) the reservoir is not presently large enough to store the additional water for later use during the winter. 2.12 COSTS AND IMPACTS OF ELECTRIC END USE CONSERVATION PROGRAMS (Contractor: Institute of Social and Economic Research) Status: Draft complete; Final report due March 1989. The contractor was directed to describe the expected costs and load reduction impacts of the most promising electric end-use conservation programs that could be devised for the Railbelt. While selection of the “most promising" programs requires a comparative assessment and preliminary screening of programs, the economic feasibility of each program will be judged in comparison with generation and transmission options during the final, integrative stage of the Railbelt studies. Based on preliminary screening criteria, nine programs were identified for further analysis. Eight of the nine programs are structured around dealer/contractor rebates, i.e., rebates to the businesses that sell or install eligible efficiency equipment, thereby reducing the price of efficiency investments faced by consumers. The ninth RIT76A 2-22 INTERIM REPORT Decision Focus Incorporated program would provide rebates to the owners and designers of new or remodeled commercial buildings based on the design efficiency of lighting and ventilation systems. All of the programs are intended to encourage the installation of efficient equipment either initially (in the case of the ninth program) or at the time of normal replacement of standard equipment. No intensive retrofit programs are proposed, primarily because they are more expensive (useful equipment is prematurely replaced) and the present cost of electrical generation in the Railbelt is relatively low. However, though the proposed programs are more cost-effective than accelerated retrofit-type programs, they need more time for their effects to fully register. Because the stock of appliances and equipment takes 10 to 20 years to turn over, programs that encourage efficiency upgrades at the time of normal replacement must be in place for 10 to 20 years to have the potential for affecting the entire appliance stock. The nine programs are summarized briefly below, with the residential programs listed first, followed by commercial. 1. Water Heater Conversions: $500 rebate for the conversion of a residential electric water heater to natural gas. 2. Efficient Electric Water Heaters: $40 rebate for the purchase of an electric water heater with an efficiency over 95 percent. 3. Gas Dryer Rebates: $170 rebate for installation of gas piping to a clothes dryer within a residence, $50 rebate for purchase of a gas clothes dryer. 4. Efficient Refrigerator Rebates: $50 rebate for purchase of refrigerator at least 28 percent more efficient than required by new federal appliance efficiency standards. 5. Efficient Freezer Rebates: $50 rebate for purchase of freezer at least 35 percent more efficient than required by new federal appliance efficiency standards. 6. Fluorescent Lamp Rebates: Rebates from $0.30 to $1.80 for purchase of energy efficient fluorescent lamps. vA Electronic Ballast Rebates: $13 rebate paid for each electronic fluorescent ballast. (A ballast is the device used to start and provide proper operating conditions for fluorescent lamps.) RI776A 2-23 INTERIM REPORT Decision Focus Incorporated is Electronic Ballast Rebates: $13 rebate paid for each electronic fluorescent ballast. (A ballast is the device used to start and provide proper operating conditions for fluorescent lamps.) 8. Incandescent to Fluorescent Conversions: $7 to $12 rebates for purchase of compact fluorescent lamps, adapters, and fixtures suitable for replacing incandescent lamps. 9. Sliding-Scale New Construction Rebates: $1 per square foot rebate for every one watt per square foot reduction in lighting or ventilation power consumption below a threshold level. This rebate applies in the commercial sector to new construction or remodel projects, and would be divided (85 percent / 15 percent) between the building owner and the architect/engineer project designer. The commercial lighting programs (#6, #7, and #8 above) generate nearly 60 percent of the expected savings from all nine programs. Within the residential category, the electric water heater conversion program appears to have the most impact and also the lowest cost per kilowatt hours saved. In estimating program impact, care was taken to avoid double counting efficiency measures already assumed to occur within the electric demand forecast (ie., "market driven" efficiency) and to base projected response rates of consumers to these incentives, not only on the available electric end-use data for the Railbelt, but also on the program participation rates reported by others. It is estimated that if the incentive payments for all nine programs were held in place over a period of 20 years, the savings in the 20th year would be approximately 7 percent of estimated load. Load reduction impact builds over the 20-year period up to this 7 percent peak and then declines over the ensuing 20 years due to the termination of incentives, the retirement and replacement of equipment bought earlier with the incentives, and the return to "normal" purchasing behavior. The amount of electricity saved, as well as program cost, is roughly proportional to the length of the program. If the programs were in place for 5 years instead of 20, program impact would peak in year 5 at roughly 2 percent of estimated load, and then decline from there. The total budgetary cost of implementing all nine programs for 20 years is estimated at $67 million in present value terms (ie., a one time "endowment" of $67 million could fund the programs for 20 years if interest earnings on that sum could be earned, retained, and reinvested). Assuming rebates increase with inflation at 4.5 percent per year, the nominal budgetary requirement over the 20-year period is estimated at $190 million. RI776A, 2-24 INTERIM REPORT Decision Focus Incorporated 2.18 TRANSMISSION SYSTEM STABILITY WITH BRADLEY LAKE AT FULL CAPACITY (Contractor: Power Technologies, Inc.) Status: Draft under development; Final report due March 1989. Concerns were raised by the Railbelt utilities that adverse electrical conditions (specifically, electrical "instability") will result if certain transmission faults occur on the Kenai Peninsula when Bradley Lake is operating at full generating capacity, unless corrective measures are implemented in advance. The corrective measure recommended by the utilities is to build the proposed new transmission line between Anchorage and the Kenai Peninsula. The Power Authority commissioned Power Technologies, Inc. (PTI) to analyze the issue and identify the costs and implications of various solutions to the instability problem. For all of the cases now under examination, it is assumed that the transmission system from Bradley Lake to Soldotna will include two paths: a new direct line between the two points plus a connection with the existing Homer Electric line that runs along the west coast of the Kenai Peninsula. The worst case from the standpoint of system stability occurs when the new line is tripped between Bradley and Soldotna while Bradley Lake is operating at peak capacity (formerly estimated at 90 MW but now re-estimated at 119 MW). As a result of that line trip, all 119 MW would be directed across the existing Homer Electric line along the west coast of the Kenai Peninsula. Regardless of stability considerations, the thermal limit of that existing line is insufficient to accommodate 119 MW. Consequently, the output of Bradley Lake must be reduced in the event of a trip on the new line from Bradley to Soldotna, whether or not a new line is built between Anchorage and the Kenai Peninsula. Such reduction in output may cause an outage depending on the amount of spinning reserve available in the system at the time of the event. The base case selected for further study requires that Bradley Lake output be "ramped back" from full output to approximately 60 MW in the event of a trip on the new Bradley to Soldotna line. Work on these issues is still in progress. However, preliminary results indicate that satisfactory electrical conditions can be maintained in that event without a new Anchorage-Kenai Peninsula intertie with the addition of certain stability aids, including a braking resistor at the Bradley Lake site, stabilizers on the Bradley Lake units, and perhaps three series capacitors installed on the Kenai Peninsula transmission system. A preliminary capital cost estimate for these aids is approximately $2.5 million. Additional or alternative stability aids may ultimately be recommended to further improve electrical conditions or to address other transmission failures or operating conditions. RIT76A 2-25 INTERIM REPORT Decision Focus Incorporated Preliminary results further indicate that, assuming the proposed new Anchorage- Kenai Peninsula intertie is in place, comparable stability can be provided with the addition of the braking resistor and stabilizers only. A preliminary capital cost estimate for these aids alone is approximately $750 thousand. RI776A 2-26 INTERIM REPORT Decision Focus Incorporated Section 3 SCREENING ANALYSIS OF POWER SUPPLY ALTERNATIVES AND END-USE PROGRAMS 3.1 OVERVIEW Technology screening is a method commonly used to eliminate uneconomic power generation technologies from further study. By removing technologies that are not advantageous under any anticipated circumstances, one can scrutinize the remaining technologies more fully and find the scenarios under which each remaining technology is optimal. In addition, technology screening allows a comparison of supply-side technologies with end-use conservation techniques to determine which technologies and conservation programs are least expensive on a lifecycle cost basis per measure of energy served or reduced. We begin this section by describing the technology screening method and then the evaluated technologies and their costs. Next, we perform some sensitivity tests to help determine the least expensive power supply technologies under various assumptions. Finally, we compare the costs of supplying loads (per unit of energy) with the costs of reducing loads (per unit of energy) by implementing end-use conservation programs. This helps us gain insight into which end-use programs may be economically attractive. 3.2 INTRODUCTION TO THE TECHNOLOGY SCREENING METHOD Despite the abundance of power generation technologies, no single technology is optimal for supplying all loads. For example, it would be inefficient to use a nuclear plant for peaking purposes. It cannot follow load changes quickly enough, and it has such large fixed costs that producing small amounts of energy would be prohibitively expensive per unit of energy. Conversely, a small combustion turbine, though well suited for following load swings, would be inefficient to operate for baseload service because its variable costs of operation are high relative to other technologies. Typically, technologies with higher capital costs exhibit higher efficiencies compared with lower capital cost technologies. By producing a screening diagram (see Figure 3-1), a visual representation of the tradeoffs between higher capital costs/higher efficiency on the one hand, and lower capital costs/lower efficiency on the other hand, RI776A 3-1 INTERIM REPORT Decision Focus Incorporated can be made to determine the least expensive technologies at each capacity factor and under various economic and performance assumptions. In the diagram, each technology is represented by a curve that relates the unit cost of producing energy (on the vertical axis) plotted against various capacity factors’ under which the technology is assumed to be operating (on the horizontal axis). Real Levelized Costs (Cents/kWh) 30 25 20 15 Technology C 10 - i [ir sores " Technology B 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Figure 3-1. A Screening Diagram In Figure 3-1, three technologies are represented. Technology A has the lowest fixed and highest variable costs. This is typical of technologies used for peaking operation (combustion turbines and internal combustion engines) that are inexpensive to build but expensive to operate. Technology B has the next lowest fixed costs and next highest variable costs. The point where Technology B intersects Technology A is the breakeven point for Technology B, i.e., beyond this capacity factor, Technology B is more economical than Technology A. If Technology A is expected to operate at a capacity factor greater than the breakeven point, then Technology B would be less expensive to operate per unit of energy produced. Finally, Technology C could be ruled out from further study because it is not economical at any capacity factor. Capacity factor represents the average capacity utilization of a plant over a time period, typically one year. For example, a 100% capacity factor represents use of a plant at full load year round; a 50% capacity factor can represent use of a plant at full load for half the year, or use of the plant at half load for the full year. RI776A 3-2 INTERIM REPORT Decision Focus Incorporated 3.3. TECHNOLOGIES EVALUATED We focused on several technologies that are currently available or have been proposed for use in the Railbelt. We evaluated four fuel types: natural gas, coal, wood, and refuse. Within each fuel type, we selected sizes that would exhibit economies of scale yet still represent reasonable capacity additions in the area. The initial data on the technologies were obtained from two sources. For coal-fired CFBs, we used the Stone & Webster report to APA dated November 1988 [1], and for all other technologies, we used the EPRI Technical Assessment Guide, Volume 1: Electricity Supply-1986 [2]. This data was also checked against available information on similar plants in the Railbelt area. A brief description of each technology follows. 8.3.1 Coal-Fired Circulating Atmospheric Fluidized Bed Combustion (CFB) Stone and Webster identified CFB as the preferred technology for future coal- fired plants in the Railbelt primarily due to expected capital cost savings relative to conventional plants. In a CFB, fuel is combusted on a bed of ash that is fluidized by a mass of air levitating it from below. CFBs have excellent fuel flexibility and can be operated with blends of standard coal and low-cost waste coal. Although the Stone & Webster report evaluated three different sizes (150 MW, 100 MW, and 50 MW), we initially decided to analyze only the 150 MW plant. Due to economies of scale, we assume that if a technology is uneconomic at the largest size, it would be uneconomic at the smaller, more expensive (per installed kW) sizes. To eliminate delivery charges, we examined minemouth plants located at Beluga and Healy. Coal price forecasts at both locations were developed by ISER (refer to Section 2). In addition, we include a third coal plant located at Healy that can handle a 25 percent waste coal blend. In a later sensitivity, we also evaluate a 50 MW plant utilizing a 75 percent waste coal blend. The estimates of waste coal percentages were derived as follows. Presently the Usibelli Coal Mine produces approximately 1.6 million tons of coal per year. If existing sales are maintained, production would increase further if a new coal plant were developed in the Interior. The following estimates of waste coal utilization are based on standard coal production of 2.0 million tons per year. It is estimated that waste coal is produced at the Healy mine in approximately a 1:10 ratio compared with standard coal. As a result, waste coal production of about 200,000 tons per year would be consistent with the selected coal production scenario. A 150 MW coal plant operating at a 75 percent capacity factor would require approximately 10,000,000 MBtu per year in fuel, given a heat rate of 10,000 Btu/kWh. Of this amount, 200,000 tons/year of waste coal could supply about 2,400,000 MBtu, assuming 6000 Btu/lb. Therefore, in the base case adopted for the coal plant screening analysis based on the costs and characteristics of a 150 MW plant, it is assumed that RIT776A 3-3 INTERIM REPORT Decision Focus Incorporated waste coal could supply 25 percent of the fuel requirement. For a 50 MW plant, the same amount of waste coal could supply nearly 75 percent of annual fuel requirements. 3.3.2 Combined Cycle (CC) In a CC plant, natural gas is first burned in a combustion turbine/electric generator. The hot exhaust gases from the turbine are then passed through a heat recovery steam generator that produces steam for a conventional steam turbine/electric generator. This results in greater efficiency than would be obtained from the combustion turbine alone. Our analysis used a 220 MW plant size. 3.3.3 Conventional Combustion Turbine (CT) Because the exhaust gases from the combustion turbine are not utilized, CTs are less efficient than CCs. However, because they do not include a heat recovery steam generator, CTs are much less expensive than CCs. We used a 75 MW plant for this analysis. 3.3.4 Wood-Fired Steam Turbine (WST) In a WST plant, wood wastes (sawmill or refuse) are burned in a boiler that produces steam to drive a steam turbine. Some of the steam can be extracted from the turbine for cogeneration purposes. We assume that this plant is operated for cogeneration and as such, the plant produces 3.8 lb/kWh of steam at a value of $3 per 1000 Ib. In addition, we assumed that this 24 MW plant was located where wood wastes could be obtained relatively inexpensively. 3.3.5 RDF Steam Turbine (RST) In an RST plant, municipal solid wastes are burned on a moving grate in a water wall incinerator to produce steam to drive a steam turbine. Although the fuel has a low heating value and high moisture content, it is usually free of negative cost because the plant operator may be able to collect a tipping fee for the refuse. The plant we analyzed was 45 MW. Because we were focusing on large capacity additions, we did not consider internal combustion engines as they are not usually economical for large capacity additions and function mainly as emergency backup. RIT76A 3-4 INTERIM REPORT Decision Focus Incorporated 3.4 EVALUATING COSTS FOR EACH TECHNOLOGY We evaluated costs for each technology by computing the net present value of total capital, operating, and fuel costs at several capacity factors over the life of each plant. Each of these cost streams was analyzed separately and discounted back to the initial operating year. We then levelized the sum of these costs to obtain real levelized yearly costs. The levelized costs were reported in cents/kWh. It was assumed that each plant is brought on line in 1995. All costs are expressed in 1987 dollars. The next two sections explain how we calculated these costs. 3.5 FIXED COSTS Fixed costs are usually divided into two main groups: the carrying charges (total plant investment costs) and the fixed operating and maintenance (O&M) costs (typically labor, some maintenance charges, and overhead). 3.5.1 Carrying Charges The carrying charges on a plant are those costs associated with amortizing the total plant investment (i.e., the total capital requirement of the plant). The total plant investment is comprised of two primary parts: Le The total cost for construction, materials and land including any royalties on the technology. This is usually expressed in overnight construction costs. For the coal-fired plants, estimates were obtained from the Stone & Webster report. For all other technologies, we obtained estimates from the EPRI Technical Assessment Guide. We adjusted these estimates for Alaska costs and conditions. These adjustments were based on the estimated time, complexity, and skilled labor requirement for plant construction. For the CC, WST, and RDF, we assumed that costs would be 50 percent higher in Alaska than in the lower 48. For the CT, we assumed that costs are 25 percent higher. These factors were selected to maintain a level of consistency with the high labor cost adjustment implied in the Stone & Webster coal plant estimates. 2. Allowance for funds used during construction (AFUDC), ie., interest paid on the funds used during construction. AFUDC was computed for each technology by assuming a uniform construction expenditure schedule over an assumed construction period. R1776A 3-5 INTERIM REPORT Decision Focus Incorporated The total plant investment costs (87$/kW) are shown in Table 3-1. Because it has been proposed that the coal plants could be built for as little as $1600/kW (refer to Section 2), we include such an assumption in a later sensitivity case. Table 3-1 TOTAL PLANT INVESTMENT COSTS Original Total Plant Alaska Plant Size Cost Plant Cost Investment Technology* (MW) (87$/kW) Multiplier (87$/kW) BCFB 150 2167 1.00 2343 HCFB 150 2079 1.00 2247 HCFB-WC 150 2136 1.00 2309 co 220 579 1.50 928 CT 75 341 ome 446 WST 24 2107 1.50 3454 RDF 45 4414 1.50 7237 *NOTE: Technology Plant Fuel Name Type Type BCFB Beluga/Circulating Fluidized Bed Beluga Coal HCFB Healy/Circulating Fluidized Bed Healy Coal HCFB-WC Healy/Circulating Fluidized Bed/Waste coal Healy Coal & Waste Coal CC Combined Cycle Gas CT Combustion Turbine Gas WST Wood Steam Turbine Wood RDF Refuse Derived Fuel Refuse (1) The Original Plant Cost column shows plant costs ($/kW) in 1987 dollars before any multipliers have been applied. The Stone & Webster coal plant estimates already incorporate Alaska cost adjustments, and also include one-time training and commissioning costs. For the Healy waste coal plant, the equipment necessary to allow processing of a 25% waste coal blend is included. Finally, the coal plant costs are discounted to 1987 dollars (from 1988 dollars). (2) The Total Plant Investment column shows plant costs with AFUDC, and adjusted for Alaska. RIT76A 3-6 INTERIM REPORT Decision Focus Incorporated 3.5.2 Fixed O&M Costs Fixed O&M costs are those expenses that are incurred even when a plant is not utilized, and that do not increase when the plant is operated. Typical fixed O&M costs are: labor, a portion of the maintenance, administration, and general overhead. In general, baseloaded plants exhibit higher fixed O&M costs than peaking plants. The fixed O&M costs for each of the technologies were determined as follows: iz Coal Plants—Fixed O&M costs for the coal plants were obtained from the Stone & Webster report (Table 7-11). Because the Stone & Webster report does not show a breakdown between fixed and variable O&M costs, we assumed that all of the labor and administration and overhead expenses, along with half of the maintenance and replacement costs were fixed, the remaining portion of the O&M costs (consumables and the other half of the maintenance and replacement costs) were assumed to be variable. 2: CC and CT—Fixed O&M costs for combined cycle and combustion turbine plants were established consistent with existing plants in the Railbelt. 3. WST and RST—Fixed O&M costs for wood and refuse plants were obtained from the EPRI data with a 25 percent adjustment for Alaska costs. The fixed O&M costs (in 1987 dollars) for all the technologies are shown in Table 3-2. Table 3-2 FIXED O&M COSTS (87$/kW/Yr) BCFB Du aa 7 HCFB 58.16 HCFB-WC 58.16 cc 10.00 ct 12.00 wsT 75.82 RDF 141.81 RI776A 3-7 INTERIM REPORT Decision Focus Incorporated 3.6 VARIABLE COSTS Variable costs are those costs that vary as a plant is operated. They are comprised of fuel costs and variable O&M costs. 3.6.1 Fuel Costs Fuel costs for each technology are a function of the unit’s heat rate (i.e., efficiency), the fuel price, and the capacity factor of the unit. Full load heat rates were used in this screening analysis. The Beluga and Healy coal prices shown are for minemouth plants, and the natural gas prices are based on Enstar contract terms (refer to Section 2). The wood prices are for bone-dry saw mill dust at $10/ton and 6000 Btu/lb. The refuse prices are for a tipping fee of $15/ton and 4900 Btw/b.” All fuel prices remain constant in real terms except for natural gas. We used the low natural gas price scenario in the base case (probability = 60%). In that scenario, gas escalates at 0.16 percent real until 2010 (i.e., negligible real escalation), and then remains constant in real terms. In a later sensitivity, we analyze higher gas prices and look at very high gas escalation in the post 2010 period due to assumed resource depletion. The fuel cost of the Healy CFB with coal technology (HCFB-WC) assumes a 25 percent coal blend. Table 3-3 lists the fuel costs for all technologies. Table 3-3 1995 FUEL COSTS (1987 dollars) Heat Rate Fuel Cost* Tech (Btu/kWh) (87$/MBtu) (87$ /MWh) BCFB 10113 2.15 11.63 HCFB 10114 1.30 1.3}. 1.5) HCFB-WC 10114 0.99 10.04 cc 8394 1.54 12.93) cT 11600 1.54 17.86 WST 16250 0.75 12.19 RDF** 16300 -1.39 22.598) *Waste coal cost = 0.07 $/MBtu **The negative RDF fuel cost reflects a tipping fee usually paid to the RDF plant operator for collecting refuse. ?Based on phone conversation with Babcock & Wilcox conveying their experience with wood and refuse-fired cogeneration facilities around the country. RIT76A 3-8 INTERIM REPORT Decision Focus Incorporated 3.6.2 Variable O&M Variable O&M costs are comprised mainly of maintenance charges and consumables such as water, oil, sorbents, and so on. Variable O&M costs for each of the technologies were estimated as follows: a Coal Plants—Variable O&M costs for the coal plants were obtained from the Stone & Webster report (Table 7-11). As reported in Section 3.5.2, these costs were assumed to be all of the consumables’ expenses and half of the maintenance and replacement costs. It was also assumed that the O&M costs shown were for a plant operating at 75 percent capacity factor. 2; CC and CT—Variable O&M costs for the combined cycle and combustion turbine plants were established consistent with existing plants in the Railbelt. 3. WST and RST—Variable O&M costs for the wood and refuse plants were obtained from the EPRI TAG report; an adjustment of 25 percent was applied over the EPRI data. The variable O&M costs (in $/MWh) for each of the technologies are shown in Table 3-4. The total plant variable O&M costs are proportionate to the capacity factor. Table 3-4 VARIABLE O&M COSTS Variable O&M Costs Tech (87$/MWwh ) BCFB 3.89 HCFB 3.83 HCFB-WC 3383 CC 1.50 ct 1.50 WwST* -1.11 RDF 14.47 *The negative WST variable O&M costs include a "credit" for the steam produced by the plant. RI776A 3-9 INTERIM REPORT Decision Focus Incorporated 3.7 ECONOMIC ASSUMPTIONS The following economic assumptions were made: Real Discount Rate = 4.5% O&M Real Escalation Rate = 0% Capital Real Escalation Rate = 0% Unit Life = 30 years 3.8 BASE CASE RESULTS Using the previous assumptions, we found that the gas-fired technologies (combustion turbine and combined cycle) were more economic than all other technologies at all capacity factors examined.* The least expensive coal plant was the Healy waste coal plant. However, even at high capacity factors, the combined cycle was less expensive. As an illustration, at 75 percent capacity factor, the levelized cost of the Healy waste coal plant was 4.43 cents/kWh while the levelized cost of the CC was 2.48 cents/kWh. On a lifecycle cost basis, the cost of the coal plant was 79 percent higher. The refuse and wood technologies were not economically viable under the base case assumptions. The refuse technology was burdened too heavily with high fixed costs. The wood-fired plant was burdened with both high fixed costs and high variable costs. Table 3-5 lists the levelized costs for each plant at various capacity factors. Figure 3-2 shows the base case screening diagram for all plants except the regular Healy coal (its costs are very close to the Beluga plant). Figure 3-3 focuses exclusively on the CT, the CC, and the Healy waste coal plant. Table 3-5 BASE CASE Levelized Total Costs at Various Capacity Factors (87 cents/KWh) Tech ONES 0.3 0.45 0.6 O75 0.9 BCFB 16.89 9.22 6.67 5.39 4.62 4.11 HCFB 16.62 9.16 6.67 5.43 4.68 4.18 HCFB-WC 16.60 8.99 6.46 5.19 4.43 3.92 cc 6.56 4.01 336 2.73 2.48 peel cT 4.96 3.46 2.96 2: 2.56 2.46 WST 23.02 12.06 8.41 6.58 5.49 4.76 RDF 43.79 21.49 14.06 10.34 8.11 6.62 5In practice, capacity factors over 90 percent are difficult to achieve because plant equivalent availabilities are limited by planned and unplanned outages. RIT76A 3-10 INTERIM REPORT Decision Focus Incorporated Real Levelized Costs (Cents/kWh) Technologies BCFB HCFB-WC cc CT 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Figure 3-2. Technology Screening (Base Case—All Technologies) Real Levelized Costs (Cents/kWh) 20 155 Technologies —— HCFB-WC 10 F -- 66 —= cT 5 | 0 L 1 ! 1 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Figure 3-3. Technology Screening (Base Case—Selected Technologies) RI776A 3-11 INTERIM REPORT Decision Focus Incorporated 3.9 SENSITIVITY ANALYSIS To test the stability of the base case results, and to see under what circumstances they would change, the following sensitivity studies were done: iS 2. Middle gas prices. High gas prices. Low coal plant capital costs. Low coal plant capital costs combined with high fraction of waste coal. 8.9.1 Sensitivity #1: Middle Gas Prices In this case, we modified the gas prices to reflect the Alaska Power Authority’s middle gas price scenario (Probability = 30%) between 1995 and 2010, followed by sharp escalation during the following 15 years (refer to Section 2): Period Period Years 1995-2010 2010-2025 Price (1st Year) $1.98/MBtu $2.28/MBtu Real Escalation per Year 0.9% 7.2% Price in Year 2025 = $6.50/MBtu As expected, this change increased the production costs of the gas-fired technologies. However, the CC plant still maintained a clear economic advantage over the Healy waste coal plant, even at high capacity factors. At 75 percent capacity factor, the levelized cost of the CC was 3.47 cents/kWh, while the levelized cost of the Healy waste coal plant (which was the lowest cost coal plant), was 4.43 cents/kWh. On a lifecycle cost basis, the cost of the coal plant was 28 percent higher. Table 3-6 displays the levelized costs for all technologies at various capacity factors, and Figure 8-4 shows the screening curve for the CT, the CC, and the Healy waste coal plant. RI1T76A 3-12 INTERIM REPORT Decision Focus Incorporated Table 3-6 SENSITIVITY CASE #1: MIDDLE GAS PRICES Levelized Total Costs at Various Capacity Factors (87 cents/KWh) Tech 0.15 0.3 0.45 0.6 0.75 0.9 BCFB 16.89 9.22 6.67 5.39 4.62 4,11 HCFB 16.62 9.16 6.67 5.43 4.68 4.18 HCFB-WC 16.60 8.99 6.46 5.19 4.43 Bawa cc 7.55 5.00 4.15 3.72 3.47 a.20 cr 6.32 4.83 4.33 4.08 3.93 3.83 WST 23.02 12.06 8.41 6.58 5.49 4.76 RDF 43.79 21.49 14.06 10.34 8.11 6.62 Real Levelized Costs (Cents/kWh) 20 Technologies —— HCFB-WC —*- CC —-2- CT 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Middle gas price scenario as defined by the APA. Probability = 30%. Figure 3-4. Technology Screening (Middle Gas Price) R1776A 3-13 INTERIM REPORT Decision Focus Incorporated 3.9.2 Sensitivity #2: High Gas Prices In this case, we modified the gas prices to reflect the Alaska Power Authority’s high gas price scenario (Probability = 10%) between 1995 and 2010, followed by sharp escalation during the following 15 years (refer to Section 2): Years Price (1st Year) Period 1995-2010 $2.31/MBtu Real Escalation per Year 1.8% Final Price in Year 2025 = $7.50/MBtu Period 2010-2025 $3.01/MBtu 6.3% In this sensitivity case, the CC plant still maintained an economic advantage over the Healy waste coal plant. 11 percent higher. Table 3-7 shows the levelized costs for all technologies. illustrates the CT, CC, and Healy waste coal costs. For example, at 75 percent capacity factor, the levelized cost of the CC was 4.00 cents/kWh while the levelized cost of the Healy waste coal plant was 4.43 cents/kWh. On a lifecycle cost basis, the cost of the coal plant was SENSITIVITY CASE #2: Table 3-7 HIGH GAS PRICES Figure 3-5 Levelized Total Costs at Various Capacity Factors (87 cents/KWh) Tech BCFB HCFB HCFB-WC cc cT WST RDF R1776A 0 LG. 16. 16: 8. un p< As Nr rPNHUUwDWwW wo 0.3 0.45 0.6 On75) 0.9 »22 6.67 5.39 4.62 4,14 -16 6.67 5.43 4.68 4.18 -99 6.46 5.19 4.43 31.92 193 4.68 4.26 4.00 383 re 5.07 4.82 4.67 4.57 -06 8.41 6.58 5.49 4.76 -49 14.06 10.34 8.11 6.62 3-14 INTERIM REPORT Decision Focus Incorporated Real Levelized Costs (Cents/kWh) 20 Technologies —— HCFB-WC —* CC — CT 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor High gas price scenario as defined by the APA. Probability = 10%. Figure 3-5. Technology Screening (High Gas Price) 3.9.3 Sensitivity #3: Low Coal Plant Capital Costs In this case, we assumed gas prices as in the base case but reduced the coal plant capital costs by 33 percent. This was equivalent to assuming a 25% adjustment factor for Alaska costs and conditions, instead of the 87 percent cost differential implied by the Stone & Webster cost analysis. The coal plant costs were reduced as follows: Plant Cost ($/kW) Pl B C Sensitivity #3 Beluga 2343 1566 Healy 2247 1502 Healy Waste 2309 1543 The gas-fired technologies remained most economic at all capacity factors. At 75 percent capacity factor, for example, the levelized cost of the Healy waste coal plant was 3.71 cents/kWh compared to 2.48 cents/kWh for the CC. Table 3-8 shows the levelized costs for all technologies, and Figure 3-6 illustrates the costs of the CT, CC, and Healy waste coal plants. R1776A 3-15 INTERIM REPORT Decision Focus Incorporated Table 3-8 SENSITIVITY CASE #3: LOW COAL PLANT CAPITAL COSTS 1. COAL PLANT CAPITAL COSTS = (1.25/1.87) * BASE CASE COSTS 2. BASE CASE GAS PRICES. Levelized Total Costs at Various Capacity Factors (87 cents/KWh) Tech 01.15) 0.3 0.45 0.6 0.75 0.9 BCFB 13.26 T 5.46 4.48 3.89 3.50 HCFB 13.14 7.42 5 491: 4.56 3.99 3.60 HCFB-WC 13/03 oral Saal 4.30 3.71 3.39 cc 6.56 4.01 Sale Zeta 2.48 2.31, ct 4.96 3.46 2.96 are, 2.56 2.46 WST 23.02 12.06 8.41 6.58 5.49 4.76 RDF 43.79 21.49 14.06 10.34 6.11 6.62 Real Levelized Costs (Cents/kWh) Technologies —— HCFB-WC —* CC 0 1 1 1 = 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Alaska cost conversion factor of 1.25 (instead of 1.87) applied to Healy waste coal plant. Costs dec:$2309/kW->$1544/kW RI1776A, Figure 3-6. Technology Screening (Low Coal Plant Capital Costs) 3-16 INTERIM REPORT Decision Focus Incorporated 8.9.4 Sensitivity #4: Low Coal Plant Capital Cost Combined with High Fraction of Waste Coal This sensitivity case incorporates the following assumptions: 1 A 50 MW plant with construction costs of $1600/kW is used. 2. Waste coal supplies 75 percent (rather than 25 percent) of the fuel requirement. 8. Fixed O&M costs are assumed at $75/kW (rather than $115/kW for the 50 MW plant in the Stone & Webster estimates). 4. Gas prices are set consistent with the base case. Under these assumptions, the gas-fired technologies continued to be most economic at all capacity factors. The levelized cost of a waste coal plant at 75 percent capacity factor fell to 3.44 cents/kWh compared with 2.48 cents/kWh for the CC. Table 3-9 shows the levelized costs for all technologies, and Figure 3-7 displays the costs of the CT, CC, and the 50 MW Healy waste coal plant. Table 3-9 SENSITIVITY CASE #4: LOW COAL PLANT CAPITAL COSTS COMBINED WITH HIGH FRACTION OF WASTE COAL . CAPITAL COSTS = $1600/KW . SIZE = 50 MW OpwnrH - BASE CASE GAS PRICES. . WASTE COAL FRACTION = 75% . FIXED O&M AT $75/KW, VARIABLE O&M AT $4.34/MWH Levelized Total Costs at Various Capacity Factors (87 cents/KWh) Tech OLS) (Oe) 0.45 0.6 O.75 0.9 HCFB-WC 13.98 1.39 5. 49) 4.10 3.44 3.00 icc 6.56 4.01 S16 2213 2.48 2s Cr 4.96 3.46 2.96 Zatd 2.56 2.46 WwsT 23302 12.06 8.41 6.58 5.49 4.76 RDF 43.79 21.49 14.06 0.34 8.321 6.62 RIT76A 3-17 INTERIM REPORT Decision Focus Incorporated Real Levelized Costs (Cents/kWh) Technologies 8 —— HCFB-WS — CC 6 “2 CT 4 2 0 4 1 4 1 0.15 0.3 0.45 0.6 0.75 0.9 Capacity Factor Coal Plant Cost = $1600/kW 75% Healy waste coal blended in 50MW Healy Location Figure 3-7. Technology Screening (Low Coal Plant Capital Costs and High Fraction of Waste Coal) 3.10 COMPARATIVE EVALUATION OF SUPPLY-SIDE TECHNOLOGIES AND END-USE CONSERVATION PROGRAMS We then compared the least expensive supply-side options to the demand-side alternatives. Table 3-10 lists the nine programs which were evaluated and their respective levelized costs [3]. The programs were evaluated with respect to base case and middle case gas prices. The results are shown in Figures 3-8 and 3-9. The base case comparison clearly suggests that three of the programs are likely to be economic: two of the commercial programs (more efficient fluorescent lamps and conversions from incandescent to fluorescent lighting), and one residential program (conversions from electric to gas water heating). In the middle case, all of the other programs (except the efficient freezer program) emerge as promising candidates for implementation. “The demand-side alternatives are end-use conservation programs that are designed to reduce loads by increasing the average efficiency of residential and commercial appliances and equipment and by converting certain electric appliances to natural gas operation. By providing consumers with monetary incentives to upgrade or replace their equipment, reductions in load can be achieved. RIT76A 3-18 INTERIM REPORT Decision Focus Incorporated Table 3-10 END-USE PROGRAMS Annual* Cents/kWh Load Program Saved Factor A Water Heater Conversions 2.0 64 B Efficient Water Heaters 2.3 -85 C Gas Dryer Rebates 342 59 D Efficient Refrigerators 351 .88 E Efficient Freezers 3.5 1.02 F Efficient Fluorescent Lamps 2.1 49 G Electronic Ballasts 3.5 49 H Incandescent Conversions 1.8 41 I Sliding-Scale Rebates 257 -48 Source (Table 1-2 ISER End use report) Annual energy saved (kWh) Load reduction on peak (kW) X 8760 (hrs/year) * Annual load factor = Real Levelized Costs (Cents/kWh) Technologies i E <a |i CC —+ CT 4 — End-use Conservation 1 ins 0 1 1 1 0.15 0.45 0.75 1.05 Capacity Factor Figure 3-8. Supply-Side Technologies and End-Use Conservation Programs (Base Case) RIT776A 3-19 INTERIM REPORT Decision Focus Incorporated Real Levelized Costs (Cents/kWh) 8 6p Technologies al —-*- CC G AN it D aE —-2- CT al B 4 End-use Conservation A Qe a OF aA 0 1 1 1 0.15 0.45 0.75 1.05 Capacity Factor Middle gas price scenario as defined by the APA. Probability = 30%. Figure 3-9. Supply-Side Technologies and End-Use Conservation Programs (Middle Gas Prices) 3.11 CONCLUSIONS Gas-fired technologies were identified as the most economic power supply alternatives under all assumptions tested. In the base case, the levelized cost of coal- fired generation exceeded the gas alternative by around 75 percent at high capacity factors. Coal-fired generation was not identified as economic under any of the assumptions tested. Wood and refuse-fired technologies were more expensive than both gas and coal. Three of the end-use conservation programs showed high potential for economic feasibility in the base case; five others may be economic under higher (middle case) gas price assumptions. 3.12 REFERENCES {1] "Estimated Costs and Environmental Impacts of Coal-Fired Power Plants in the Alaska Railbelt Region,” Stone & Webster Engineering Corporation report to Alaska Power Authority, November 1988. (2] Technical Assessment Guide (TAG), Volume 1: Electricity Supply 1986, Electric Power Research Institute. [3] Analysis of Electrical End-Use Efficiency Programs for the Alaskan Railbelt, Institute of Social and Economic Research, University of Alaska, Anchorage, November 1988, Draft. RIT76A 3-20 INTERIM REPORT Decision Focus Incorporated Section 4 IMPACT OF PROPOSED INTERTIES ON SYSTEM RELIABILITY 41 OVERVIEW This section analyzes the effects that upgraded or new interties would have on Railbelt service reliability. Reliability is an important attribute of electric power systems because the value of most power used greatly exceeds its cost. The costs to utilities of consumer outages may be relatively small when measured in terms of lost revenues. However, the costs incurred by consumers can be large, particularly for certain types of industrial and commercial customers. Upgraded or new interties could affect both the frequency and the duration of customer outages. This study considers the effects on reliability of: 1. The new Kenai-Anchorage line. 2. The Anchorage-Fairbanks upgrade. 3. The new Anchorage-Fairbanks Northeast intertie. The assessment of the value of improved system reliability due to new/upgraded transmission lines requires estimates of the intertie impacts on customer outages and of the costs of avoided outages. The costs are difficult to calculate and the analysis of potential changes in customer outages in the Railbelt is complicated by the addition of Bradley Lake. Because an in-depth study was not possible within the time constraints of this study, we used the results of recent research on the costs of consumer outages compiled by the Electric Power Research Institute and a detailed study performed by Ontario Hydro that we judged to be the most applicable to the Railbelt. We also analyzed historical outages data in the Railbelt and discussed expectations for potential changes in customer outages with the Railbelt utilities, APA, and Power Technologies Incorporated. This section examines historical customer outages in the Railbelt, describes the potential changes in customer outages that would result from the proposed RIT776A 4-1 INTERIM REPORT Decision Focus Incorporated new/upgraded transmission lines, and summarizes the costs of customer outages and the value of improved system reliability. 4.2 ANALYSIS OF HISTORICAL CUSTOMER OUTAGES 4.2.1 Customer Outages To determine the historical level of customer outages in the study area, we collected information on outages in 1986 and 1987 from the eight utilities’ that could be affected by one or more of the intertie proposals. Because only outages that could have been eliminated with the implementation of new or upgraded interties were important to this study, only transmission and generation related outages were considered.” The information we received from the utilities varied in format and detail quite considerably. Some utilities reported individual outages down to the feeder level, while others simply summarized the total outage hours per customer per year. Table 4-1 summarizes the form of the data we received from each utility. Appendix A includes the detailed outages data. 4.2.2 Customer Unserved Energy Customer unserved energy due to a customer outage is the electric energy that would have been demanded by the customer if the customer was not subject to the outage. The total unserved energy of an outage is the total number of customers affected by the outage multiplied by the average demand per customer and by the duration of the outage. For example, if 2500 customers were without power for one hour and the average demand per customer was 5 kWh/hour, then the total unserved energy would be 2500 customers * 5 kWh/hr * 1 hour = 12500 kWh or 12.5 MWh. For outages where different numbers of customers were affected for different amounts of time, the utilities often recorded multiple outages affecting different numbers of customers for different durations. For example, if an outage affects 5000 customers, but half are restored within half an hour and the other half within an hour, the outage would be recorded as one outage for half an hour for 2500 customers and another outage for one hour for the 2500 other customers. "AMLP, CEA, MEA, HEA, SES, GVEA, FMUS, and CVEA. "Transmission line outages occur when a transmission line goes down and the receiving area is unable to meet the new demand through spinning reserves. Generation-related outages occur when a generating unit goes down, and demand cannot be met because of insufficient spinning reserves or inadequate access to spinning reserves through the transmission system. RI776A 4-2 INTERIM REPORT Decision Focus Incorporated Table 4-1 FORM OF UTILITY DATA ON CUSTOMER OUTAGES Utility How Outages Were Reported AMLP Date, time, duration, cause, location, and number of customers affected for each outage. CEA Total unserved energy per customer from REA Form 7 (Chugach detailed outage information was not available at the time of the survey). MEA Date, time, duration, cause, location, and number of customers affected for each outage. HEA Date, time, duration, cause, location, and number of customers affected for each outage. SES Date and average outage time of transmission line outages only. FMUS Date, time, duration, cause, location, feeder, and type of feeder demand for each feeder. GVEA Outage hours per customer from REA Form 7, along with breakdowns of outages by duration for five years (1983 to 1987). CVEA Outage hours per customer from REA Form 7 for six years (1982 to 1987). Customer demand varies by the time of day and by the time of year. For example, load is typically higher in the winter than in the summer, and load is higher during the workday than during the weekend. Thus the average demand was modified to reflect the time of the outage by dividing the year into two time periods: peak and off-peak. Peak hours were defined as the highest half demand hours of the year, and off-peak hours were defined as the lowest half demand hours of the year. A peak to off-peak ratio was then determined for each utility, and a corresponding weight was applied to each outage’s average customer demand depending upon whether an outage occurred during the peak or off-peak period. In addition, using the residential demand for each utility [1] and [2], we determined the load splits between residential and industrial/commercial customers. This information was used to determine the unserved energy breakdowns by customer type and proved useful later in evaluating customer outage costs. Table 4-2 shows the average demand per customer, the off-peak to peak load ratios, and the residential and industrial/commercial load fractions for each utility. RIT76A 4-3 INTERIM REPORT Decision Focus Incorporated Table 4-2 DEMAND BY UTILITY AND RESIDENTIAL-INDUSTRIAL/COMMERCIAL SPLIT (1986) Average Total Number Demand/ Resid’al Demand (Fraction) Off-Peak Demand of Customer Demand ----------------- to Peak Utility (MWh/yr) Customer (kWh/hr) (MWh/yr) Resid’al Ind/Comm Ratio AMLP 817217 30311 3.08 178375 0.22 0.78 0672 CEA 918322 61222 alee 478040 0.52 0.48 0.74 MEA 418656 PHELPS Doe 272746 0.65 0.35 0.61 HEA 397024 16914 2.68 139903 0.35 0.65 0.78 SES 33315 1626 2.34 11873 0.36 0.64 On 75 CVEA 43570 2310 2229) 11750 0.27 0.73) Ons GVEA 451716 26053 1.98 181389 0.40 0.60 0.74 FMUS 145865 6334 2.63 26554 0.18 0.82 0.74 Total 3225685 172495 1300630 0.40 0.60 Utility Outages. Evaluating the unserved energy by utility, we found that most of the utilities had outage hours ranging from a little more than four hours per customer per year, to less than one hour per customer per year. SES was somewhat unusual. SES exceeded all other utilities in terms of outage hours (10.3 hours/customer/year) and unserved energy (24.2 kWh/customer/year). Since SES imports all of its power from CEA, mostly via the existing Anchorage-Kenai line (University-Daves Creek), whenever that line or SES’s link to it goes down, all the SES customers are without power until either the transmission line is restored or the SES can get its diesel generators on-line, which takes 15 to 45 minutes. Among the other utilities, MEA, which also buys all of its power from CEA, had the next largest outages with 4.4 outage hours per customer per year and 7.5 kWh of unserved energy per customer per year. CEA had the next largest outages with approximately 2.7 outage hours per customer per year although AMLP had more unserved energy (7.1 kWh/customer/year versus 4.7 kWh/customer/ year at CEA) due to its much greater fraction of industrial/commercial load.* The Copper Valley area (CVEA) and the Fairbanks area have the lowest outage hours (and unserved energy) per customer. Figure 4-1 shows the total outage hours per customer by utility for the survey period, Figure 4-2 shows the corresponding total unserved energy per customer by utility, and Figure 4-3 shows the split of unserved energy by utility. 378% of load at AMLP compared to 48% of load at CEA. R1776A 4-4 INTERIM REPORT Decision Focus Incorporated Les 4-4 34 a d Ave iu F | Lal le: SES Figure 4-1. Outa, - i CC lta energy = 24.19 2. Un served E Te ey f Railbelt z g ® ° iS) cS ro) © ft es 1 l ce 5 2 g REE 3 Uv a3 m @ S 8 INTERIM REPORT Decision Focus Incorporated MEA 208.8MWh 22% AMLP 216.4MWh 23% HEA 95.1MWh 10% SES 39.3MWh 4% FMUS 27.3MWh 3% GVEA 76.8MWh 8% CVEA 4.5MWh 0% CEA 286.9MWh 30% Total Railbelt Unserved Energy=955MWh/yr Figure 4-3. Distribution of Railbelt Unserved Energy by Utility (1986/87 Average) Area Outages. The Railbelt averaged 955 MWh of unserved energy over the survey period (see Table 4-3). This came to approximately 2.6 outage hours per customer per year, and about 5.5 kWh of unserved energy per customer per year. Residential customers experienced approximately 44 percent of the unserved energy. Anchorage and Kenai experienced the greatest number of outage hours per customer: 2.91 and 2.71 hours per customer, 6.12 and 6.23 kWh of unserved energy per customer per year. Fairbanks and Copper Valley experienced the lowest outage hours per customer: 1.53 and 0.91 outage hours per customer, 3.21 and 1.95 kWh of unserved energy per customer. Most of the total unserved energy (69 percent) was in Anchorage; Kenai had 20 percent and Fairbanks had the other 11 percent; Copper Valley had less than 1 percent. Figure 4-4 illustrates customer total outage hours by area, Figure 4-5 shows customer unserved energy by area, and Figure 4-6 shows the unserved energy split by area. R1776A 4-6 INTERIM REPORT Decision Focus Incorporated Table 4-3 SUMMARY OF OUTAGE HOURS AND UNSERVED ENERGY FOR THE RAILBELT Average 1986 1987 (1986/87) Outage Hours Per Customer Seat 1.82 2.59 Unserved Energy (kWh per customer) 7.19 3.89 5.54 Unserved Energy (MWh) 1239.79 670.36 955.07 Outage Hours per Customer per year Anchorage Kenai Peninsula Fairbanks Copper Valley AREA Figure 4-4. Outage Hours of Railbelt Areas (1986/87 Average) R1776A 4- 7 INTERIM REPORT Decision Focus Incorporated Unserved Energy (kWh/Customer/yr) = = Anchorage Kenai Peninsula Fairbanks Copper Valley AREA Figure 4-5. Unserved Energy of Railbelt Areas (1986/87 Average) Anchorage 654.7MWh 68.5% Copper Valley 4.5MWh 0.5% Fairbanks 104.1MWh 10.9% Kenai Peninsula 191.8MWh 20.1% Total Unserved Energy = 955 MWh/Yr Figure 4-6. Distribution of Unserved Energy by Area (1986/87 Average) RI776A 4-8 INTERIM REPORT Decision Focus Incorporated Duration of Outages. Most of the unserved energy in Anchorage and Kenai results from outages longer than one hour (81 and 62 percent, respectively); in Fairbanks, most of the unserved energy occurs in outages of less than 20 minutes (around 60 percent); in Copper Valley, all outages have short duration (less than 20 minutes). The breakdown of outage duration is shown by area in Figure 4-7 and by utility in Figure 4-8. 4.2.3 Causes of Outages At a meeting with representatives from the Railbelt utilities,“ we presented this data and discussed the major causes of the recorded customer outages. The utilities identified natural causes (wind, birds, storms, avalanches, and small animals), faulty equipment, and other (airplanes and operator error) as the major causes of transmission related outages; and improper reserves coordination, faulty equipment and human error as the major causes of generation related outages. Most of the outages in Kenai and Anchorage are transmission line related, primarily due to natural causes. Chugach for example, reported that 9 out of 10 outages were due to transmission line failure. However in Fairbanks, the outages are estimated to be almost equally split between generation and transmission failures. 4.3 POTENTIAL CHANGES IN CUSTOMER OUTAGES The analysis of potential changes in customer outages due to potential upgrade and/or construction of new interties in the Railbelt is complicated by the addition of Bradley Lake. The integration of Bradley Lake in the system will change the pattern of system operations, particularly in the southern Railbelt, and produce a reversal in the direction of power flow between Anchorage and the Kenai Peninsula. We reviewed the historical outages data with the utilities and discussed realistic expectations for potential changes in customer outages. We also met with Power Technologies Incorporated (PTI) and discussed system stability and reliability issues. These discussions also shed light on the potential impacts of Bradley Lake (without any new interties or upgrades) on customer outages in the Kenai Peninsula and in Anchorage. This was important primarily for the analysis of the impacts of the proposed new Kenai-Anchorage intertie. 420 December 1988 with AMLP, CEA, MEA, HEA, and GVEA represented. RIT76A 4-9 INTERIM REPORT Decision Focus Incorporated 100 80 60 40 20 Anchorage Kenai Peninsula Fairbanks Copper Valley AREA Duration of Outage HB 0-20 minutes 20 minutes-1 hr EEG} 1-4 hours > 4 hours (1986/87 Avg) Figure 4-7. Railbelt Areas Unserved Energy by Outage Duration (1986/87 Average) 100 80 AMLP MEA HEA SES FMUS GVEA CVEA CEA Utility Duration of Outage HE 0-20 minutes 20 minutes-1hr EEE) 1-4 hours > 4 hours (1986/87 Avg) Figure 4-8. Railbelt Utilities Unserved Energy by Outage Duration R1776A 4-10 INTERIM REPORT Decision Focus Incorporated Next, we present our analysis of the potential changes in customer outages for each of the three proposed interties separately. We present the impacts of each intertie for each of the three areas, Kenai, Anchorage and Fairbanks.® 4.3.1 Kenai-Anchorage New Intertie The Kenai-Anchorage new intertie would impact customer outages in the Railbelt, primarily in Kenai and Anchorage.® Kenai Outages. Historically, the customer outages on the Kenai Peninsula have been primarily caused by failures of the existing Kenai-Anchorage transmission line.’ With the addition of Bradley Lake, we are advised that most of these outages (estimated at 80 percent excluding Seward’s outages) can be avoided.* With Bradley Lake and the new Kenai-Anchorage transmission line, all of these Kenai Peninsula outages (excluding Seward’s outages) can be avoided. Therefore, the new Kenai-Anchorage line would allow the system to avoid an estimated 20 percent of the customer outages on the Kenai Peninsula (excluding Seward’s outages). Anchorage Outages. With the addition of Bradley Lake, power will be exported from Kenai into Anchorage much of the time. The physical characteristics of the current transmission system limit these transfers to approximately 60 MW.° The reliability impact on Anchorage will depend on the level of export from Kenai to Anchorage and the level of accessible spinning reserve at the time of a failure on the existing Anchorage-Kenai Peninsula line. Anchorage utilities routinely carry spinning reserves in excess of 60 MW, though some of this is presently carried on the Kenai Peninsula. In the post-Bradley Lake period, we assume that at least 60 MW of spinning reserve ®The analysis of the Northeast intertie includes four areas where we add the Valdez-Glennallen area. ®In this outages analysis, we assumed that either of the routes proposed for the new Kenai-Anchorage intertie (i.e., Tesoro or Enstar) would have the same impacts on system reliability. "Note again that all reference to customer outages in this analysis refers exclusively to generation and transmission related outages; ie., only those outages that might be affected by transmission improvements. No outages related to the distribution system are included. ®Based on our discussions with Power Technologies Incorporated, under most operating conditions with Bradley Lake producing energy, Kenai Peninsula generation should be able to adjust to failures of the existing Kenai-Anchorage line such that Kenai Peninsula outages are avoided for these events. This implies substantial improvement in Kenai Peninsula reliability due to Bradley Lake. “Transfer limits will depend on the extent of reconductoring performed by Chugach and on the stability aids added to the system. At this point, the base case analysis has assumed a 75 MW limit at the sending node on the Kenai Peninsula and roughly 60 MW at the receiving node in Anchorage. RIT76A 4-11 INTERIM REPORT Decision Focus Incorporated will be accessible in the Anchorage area, and that failures of the existing Anchorage- Kenai Peninsula line will therefore not cause an appreciable increase in Anchorage area outages. Although it would seem that spinning reserve in Anchorage would be sufficient to cover any outage of the existing Anchorage-Kenai line for transfer levels up to 60 MW, it can be anticipated that such coverage will not be perfect, and that occasional outages in Anchorage may result from failure of the existing line (in the absence of the proposed new line). On the other hand, although the new line should mitigate that problem, it could also create an additional opportunity for outages in Anchorage. This could occur if the new line were to fail at a time when exports from Kenai to Anchorage exceed 60 MW. For exports less than 60 MW, the old line could provide sufficient backup to prevent an outage. We assumed that these two potential impacts of the new line would produce no net change in Anchorage reliability. Fairbanks Outages. The Kenai-Anchorage new intertie would only impact customer outages in Fairbanks indirectly. In this analysis, we assumed that the benefits of the Kenai-Anchorage new intertie would be limited to the Kenai-Anchorage area; i.e., there would be no impacts on customer outages in Fairbanks. 4.3.2 Anchorage-Fairbanks Intertie Upgrade The Anchorage-Fairbanks intertie upgrade would have minor impacts on customer outages in the Railbelt. Kenai Outages. Based on our conversations with the utilities, we assumed that the Anchorage-Fairbanks intertie upgrade would have no impact on customer outages in Kenai. Anchorage Outages. The Anchorage-Fairbanks intertie upgrade would have little impact on customer outages in Anchorage. However, according to our conversation with MEA, around 10 percent of MEA’s unserved energy could be avoided. Fairbanks Outages. According to our conversations with GVEA, the Anchorage- Fairbanks upgrade would have no impacts on customer outages in Fairbanks. 4.3.3 Anchorage-Fairbanks Northeast Intertie The Anchorage-Fairbanks Northeast intertie would impact customer outages in the Railbelt, primarily in Fairbanks and the Copper Valley areas. RI776A 4-12 INTERIM REPORT Decision Focus Incorporated Kenai Outages. Based on our conversations with the utilities, we assumed that the Anchorage-Fairbanks Northeast intertie would have no impact on customer outages in Kenai. Anchorage Outages. The Anchorage-Fairbanks northeast intertie would have little impact on customer outages in Anchorage. However, according to our conversation with MEA, the northeast intertie would help reduce the duration of customer outages around 50 percent of the time. It is therefore estimated that around 25 percent of MEA’s unserved energy could be avoided. Fairbanks Outages. According to estimates provided by GVEA, the Anchorage- Fairbanks Northeast intertie would have no impacts on generation related outages in Fairbanks, but it would have saved eight out of the ten (80 percent) and nine out of the fifteen (60 percent) transmission related outages in 1986 and 1987, respectively, because it provides a redundant path for imports from Anchorage.” We assumed that the Anchorage-Fairbanks Northeast intertie would save 70 percent of the transmission related customer outages in Fairbanks. Since around half of the outages in Fairbanks are generation related, we estimated that the Anchorage-Fairbanks intertie would save 35 percent of the customer outages in Fairbanks. Copper Valley Outages. Because of the small fraction of unserved energy in this area (less than one-half percent of the Railbelt unserved energy, refer to Figure 4-3), and because the Anchorage-Fairbanks Northeast intertie would link this electrically isolated area to the Railbelt system, we assumed that all of the Copper Valley outages would be avoided. Table 4-4 summarizes this information. 4.4 COSTS OF CUSTOMER OUTAGES The costs to consumers of unexpected outages has been studied in great depth using numerous different methods [3]. Consumers suffer damages from outages through lost production, equipment damage, increased labor costs, lost leisure time, and in several other ways. If service reliability is poor enough, they may be prompted to buy emergency backup sources like batteries or generators, or in drastic cases, to leave the utility’s service area. lPxisting line has 70 MW capacity. The NE intertie would have 150 MW capacity. Outages on existing line would be fully backed up. Outages on NE line would be backed up only when transfers are at 70 MW or below. RI776A 4-13 INTERIM REPORT Decision Focus Incorporated Table 4-4 SUMMARY OF UNSERVED ENERGY REDUCTION BY INTERTIE Northeast Upgraded Intertie New Kenai- Anchorage- (Anchorage- Area Anchorage Fairbanks Fairbanks) Kenai 20% of all outages saved (except SES’s 0 0 outages) Anchorage 0 10% of MEA’s 25% of MEA’s outages saved. outages saved. Fairbanks 0 0 85% of all outages saved. Copper Valley 0 0 100% of all outages saved. 4.4.1 Basic Elements of Outage Costs There are two ways to determine customer outage costs: one can try to measure the actual losses that a customer sustained during an outage, or one can measure how much a customer would be willing to pay to avoid an outage. Since people do not always act rationally, the two are not always the same. We use studies from Sanghvi [4] and Ontario Hydro [5] which utilize the two different methods. Because outages at different times, of different lengths, to different people, incur different costs, it is most accurate if costs are modeled as a function of the most important attributes. According to the literature [4,5], the two most important attributes are duration and type of customer affected. The duration or how long an outage lasts is important because as duration increases, total costs increase. The longer the outage, the more expensive it is. Typically, the customers prefer a single long outage to several short outages totalling the same amount of time. As a result, outage costs per unit of energy go down with time.” "In some cases this does not hold, for example, a grocery’s freezer may maintain food for four hours, but beyond this, it starts to spoil. RIT76A 4-14 INTERIM REPORT Decision Focus Incorporated Outage costs also vary significantly by customer type. Expensive machinery may be damaged by an outage for large industrial customers, or a retailer may see his/her shop emptied when the lights go out, but residential customers might just miss their favorite television show. Usually studies group the customers into three groups: residential, commercial, and industrial. However, the Ontario Hydro study has shown that there are significant differences within the industrial and commercial customer classes and so those two groups are further subdivided. In the next subsection, we outline costs by each of these customer types and delve in more detail into the differences between these groups. 4.4.2 Customer Outage Costs Because an in-depth study was not possible within the time constraints of this study, we have applied the results of past studies in this analysis. We focused on the Ontario Hydro work because it is most applicable to the Railbelt, and we used other studies [3,4] as needed. Residential Customers. During an outage, a household’s preferred consumption pattern is disrupted. Some activities must be postponed until the power resumes, while others like watching television, may be lost. More significantly, residential heating systems may be interrupted during cold weather. It is clear that the residential customers get at least as much value from the electricity they purchase as they pay for it, otherwise they would not purchase it. However, measuring the consumer surplus above and beyond the purchase price is difficult. Some studies have estimated the cost of unserved energy as the value of lost leisure, since it is what the household member could have earned had they chosen to go to work. Other studies have utilized the concept of consumer surplus and created a consumer demand function to compute it. After doing an exhaustive survey of the available studies, Sanghvi has estimated that the costs to consumers of inconvenience and lost leisure and their willingness to pay to avoid such interruptions is in the range of $0.07/kWh to $2.00/kWh (1987 dollars). Because of the impact of outages on residential heating systems in Alaska, we selected the upper boundary for use in this analysis, i.e., $2.00/kWh. Industrial Customers. During an outage, industrial customers suffer damages resulting from lost opportunity costs that are more easily measured and greater than the costs sustained by residential customers. Because industrial customers incur process restart costs, studies (Ontario Hydro and Sanghvi) have shown that they suffer very high initial outage costs. For longer outages, idle resource costs tend to dominate although some of the lost production may be made up by working longer hours or by utilizing excess capacity. Industrial customers also suffer damages to equipment and products (e.g., gears grinding to a halt) and from employee hazards associated with loss of power to machinery and insufficient lighting in the workplace. Because small RI776A 4-15 INTERIM REPORT Decision Focus Incorporated industrial customers typically have less emergency backup, they have slightly higher outage costs than large industrial customers. In the Railbelt, there are only a few large industrial customers (petroleum refineries); the remaining industrial load seems to consist of small industrial customers. We used costs from the Ontario Hydro study that range from $69.00/kWh (1980 dollars) for a one minute outage to $1.34/kWh (1980 dollars) for an outage of ten hours duration.” Commercial Customers. Commercial customers are the hardest to classify because the range of activities they cover is very broad. Therefore, commercial customers display a broad range of outage costs. For this reason, the Ontario Hydro study divided up commercial customers into three subgroups: 1. Office Buildings 2. Retail Trade and Service, i.e., establishments providing services to the general public and to other businesses including major chain and independent retailers. 3. Institutions, i.e., schools, medical facilities, municipal services, and so on.® Office buildings can face substantial problems from short-term power outages, and as a result they experience the greatest short-term outage costs.“ In addition, the cumulative lost labor costs are substantial in office buildings. Outage costs in this group range from $8.57/kWh to $195.00/kWh (1980 dollars). Retail trade and service customers have fairly low short-term outage costs; longer outages tend to prevent them from doing business. Because few retailers maintain standby generation, they have the greatest outage costs beyond four hours as they start to sustain product spoilage. Outage costs for this group range from $3.18/kWh to $23.40/kWh (1980 dollars). “although the average cost of unserved energy (in $/kWh) decreases as outage length increases, the total cost of the outage (in dollars) increases as outage length increases. 8T> obtain Railbelt distributions on commercial customers, we mapped these three customer types into the following ISER building types and used the ISER Load Forecast Report for distributions: (1) Office Buildings—ISER building types: large and small office buildings; (2) Retail Trade and Service—ISER building types: restaurant, large and small retail, grocery, lodging, car service, and warehouse; (3) Institutions—ISER building types: medical, school, college, assembly. ™Power outage in office buildings lead to problems such as disruption of computer systems, difficult evacuation without elevators, air conditioning and ventilation systems stopping which interrupts work in many offices where the windows can’t be opened. RITI6A 4-16 INTERIM REPORT Decision Focus Incorporated Institutions have the smallest overall outage costs. The only part of this group with high possible outage costs are medical facilities that generally have standby generation. Costs for this group range from $0.35/kWh to $2.20/kWh. The distribution for each of the classes in the Railbelt region is listed in Table 4-5. The costs at various durations for each of the customer types is shown in Table 4-6. Table 4-5 DISTRIBUTION OF CUSTOMER CLASSES IN THE RAILBELT Total Energy Customer (GWh ) Percent Residential 1245 a1) Large Industrial 179 530) Small Industrial tr re Commercial/Retail ves 25.9 Commercial/Office Bldgs 344 ala Hate) Commercial/Institutions 368 yh Total (Excluding Distribution Losses) = 2984 100 (Source: [6]) Table 4-6 SUMMARY OF CUSTOMER OUTAGE COSTS Total Average Comm/Ind Outage Outage Cost ($/kWh, 1980 $) Outage Cost, $/MWh Cost 1987 Outage Lg Ind Sm Ind Com/Bldg Com/Retl Com/Inst 1980 $ 1987 $ $/MW 1 sec 61.80 69.00 195.00 23.40 1.80 58781 80001 0 1 min 61.80 69.00 195.00 23.40 1.80 58781 80001 i333 5 min 15..97 18.68 a7 201 gi. 57, 0.92 15923 21671 1805 10 min 10.24 12.38 29235 6.71 0.80 10563 14376 2395 15 min 8.33 10.28 23521 6.09 0.77 8772 11939 2985 20 min eae 9.24 20.16 Sato O17 7887 10735 so75 1 hour 3.97 6.31 14333) Ticoe) 0g 6987 9509 9509 2 hours Bie Soa 13.02 8.33 1.06 7056 9604 19208 4 hours 2.26 ao OL iva. 9.05 Deas 7120 9691 38763 8 hours 1.66 4.03 10.14 12.28 Zee 8273 11259 90071 10 hours to4 sk S257 15623 2.2 9198 12518 125179 Fraction of Load 0.103 0.044 0.198 0.444 0.211 1 Outage cost for residential customers: $2/kWh or $2000/MWh (1987S) RIT76A 4-17 INTERIM REPORT Decision Focus Incorporated Knowing the total unserved energy, customer type distribution, and duration for each outage, we can compute outage costs by multiplying the total unserved energy by the cost per unit of energy to obtain a total consumer cost. Because the outage costs will vary by customer type and duration of outage, we modify the costs accordingly. The equation to compute customer outage costs is as follows: Customer Outage Cost ($) = Unserved Energy (MWh) * Cost of Unserved Energy (Customer Type, Duration) ($/MWh) where cost of unserved energy (customer type, duration) is taken from Table 4-6. 4.5 VALUE OF IMPROVED SYSTEM RELIABILITY This subsection uses the potential changes in customer outages (Section 4.3) and the cost of customer outages (Section 4.4) to determine the value of improved system reliability. The value of improved system reliability is the lesser of customer outage costs achieved through the interties and the cost of increased spinning reserves to achieve a similar level of customer outage costs. For example, if it is cheaper to attain the same level of reliability through increased spinning reserves, then the costs of increased spinning reserves is the true value of increased system reliability. The following section addresses only the estimates of customer outage costs. The cost of increased spinning reserves to match the reliability impacts of the intertie proposals will be addressed in the final report. The estimates of reliability benefit described below may be reduced depending on the outcome of the spinning reserve analysis. 4.5.1 Results of Analysis We applied the potential improvements in system reliability (Table 4-4) and determined unserved energy savings. Tables 4-7 and 4-8 summarize the results for each area and each intertie. Table 4-9 shows the present value of reliability benefit for each intertie proposal assuming the identified annual benefits were maintained throughout a 35-year economic life. RI776A 4-18 INTERIM REPORT Decision Focus Incorporated UNSERVED ENERGY SAVED BY THE INTERTIES Kenai Anchorage Fairbanks Copper Valley A-F Northeast VALUE OF UNSERVED ENERGY SAVED BY THE INTERTIES Kenai Anchorage Fairbanks Copper Valley A-F Table 4-7 (MWh/yr) INTERTIE New Upgraded K-A A-F 21.3 0.0 0.0 20.9 0.0 0.0 0.0 0.0 21,3 20.9 Table 4-8 (M$/yr) INTERTIE New Upgraded K-A A-F 0.138 0.000 0.000 0.078 0.000 0.000 0.000 0.000 0.138 0.078 Table 4-9 PRESENT VALUE OF RELIABILITY BENEFIT (4.5% real discount rate) Kenai-Anchorage New Intertie Anchorage-Fairbanks Intertie Upgrade Anchorage-Fairbanks Northeast Intertie R1776A 4-19 $1987 Millions OrN Widow INTERIM REPORT Decision Focus Incorporated 4.5.2 Sensitivity Analysis There is more uncertainty surrounding the reliability analysis for the new Kenai-Anchorage intertie than for either of the Anchorage-Fairbanks proposals. As a result, two sensitivity cases were examined to further test the base case results. 4.5.3 Sensitivity Case #1 The purpose of the first sensitivity case was to estimate the impact of substantially different assumptions on the reliability benefits of the new Kenai- Anchorage line. Two changes were made to the base case assumptions: i In the base case, it was assumed that operation of Bradley Lake would result in an 80 percent reduction in Kenai Peninsula outages caused by failure of the existing Kenai-Anchorage line. The new intertie produced an additional 20 percent reduction. In this sensitivity case, we assume that Bradley Lake causes a 40 percent reduction in those same Kenai Peninsula outages, and the new intertie reduces them by the remaining 60 percent. In other words, the reliability impact of the new intertie on the Kenai Peninsula is tripled relative to the base case. 2: In the base case, it was assumed that outages in Anchorage due to export from the Kenai Peninsula and insufficient spinning reserve would be rare, and would be offset by occasional outages in Anchorage caused by the new line. In this sensitivity case, we assume that Anchorage outages caused by Kenai Peninsula exports would be significant in the absence of the new line due to insufficient spinning reserves. Specifically, we assume two outages per year, each resulting in the loss of 30 MW of load for one hour. It is further assumed that these two outages per year would not occur if the new line were built. Table 4-10 shows the result of this sensitivity case. The present value of reliability benefits in this case (35 years, 4.5 percent real discount rate) is $13.7 million. RI776A 4-20 INTERIM REPORT Decision Focus Incorporated Table 4-10 AMOUNT AND VALUE OF UNSERVED ENERGY SAVED BY NEW KENAI-ANCHORAGE LINE: SENSITIVITY CASE #1 Amount Value (MWh/yr) ($1987 million/yr) Kenai 63.9 0.414 Anchorage 60.0 03373 Fairbanks 0.0 0.000 Copper Valley 0.0 0.000 TOTAL 123.9 0.787 4.5.4 Sensitivity Case #2 In both the base case and the first sensitivity case, it was assumed that outage hours were allocated to residential and commercial customers in proportion to their respective fractions of system demand. This resulted in a 40 percent allocation to residential and 60 percent to commercial. Yet the value of unserved energy is much higher for commercial than for residential customers, and it is an accepted utility practice to allocate limited outages to residential customers first to the extent possible. In this sensitivity case, we carry forward all of the assumptions from sensitivity case #1, except we assume that the 30 MW outages in Anchorage are allocated 80 percent to residential customers and 20 percent to commercial customers. Table 4-11 shows the results. The present value of reliability benefits in this case (again for 35 years, 4.5 percent real discount rate) is $11.0 million. Table 4-11 AMOUNT AND VALUE OF UNSERVED ENERGY SAVED BY NEW KENAI-ANCHORAGE INTERTIE: SENSITIVITY CASE #2 Amount Value (MWh/yr) ($1987 million/yr) Kenai 63.9 0.414 Anchorage 60.0 0.214 Fairbanks 0.0 0.000 Copper Valley 0.0 0.000 TOTAL L239 0.628 R1776A 4-21 INTERIM REPORT 4.6 Decision Focus Incorporated USEFUL READINGS W. B. Shew, "How to Assess the Value of Electricity Reliability," The Value of Service Reliability to Consumers, EPRI, EA-4494, May 1986. E. Mosbak, “Shortage Costs: Results of Empirical Studies," The Value of Service Reliability to Consumers, EPRI, EA-4494, May 1986. A. P. Sanghvi, “Optimal Electricity Supply Realiability Using Customer Shortage Costs," 4.7 (1) [2] [3] [4] [5] [6] R1776A, The Value of Service Reliability to Consumers, EPRI, EA-4494, May 1986. REFERENCES Alaska Electric Power Statistics, 1960-1986, Alaska Power Authority, November 1987. Summary Supplement on Railbelt Utilities, Alaska Power Authority, November 1987. F. J. Alessio, P. Lewin, and S. G. Parsons, "The Layman’s Guide to the Value of Electric Power Reliability," The Value of Service Reliability to Consumers, EPRI, EA-4494, May 1986. A. P. Sanghvi, "Economic Costs of Electricity Supply Interruptions: U.S. and Foreign Experience," The Value of Service Reliability to Consumers, EPRI, EA- 4494, May 1986. L. V. Skof, "Ontario Hydro Surveys on Power System Reliability: Summary of Customer Viewpoints," presented at EPRI Seminar on the Value of Service Reliability to Customers, St. Louis, October 1983. Forecast of Electricity Demand in the Alaska Railbelt Region: 1988-2010, Institute of Social and Economic Research, University of Alaska, Anchorage, November 1988, Draft. 4-22 INTERIM REPORT Decision Focus Incorporated Section 5 ANALYSIS OF COGENERATION POTENTIAL 5.1 OVERVIEW Cogeneration is the simultaneous or sequential production of both thermal (typically hot water or steam) and electric energy. The efficiency of cogeneration systems, generally within a range of 67 to 92 percent, exceeds the efficiency of conventional systems that provide electricity or thermal energy separately. Cogeneration equipment is available in a wide range of sizes. Large systems, similar in size to conventional power plants are suitable for industrial applications; smaller systems, as small as 10 kW, are suitable for commercial applications. The technical and market potential of cogeneration depends on several factors, primarily: ° Fuel/Electricity Cost Differential: Cogeneration systems are most attractive where there is a large difference between the retail price of electricity and the price of fuel used in a cogeneration system. ° Coincidence of Thermal and Electric Loads: The value of cogeneration is significantly enhanced when the demands for thermal energy and electric energy coincide. Coincidence can be created by selling excess electricity to the utility or storing excess thermal energy. ° High Thermal Load Factor: Cogeneration is most attractive when the demand for thermal energy is relatively constant on a year- round basis. The first factor appears to be characteristic of the Railbelt. The second and third factors are not as commonly found and depend on the nature of the business. The objective of this analysis was to forecast installations of cogeneration equipment by the commercial and industrial sectors of the Railbelt during the period RI776A 5-1 INTERIM REPORT Decision Focus Incorporated 1990 to 2010. We collected and developed information on energy needs in each sector to determine the total technical potential for cogeneration, that is, the potential without regard to economics. We then determined the market potential, taking into account the uncertainties in fuel price and electric price growth, capital cost, and customer discount rates. The remainder of this section has the following organization. First, we describe how we characterized the potential cogeneration market in the Railbelt. Second, we present our characterization of the potential cogeneration technologies. Third, we discuss the market potential analysis approach. Finally, we present the results of the analysis. 5.2 CHARACTERIZATION OF COGENERATION TECHNICAL POTENTIAL The cogeneration technical potential is the total capacity that could be installed regardless of its economics. We performed some preliminary economic analyses to determine how electric customers in the commercial and industrial sectors would size their cogeneration systems. We found that customers would typically choose a system size to meet their peak electric demand or less. Thus, the cogeneration technical potential is equivalent to the peak demand for electricity. The subsections below summarize how we characterized the technical potential within each component of the commercial sector and the industrial sector. Appendix B describes the detailed characterization of the cogeneration technical potential. 5.2.1 Commercial Sector We characterized the types of customers in the commercial sector according to the fifteen building types shown in Table 5-1. Our main source of information was an ISER survey of commercial buildings in the Anchorage, Fairbanks, Kenai, and Matsu regions [1]. The survey covered buildings representing approximately six percent of total Railbelt commercial sector electricity consumption. We excluded the miscellaneous, vacant, and assembly building types, for two reasons: First, the Anchorage International Airport (AIA), which is treated separately, makes up a major portion of these groups. Second, the buildings in these groups have diverse electric and thermal requirements making an average somewhat meaningless. The survey contained information on building type, annual electricity consumption, sources of space and water heating (electric, gas, or oil), floor area, and the fraction of the floor area served by these sources. Appendix C includes the detailed customer-specific information provided by the survey. Since only information on annual RIT76A 5-2 INTERIM REPORT Decision Focus Incorporated electricity consumption (kWh per year) is available, but not demand (kW), we derived the peak demand to estimate the technical potential in the commercial sector. Table 5-1 COMMERCIAL SECTOR BUILDING TYPES 1. Small Office 6. Grocery 11. School 2. Large Office 7. Warehouse 12. College 3. Restaurant 8. Car Service 13. Assembly 4. Large Retail 9. Lodging 14. Miscellaneous 5. Small Retail 10. Medical 15. Vacant We also calculated the demand for space and water heating in each building type to determine thermal energy requirements.’ We refer to the demand for space and water heating as "thermal" demand in this section. We used a three-step approach to characterize the technical potential. First, we calculated the total annual electricity and thermal demand for each surveyed customer in each building type. We assumed that the average electric and thermal demand per square foot for each building type within the survey sample was representative of all commercial buildings within that type. Second, we developed annual load duration curves to represent the different levels of thermal and electricity demand over the year for each building type. This allowed us to determine the peak electric and thermal demand for each average customer.? Third, we estimated the total peak electric demand for the building type based on the peak electric demand for each average customer. The sum of the peak electric demands for each building type was the technical potential. We sorted the building types according to their electric demand during the winter day period (called peak demand), and we identified four groups. The building types in the first group have peak demand less than 50 kW and consist of the car service, small office, warehouse, restaurant, and small retail building types. The building types in the second group have peak demands between 50 and 200 kW and We considered space cooling as a possible end use to be served by cogeneration, but cooling in the Railbelt is done primarily with compression chilling and outside air. Compression chilling usually precludes a switch to absorption chillers due to the fairly extensive change in equipment that would be involved. "The size of an "average customer” for each building type in the total market was assumed to be equal to the size of the average customer for each building type in the sample. RIT76A 5-3 INTERIM REPORT Decision Focus Incorporated include the medical, grocery, and large retail types. The third group consists solely of schools due to their unique load duration curve with a peak demand around 300 kW. The building types in the fourth group have peak demands greater than 300 kW and include the lodging, college, and large office types. We included a fifth group in the commercial sector: hospitals. Although we did not have information on hospitals from the ISER survey, we had monthly electricity billing histories for the two hospitals in Anchorage [2]. Information was not publicly available for the hospital in Fairbanks. Since information on hospital thermal demand was not available, we used a ratio of electric to thermal demand of 0.2 MWh/MBtu to derive the thermal demand.® The total technical potential was estimated at 211.3 MW. Table 5-2 summarizes the cogeneration technical potential by group and region for the commercial sector. Table 5-2 COMMERCIAL SECTOR COGENERATION TECHNICAL POTENTIAL Group Anchorage Fairbanks Kenai Railbelt al 62.6 E57 10-3 88.6 2 18.9 5.3 2.0 26.2 3 39 BES 3.5 20.7 4 63.1 3.9 2.2 69.2 5 4.4 2.2 0.0 6.6 Total V6259 30.3 18.1 2a 5.2.2 Industrial Sector Information on historical electric demand was available for many industrial facilities, which formed the basis for calculating their cogeneration technical potential. We obtained data on peak electric demand and consumption by industry for almost all the companies in each industry [3]. The major industries in the Railbelt are petroleum processing and transportation, manufacturing, fish processing, construction, and mining. Table 5-3 shows peak electricity demand and consumption for each industry for the four Railbelt Personal communication with Science Applications International Corporation indicated a ratio of about 0.25 MWh/MBtu for the lower 48 states. We assumed that thermal requirements would be slightly higher in Alaska, hence the 0.20 MWh/MBtu. RI776A 5-4 INTERIM REPORT Decision Focus Incorporated regions. We aggregated this information over the regions and sorted the industries by decreasing peak electric demand as shown in Table 5-4. Table 5-3 INDUSTRIAL SECTOR ELECTRICITY DEMAND Demand Energy Region Industry (MW) (GWh) Anchorage Manufacturing 14 32 Fairbanks Petroleum Processing 7d 50 Mining 3 10 Petroleum Transportation a 4 Construction nt a Kenai Petroleum Processing ZA. 129 Manufacturing 7 20 Fish Processing 6 9 Matsu Construction 2 1 Totals 62 256 Table 5-4 ELECTRICITY DEMAND BY TYPE OF INDUSTRY Industry Petroleum Processing Manufacturing Fish Processing Construction Mining Petroleum Transportation Totals Demand Energy (MW) (GWh) 28 179 21 52 6 9 = 2 2 10 2 4 62 256 Petroleum processing is the largest industry, with a demand of 28 MW. This industry has substantial thermal demands and several refineries have or are considering building cogeneration systems. Because Tesoro has recently installed a 4 MW cogeneration unit, we subtracted 4 MW from the technical potential for a total of 24 MW. RI776A 5-5 INTERIM REPORT Decision Focus Incorporated Although manufacturing is the second largest industry in the Railbelt in terms of electric demand, we excluded it because of limited information and the assumption that their thermal requirements would be relatively small. The fish processing industry is the third largest in terms of electric demand. This industry has substantial thermal demand during its summer processing season. Its technical potential is 6 MW. We excluded the construction and petroleum transportation industries because they are not major electric consumers in the Railbelt, and probably have little or no process heat needs. The mining industry consists of the Usibelli Coal Mine; its demand shown in Table 5-4 includes only its present, utility-supplied consumption. A major cogeneration proposal now under consideration by Usibelli to supply thermal energy to a coal-drying process will be examined further during the next two months as part of this effort. Based on these considerations, we define the technical potential in the industrial sector as the sum of the technical potential in the petroleum processing and fish processing industries, for a total of 30 MW (excluding the 3.0 MW of mining). 5.3. COST AND PERFORMANCE OF POTENTIAL COGENERATION TECHNOLOGIES Because of the small-scale potential cogeneration applications in the Railbelt, we focused on small (less than 5 MW) cogeneration technologies. Small cogeneration technologies typically use an internal combustion engine (ICE) or a combustion turbine as the prime mover. Appendix D describes the two technologies. The choice and size of prime mover can be a function of many factors. In this analysis, we focused on peak electric demand and the ratio of electric to thermal demand as the main determinants of the cogeneration system. We chose cogeneration systems that best matched the thermal and electric needs of the average customer in each group. Table 5-5 shows the cogeneration unit cost and performance information used for the commercial and industrial sector analyses [4] and [5]. RIT76A 5-6 INTERIM REPORT Decision Focus Incorporated Table 5-5 COST AND PERFORMANCE OF SELECTED COGENERATION UNITS (1987 dollars) Fixed Capital Initial O&M Heat Power to Size* Cost Investment Cost Rate Heat Ratio Group (kW) ($/kW) ($M) ($/kW/yr) (Btu/kWh) (MW/MBtu/hr) 1 50 1800 0 90 12,000 0.20 2 200 1100 0.06 70 11,000 0.45 3 300 800 0.12 60 11,000 0.20 4 500 550 0.175 45 11,000 0.40 5 2500 565 o.3 35 10,000 0.50 Petro- leum 4300 1072 1.027 65 12,000 0.09 Fish 200 1100 0.06 70 11,000 0.50 *All are internal combustion engines, except the 4300 kW system, which is a combustion turbine with heat recovery. 5.4 COGENERATION MARKET POTENTIAL The cogeneration market potential is the portion of the technical potential that is economic from the viewpoint of the investor. We analyzed the market potential under conditions of uncertainty regarding certain major factors in the economics of cogeneration. We represented the uncertainty by combining uncertain scenarios for these factors into scenario combinations, each associated with a likelihood, or probability, of occurrence. In this way, we calculated the market potential for each scenario combination, weighted it by its associated probability, and then summed these values to achieve an estimated market potential. We used the cogeneration computer model, COGENOPT [6], to calculate the economics of cogeneration. COGENOPT takes into account all of the factors important for the economics of cogeneration, including cogeneration system cost and performance, the potential noncoincidence of electric and thermal demand, electric retail rates, utility power purchase (or buyback) rates, the value of the thermal output, and the customer’s discount (or hurdle) rate for the investment. We used the net present value (NPV) of the cogeneration investment over a 20-year period (1990 to 2010) as the criterion for the economics. COGENOPT discounted the annual cash flows using the customer’s hurdle rate. If the NPV was greater than or equal to zero, then the investment was considered economic from the investor’s viewpoint. RIT76A 5-7 INTERIM REPORT Decision Focus Incorporated The market potential analysis took into account the uncertainty in the following three factors: (1) cogeneration system capital costs; (2) the real annual compound growth rate of electric rates, gas and oil prices, and the buyback rate; and (3) the customer’s nominal discount rate. We grouped the uncertainties in the energy prices together because these prices are tied together through gas and oil prices. We characterized the uncertainty in each factor with three scenarios: low, middle, and high. The uncertainty in capital costs reflects a wide range of possible outcomes. We used the values shown in Table 5-5 for the middle scenario, 75 percent of those values for the low scenario, and 125 percent of those values for the high scenario. For the uncertainty in the real annual compound growth rate in the various energy prices, we used data provided by ISER and the APA [7]. For electricity rates in the commercial sector, we used the low, middle, and high forecast of rates in each of the four Railbelt regions. In the industrial sector, we used Homer Electric’s wholesale rates plus a 25 percent margin to approximate the rates paid by industrial customers. We used Homer’s rates because all of the fish processing and most of the petroleum processing occurs in the Kenai region. Table 5-6 shows the 1987 rates, the rates in 2010, and the real annual compound growth rate for the low, middle, and high electric rate scenarios. Table 5-6 RAILBELT COMMERCIAL AND INDUSTRIAL ELECTRICITY PRICES (1987 cents/kWh) BASE LOW SCENARIO MIDDLE SCENARIO HIGH SCENARIO Growth Growth Growth 1987 2010 Rate 2010 Rate 2010 Rate Commercial Anchorage 6.5 6.3) -0.1% 7.0 0.3% 728) 0.8% Fairbanks 9.0 9.1 0.0% 9.9 0.4% 10.8 0.8% Kenai 8.6 9.6 0.5% 10.1 0.7% 10.6 0.9% Matsu Teil oe 0.7% 9.6 1.0% 10.2 Lee Industrial 4.0 Sa0 1.2% 5.8 1.6% 6-3) 2.0% Source: Memo from Steve Colt to Dick Emerman, 8/23/88. For gas prices, we used ENSTAR’s base price in 1989 in combination with their stated margins for small commercial and large commercial (industrial) customers. The base price in 1987 was estimated at $3.03/MBtu for the commercial sector, and $2.611/MBtu for the industrial sector. The margins for the commercial and industrial RI776A 5-8 INTERIM REPORT Decision Focus Incorporated sectors were $1.61/MBtu and $1.15/MBtu, respectively [8]. These prices applied to all regions. Table 5-7 shows the estimate of 1987 prices, the prices in 2010, and the real annual compound growth rate for the low, middle, and high scenarios. For oil prices, we used the price forecasts developed for this study [9]. This price was necessary because we analyzed the economic potential in Fairbanks for two cases: gas available and gas not available (oil only). Table 5-8 shows the 1987 price, the prices in 2010, and the real annual compound growth rate for the low, middle, and high scenarios. Table 5-7 GAS PRICES (1987 $/MBtu) BASE LOW SCENARIO MIDDLE SCENARIO HIGH SCENARIO Growth Growth Growth 1987 2010 Rate 2010 Rate 2010 Rate Commercial 3.030 3.174 0.2% 3.895 1.1% 4.616 1.8% Industrial 21612 2.714 0.2% 3.435 1.2% 4.156 2.0% Source: ENSTAR forecast - Letter to Dick Emerman, 7/19/88 (prices include Margin on sales to small commercial customers of $1.61/MBtu, and to industrial customers of $1.15/MBtu) Table 5-8 OIL PRICES (1987 $/MBtu) BASE LOW SCENARIO MIDDLE SCENARIO HIGH SCENARIO Growth Growth Growth 1987 2010 Rate 2010 Rate 2010 Rate Fairbanks 4.930 5.140 0.2% 6.880 2558 8.550 2.4% Source: "OIL2" used in Over/Under model for Fairbanks For the buyback rate, we used the retail electricity rate assumptions for each customer group. Because buyback rates are unknown, sensitivity analysis was performed and indicated that most of the savings due to cogeneration would be due to avoided electricity rates, not sales of electricity; a reduction in the buyback rate of 50 RIT76A 5-9 INTERIM REPORT Decision Focus Incorporated percent reduced the economic potential by only 2 MW. (Only a few customer groups in a few regions had any significant sales of electricity.) We also looked at three scenarios for the customer’s nominal hurdle rate. These scenarios reflect different attitudes toward the time value of money. A low hurdle rate causes future cash flows to be weighted more heavily than a high hurdle rate. Table 5-9 shows the three scenarios of hurdle rates that we used. Table 5-9 CUSTOMER NOMINAL HURDLE RATES Percent Low as Middle 30 High 50 Given the above information for each scenario, we developed probabilities for each scenario. For capital costs, we used the following probability distribution for the low, middle, and high scenarios: 10, 40, and 50 percent, respectively.“ For the uncertainty in the growth rates of energy prices for the low, middle, and high scenarios, we used 60, 30, and 10, respectively, consistent with APA directive followed throughout this study. For the hurdle rate, we used the following probability distribution for the low, middle and high scenarios: 20, 60, and 20 percent, respectively. Figure 5-1 illustrates these probability distributions with a probability tree—a typical way of showing the structure of an uncertainty analysis. This figure is a shorthand way of expressing the full combination of scenarios. For example, for each scenario of capital cost, there are three scenarios of growth in energy prices. For each scenario of energy prices, there are three scenarios of customer hurdle rate. Thus, there are 3 x 3 x 3, or 27, scenario combinations. Each scenario combination has a corresponding probability calculated by taking the product of the probabilities of each scenario. ‘Middle case cost estimates were based on lower 48 data. Consequently, we assumed that Alaska installed costs would likely be higher rather than lower. RIT76A 5-10 INTERIM REPORT Decision Focus Incorporated Growth in: Fuel Capital Cost, Electric Hurdle Costs Price, Buyback Rate Rate Low Low Low 0.10 0.60 0.20 0.40 Middle 0.30 Middle 0.60 Middle Figure 5-1. Probability Tree We calculated the NPV for each of the 27 scenarios for each of the average customers in each commercial and industrial sector group in each region of the Railbelt. If the NPV was greater than zero, then we multiplied the probability of that scenario times the cogeneration unit size selected by the model. We summed these values over the 27 scenarios for each case. This was the fraction of the technical potential which was economic from the investor’s viewpoint; i.e., the market potential. 5.5 OTHER POTENTIAL RAILBELT COGENERATION A number of specific cogeneration possibilities were brought to our attention during this review and are discussed in greater detail in Appendix E. Of these possibilities, only one has thus far been identified as a likely development: 1.05 MW at Anchorage International Airport. 5.6 RESULTS AND CONCLUSIONS The market potential was 22 MW if Fairbanks has gas supplies, and 20 MW if it does not.© The results of the market potential calculations for each group in the commercial and industrial sectors are shown in Table 5-10 for the case where gas is not available in Fairbanks. Table 5-11 illustrates the detailed results by region. Around one-half of the estimated market potential is in Anchorage (commercial cogeneration), one-third is in Kenai (mostly industrial cogeneration), and less than one- fourth is in Fairbanks (mostly commercial cogeneration). Commercial cogeneration 5These results do not include the potential cogeneration projects described in Appendix E. RI776A 5-11 INTERIM REPORT Decision Focus Incorporated contributed two-thirds to three-fourths of the estimated market potential. Schools, hospitals, lodging, colleges, and large offices contributed most of the potential in the commercial sector; petroleum refining was dominant in the industrial potential. Table 5-10 MARKET POTENTIAL RESULTS* (Assuming Gas Is Not Available in Fairbanks) Technical Market Potential Potential Market Commercial Sector (MW) Fraction Potential Group » 88.6 0.01 em 2 26.2 0.07 La? s 20a S. id. aie 4 69.2 0.06 3.9 5 6.6 0.68 4.5 Industrial Sector Petroleum Processing 2 Fish Processing 0 +26 0 -00 3 19.6 ane oo oo ON 24 ay ©) *Does not include the potential cogeneration projects described in Appendix E. Table 5-11 DETAILED MARKET POTENTIAL RESULTS BY REGION (MW) Anchorage Fairbanks Kenai Railbelt w/gas w/o gas w/gas w/o gas Commercial Sector Group 1 0.8 0.9 0.2 0.1 p aod 2 1.0 On 0.4 0.3 vapak LST 3 a 0.6 0.4 0.5: Z10) Zee 4 Sez 01515) 0.4 0-3 4.0 3.9 5 3126. 1.8 0.9 0.0 5.4 a5) Total 10.1 4.5 2e3 a2 16.0 134 Industrial Sector Petroleum Processing 0.0 0.06 0.03 6.18 6.2 6.2 Fish Processing 0.0 0.0 0.0 0.0 0.0 0.0 Total 0.0 0.06 0.03 6.18 6.2 6.2 Railbelt Total 210 31. 4.6 23) 7.4 2202 19.6 RIT76A 5-12 INTERIM REPORT Decision Focus Incorporated We also performed several sensitivity analyses to determine which of the uncertain factors were most important. Table 5-12 summarizes these results. The analyses showed that, for all sectors, the most important factor was the customer discount (or hurdle) rate assumption. With low hurdle rates, the market potential would be around 75 MW. With high hurdle rates, the market potential would be less than 1 MW. These are extreme cases, but they illustrate the importance of customer attitude toward the time value of money. Table 5-12 MARKET POTENTIAL SENSITIVITY ANALYSIS RESULTS Capital Energy Costs Growth Rates Hurdle Rate Base Low High Low High Low High Anchorage 10.1 Zid, 5.4 10.2 LON 39.0 0.4 Kenai 7.4 aa57 5.6 7.4 7.4 28.5 0.0 Fairbanks (with gas) 4.6 9.8 or ac3 5.6 13.4 05.2) Fairbanks (without gas) au 5.0 1.9 2.4 aad 7.6 0.1 Total w/gas in Fairbanks 2261 $5.2 223.9 21.9 23. 80.9 0.6 w/o gas in Fairbanks 19.8 50.4 12.6 20.0 19.6 Tica; 0.5 Capital cost uncertainty was also very important. With low capital costs, the market potential would be around 50 MW. With high capital costs, the market potential would be only 13 MW. The uncertainty in the growth in electricity and fuel prices was insignificant. This was due to the fact that gas and electricity prices move together in the Railbelt, negating the impacts they would have if they moved separately. We also examined the importance of the utility buyback rates (the analysis assumed they would be equal to retail rates). We found that most customers were not generating more electricity than they needed. We performed another sensitivity analysis in which we set the buyback rate to be one-half the retail rate. This reduced the market potential by only 2 MW (or 10 percent). In conclusion, cogeneration has some potential in the Railbelt during the study period. The cogeneration market potential is estimated at around 20 MW, but could be much smaller under unfavorable market conditions or much larger (three to four times larger) under favorable market conditions. RI776A 5-13 INTERIM REPORT 5.7 (1) [2] [3] [4] [5] [6] (7] [8] [9] RIT76A Decision Focus Incorporated REFERENCES Commercial End Use Survey, Institute for Social and Economic Research, 1988. Provided by Steve Colt, ISER, November, 1988. ISER industrial sector database, provided by Steve Colt, November 1988. Reference Guide to Small Cogeneration Systems for Utilities, RMR Associates, Electric Power Research Institute, EM-4371, final report, February 1986. Small Cogeneration System Costs and Performance, Science Applications International Corporation, Electric Power Research Institute, EM-5954, final report, August 1988. P.N. Estey and S. J. Jabbour, COGENOPT Users Manual, September, 1983. Memo to Railbelt Utilities from Richard Emerman, Alaska Power Authority, Subject: Retail electricity price inputs to Railbelt load forecasting models, August 23, 1988. Letter to Richard Emerman, Alaska Power Authority, from Bill Hickman, ENSTAR, July 19, 1988. Fuel Price Outlooks: Crude Oil, Natural Gas, and Fuel Oil, report to the Alaska Power Authority, ICF Incorporated, August 1988. 5-14 INTERIM REPORT Decision Focus Incorporated Section 6 INPUT DATA AND MODELING ASSUMPTIONS 6.1 OBJECTIVE/OVERVIEW This section describes the data we collected from the Railbelt utilities, APA, and APA contractors. The data consisted of system load information, fuel price forecasts, existing generating plant information, and details of existing and potential new/upgraded _interties. These data are primarily required for use in the comprehensive Over/Under production simulation model to calculate economy energy transfers and transmission efficiency losses, and will also be used for the calculation of operating and planning reserve sharing benefits. For modeling purposes, we divided the Railbelt region into four principal load and supply centers: the Kenai Peninsula, the Anchorage-Matsu Valley area (hereinafter called the Anchorage area), the Fairbanks area, and the Copper Valley area. 6.2 SYSTEM LOADS We collected the system load information for each of the four areas in the Railbelt. The data included peak load growth forecasts and load duration curves. 6.2.1 Peak Load Growth Forecasts The peak load growth forecasts were provided by the Institute for Social and Economic Research at the University of Alaska for three equally probable cases: high, middle, and low growth scenarios. The assumptions and methodology used in the analysis are documented in [1] and [2]. Table 6-1 summarizes the load growth rates. 6.2.2 Load Duration Curves Using the 1987 hourly load data provided by the utilities [3], [4], [5], [6], [7], [8], {9}, [10], [11], [12], we developed load duration curves for all utilities in the Railbelt. We divided the year into a peak and off-peak season. The peak demand season (winter RIT76A 6-1 INTERIM REPORT Decision Focus Incorporated months) include November through February. The off-peak demand season includes the remaining eight months of the year: March through October. The utilities were then aggregated into the four areas based on their service areas. Table 6-2 lists the service areas of all eight utilities. Table 6-1 FORECAST OF LOAD GROWTH RATES Peak Demand Copper Case Year Anchorage Kenai Fairbanks Valley LOW 1994 383 65 119 10.6 2010 422 68 135 12 Growth/Yr 0.0057 0.0027 0.0074 0.0073 MID 1994 386 76 118 14.3 2010 494 88 143 16.4 Growth/Yr 0.0146 0.0087 0.0114 0.0081 HIGH 1994 404 80 125 19.5 2010 532 97 163 20 Growth/Yr 0.0163 0.0114 0.0157 0.0015 Table 6-2 SERVICE AREAS OF RAILBELT UTILITIES Area Utilities Kenai HEA, SES, CEA* Anchorage AMLP, MEA, CEA* Fairbanks GVEA, FMUS Copper Valley CVEA *CEA serves retail loads in both Anchorage and Kenai. Table 6-3 shows that CEA’s retail load is mainly in Anchorage (96 percent); only 4 percent is in Kenai. Using the energy breakdown within each area (see Table 6-4), RIT76A 6-2 INTERIM REPORT Decision Focus Incorporated we constructed the area load duration curves as the energy weighted sum of each utility’s individual load duration curve for all utilities within each area.’ The peak and off-peak load duration curves for Kenai, Anchorage and Fairbanks are shown in Figures 6-1, 6-2, and 6-3. The load duration curve for the Copper Valley area was not completed at the time of this write-up. Table 6-3 ENERGY AND DEMAND BREAKDOWN FOR CEA Energy Peak Demand Area (GWh) % of Total (MW) Kenai 34.6 4 qd Anchorage 846.4 96 161 Source: [13] Table 6-4 ENERGY BREAKDOWN BY AREA AND BY UTILITY Energy Area Utility (GWh) % of Area Kenai HEA 414.5 85.5 SES 35.5 ies CEA 34.6 ase: Total 484.6 100.0 Anchorage CEA 846.5 40.8 AMLP 789.2 38.0 MEA 439107, 22) Total 2075.4 100.0 Fairbanks GVEA 409.8 75.0 FMUS 136.9 25.0 Total 546.7 100.0 Copper Valley CVEA 43.6 100.0 Sources: /[(3]/,, (77, ((12i. (22), (131 Because of the high coincidence factor among peaks in the Railbelt (over 97 percent), we added the utility loads within each area. This is referenced in Susitna FERC License Application, Nov. 1985 draft, Exhibit B. RIT76A 6-3 INTERIM REPORT Decision Focus Incorporated Fraction of Peak Demand 1987 Demand (MW) 80 oe Peak Season 7 60 0.6 Off-Peak Season J 40 04+ = 20 0.27 0 1 4 1 4 J 1 1 1 1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Fraction of Season Peak Season: Nov-Feb Figure 6-1. Load Duration Curves for the Kenai Peninsula Fraction of Peak Demand 1987 Demand (MW) 1 Peak Season 0.8 >= 300 0.6 Off-Peak Season 200 0.4 100 0.2 1 1 4 1 1 J 0 01 o2 O08 04 O85 06 O07 O8 089 1 Fraction of Season 1 Peak Season: Nov-Feb Figure 6-2. Load Duration Curves for Anchorage RIT76A 6-4 INTERIM REPORT Decision Focus Incorporated Fraction of Peak Demand 1987 Demand (MW) + 100 os Peak Season = 80 0.6 60 fo Off-Peak Season | + 40 0.2 + 20 0 L 1 1 1 1 1 1 1 1 0 0 (0:4) ||) (0.2) |0'3) ||) |'0'4)| ||| '0:5||)/:0:6)| ||10:7, ||| /o!a'| || 019 1 Fraction of Season Peak Season: Nov-Feb Figure 6-3. Load Duration Curves for Fairbanks 6.3. FUEL PRICE FORECASTS The four primary fuels used in the Railbelt are natural gas, coal, No. 4 fuel oil, and No. 2 fuel oil. Coal price forecasts for various minemouth and delivered locations were provided by the Institute for Social and Economic Research [14] and are tied to the cost of production. A single price forecast at each of these locations was determined. The prices stay constant in real terms. The fuel oil price forecasts were provided by ICF, Inc. [15]. They are tied to long-term crude oil price forecasts. Three price scenarios were developed: high, middle, and low, with respective probabilities of 10, 30, and 60 percent. The natural gas price forecasts were determined from recently negotiated contracts between Enstar and Marathon, and Chugach and Marathon. The Enstar gas prices include a Kenai-Anchorage delivery premium. The Chugach gas prices are used for the Beluga and Bernice Lake plants. The Enstar gas prices are used for the remaining Anchorage area plants.? The natural gas contract prices are adjusted by changes in the price of crude oil. All fuel forecasts for the three scenarios are shown in Table 6-5. ?The Soldotna plant requires an additional delivery charge. R1776A 6-5 INTERIM REPORT Decision Focus Incorporated Table 6-5 FORECAST OF FUEL PRICES (1987 $/MBtu) NATURAL GAS CHUGACH ENSTAR Year Low Mid High Low Mid High 1990 iat 1.43 1.86 1.43 1.60 1.69 1994 1.30 1.64 1.88 1.52 1.90 2.19 1995 Se 1.69 1.98 1.54 1.98 26S 2010 665) 2.41 Srl 1256, 2.28 SiO: FUEL OIL No. 4 No. 2 Year Low Mid High Low Mid High 1990 Zi100) 319 3.54 4513 4.85 Sn 24 1994 Zeid 3.58 4.18 4.42 5.26 5.78 1995 2.84 3.68 4.34 4.49 5.536 5.94 2010 3.54 Sis 6.75 5.14 6.88 8.54 COAL Type Price Healy Minemouth 30) Healy delivered to Fairbanks 2052 Healy delivered to Nenana 1.69 Healy Waste Coal 0.07 Matanuska Minemouth To15) Beluga Minemouth v5) Waste Coal (Healy and Beluga) 0.07 Note: There is no growth rate for coal prices in real terms. 6.4 POWER GENERATION PLANTS The initial generating plants database was compiled from the Railbelt Intertie Proposal, Preliminary Economic Assessment report [16]. We then surveyed the utilities for any changes in the following plant-specific information: RIT76A, 6-6 INTERIM REPORT Decision Focus Incorporated ° Rated capacity (MW). . Committed capacity additions and year of additions. ° Planned retirement date. ° Fixed operating and maintenance costs ($/kW-yr). . Forced outage rate (%), not including scheduled maintenance. ° Equivalent availability after maintenance and forced outages.* ° Fraction of maintenance scheduled during the peak season. ° Heat rate at full load (Btu/kWh). . Variable operating and maintenance costs ($/MWh). Appendix F summarizes the power plants data [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [16]. 6.4.1 Technology Groupings For the purposes of this study, it was necessary to convert the plant-specific information into technology groupings. To do this, we first computed the variable operating costs of each plant.‘ Then we sorted the plants by area, fuel type, and technology. We then grouped similar plants into technologies ensuring that each technology included plants in the same area, and with similar heat rates and identical fuel and plant types. We determined the technology characteristics by taking capacity weighted averages of the individual plants in that technology. The plant technology groupings and the technology characteristics are shown in Appendix F. 6.4.2 Hydro Plants The average energy by month for the four hydro power plants (Bradley Lake, Eklutna, Cooper Lake, and Solomon Gulch) is shown in Table 6-6. The four-month winter peak season has approximately 41 percent of the annual hydro energy. 6.5 TRANSMISSION SYSTEM Transmission voltages and efficiencies for each of the existing, new, and upgraded interties are listed in Table 6-7. Losses are shown both for the total transfer, and for that particular transfer increment. ’Equivalent Availability = (1 - Forced Outage Rate) x (1 - Annual Maintenance Rate). ‘Variable Operating Costs ($/MWh) = Fuel Costs ($/MWh) + Variable O&M ($/MWh). Fuel Costs ($/MWh) = Heat Rate (BtwkWh) x Fuel Cost ($/MBtu)/1000. RI776A, 6-7 INTERIM REPORT Decision Focus Incorporated Source: Anchorage Anchorage Anchorage Bradley Lake Beluga Anchorage RI776A (17) Fairbanks Soldotna Soldotna Soldotna Anchorage Fairbanks Existing line Upgrade Existing line New line New line Existing line Northeast intertie Table 6-6 ENERGY OF RAILBELT HYDRO POWER PLANTS (GWh/yr) Solomon Bradley Cooper Eklutna Gulch Lake Lake Copper Area Anchorage Valley Kenai Kenai Jan 14.0 2.9 49.7 6.0 Feb 12.0 3.2 43.6 4.0 Mar 12.0 3.2 31.2 3.0 Apr 10.0 2.2 16.4 4.0 May 12.0 3.6 14.6 1.5 Jun 12.0 4.0 oa 1.5 Jul 13.0 4.3 14.0 1.0 Aug 13.0 4.1 28.5 2.0 Sep 13.0 4.2 31.0 2.0 Oct 14.0 4.3 36.8 4.0 Nov 14.0 3.4 42.6 6.0 Dec 14.0 3.3 50.7 6.0 Total 153.0 42.5 366.4 41.0 Peak 54.0 1257 186.6 22.0 Off-Peak 99.0 29.8 179.8 19.0 Area (GWH/yr) (% on Peak) Kenai 407.4 51.2 Anchorage 153.0 35.3 Fairbanks 0.0 NA Copper Valley 42.5 29.9 Railbelt 560.4 49.19 Table 6-7 INTERTIE EFFICIENCIES Transfer Incremental Voltage Transfer Inc. Total Loss Loss (Kv) (MW) (MW) = (Frac) (aw) (MW) (Frac) 138 10 10 0.03 0.3 0.3 0.030 45 35 0.08 3.6 3.3 0.094 70 25 0.12 8.4 4.8 0.192 230 70 70 0.03 2.1 2.1 0.030 225 155 0.06 13.5 11.4 0.074 115 45 45 0.09 4.0 4.0 0.089 70 25 0.13 8.8 4.8 0.193 230 250 250 0.02 5.0 5.0 0.020 115 45 45 0.02 1.1 1.1 0.024 90 45 0.05 4.4 3.3 0.074 120 30 0.07 8.2 3.9 0.125 230 350 350 0.02 7.0 7.0 0.020 138 10 10 0.03 0.3 0.3 0.030 45 35 0.08 3.6 3.3 0.090 70 25 0.12 8.4 4.8 0.190 150 80 0.12 «18.0 9.6 0.120 6-8 INTERIM REPORT 6.6 {1 (2 [3 [4 [5 (6 (7 [8 [9 (10 {11 {12 (13 (14 (15 (16 {17 R1776A Decision Focus Incorporated REFERENCES Forecast of Electricity Demand in the Alaska Railbelt Region: 1988-2010, Institute for Social and Economic Research, University of Alaska, Anchorage, November 1988, Draft. NE Intertie Load Forecast, Institute for Social and Economic Research, University of Alaska, Anchorage, January 1989. Letter from Moe Aslam (AMLP) to Jennie Rice (DFI), 30 August 1988. Letter from Gerald Mackey (CEA) to Jennie Rice (DFI), 19 September 1988. Letter from Gerald Mackey (CEA) to Salim Jabbour (DFI), 13 October 1988. Letter from Gerald Mackey (CEA) to Mike Gordon (DFI), 12 December 1988. Letter from Lowell Highbargain (CVEA) to Jennie Rice (DFI), 26 August 1988. Phone conversation between Paul Diener (SES) and Salim Jabbour (DFI). Phone conversation between Kent Wick (HEA) and Salim Jabbour (DF). Letter from Adam Choi (FMUS) to Salim Jabbour (DFI), 23 September 1988. Covers loads. Phone conversation between Marty Lanum (FMUS) and Salim Jabbour (DFI). Letter from Steve Haagenson (GVEA) to Jennie Rice (DFI), 24 August 1988. 1987 Power Requirements Study, Chugach Electric Association, November 1987. Letter from Scott Goldsmith (ISER) to APA Coal File, 3 November 1988. Fuel Price Outlooks: Crude Oil, Natural Gas, and Fuel Oil, report to the Alaska Power Authority, ICF Incorporated, August 1988. Railbelt Intertie Proposal, Preliminary Economic Assessment, March 1987. Letter from Dick Emerman (APA) to Salim Jabbour (DF). 6-9 INTERIM REPORT R1776A Date 87/11/27 87/06/21 87/06/21 87/06/21 87/04/01 87/04/01 87/04/01 87/04/01 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/08/30 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 86/03/04 Decision Focus Incorporated Appendix A SUMMARY OF UTILITIES DATA ON CUSTOMER OUTAGES Table A-1 AMLP Duration (hours) NNNEFP PRP RPP EP BP BHORBRBPHHHHHOODOOODOOOOrHFO +42 ua? «ve - 82 -50 - 68 a -90 - 80 -87 -13 «k3 -03 Ze wee mee weil 13 -95 -20 -20 -48 87 -90 393 95 -98 -00 -02 ~12 Number of Customers 5553 3602 6857 8313 1981 3684 10693 5312, 5030 960 7721 5115 3072 607 3148 1807 4914 5638 855 946 396 1006 4095 1055 5397 5689 1299 4695 1981 2686 affected INTERIM REPORT Decision Focus Incorporated RI776A Date 87/01/23 87/03/18 87/04/01 87/06/06 87/06/21 87/07/26 87/08/16 87/08/22 87/11/12 87/12/23 86/03/04 86/05/26 86/07/10 86/08/30 Table A-2 MEA Duration (hours) 0. a A-2 25 -78 +33 +33 -92 +33 -28 -08 -13 -00 ate -58 42 -38 Number of Customers affected 350 17137 11000 436 10000 1000 15000 1259 1173 9000 23000 1800 1200 8445 INTERIM REPORT RIT776A Date 87/04/01 87/04/01 87/04/01 87/06/21 87/06/21 87/06/21 87/06/21 87/11/11 87/11/11 87/11/11 87/11/11 STA fA 87/11/14. 87/11/11 87/04/01 87/04/01 87/04/01 87/04/01 87/04/01 87/05/13 87/06/21 87/06/21 87/06/21 87/06/21 87/07/26 87/07/26 87/07/26 87/07/26 87/11/11 87/11/11 87/11/11 87/11/11 87/11/11 87/11/11 87/11/11 87/11/11 Table A-3 Duration (hours) RFPrRRrROOrFrFON NRFPRrRPRrRPOOCOSCS COCOCOCOFRFRN oOoCoCoOorRrRrFFH -00 oon -76 « 16 +70 -70 qo -70 «45 - 68 +63 33 -30 sok 85 133 -80 01 -021 +50 13 eS +90 -90 -08 705 41 421 203) -03 7O1 +96 +46 46 -20 -10 Decision Focus Incorporated Number of Customers affected 814 663 738 356 843 500 446 664 1040 705 593 705 593 664 898 3337, 613 1481 550 5 3337 1481 613 898 550 1481 30 566 550 1481 30 566 613 898 775 SSou INTERIM REPORT LuOdgu WIYSLNI 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 0022 poqoosse szewoysng FO zZequnN SZ SL° SZ° SL° SZ° SL° oooo;co SZ° SL° GZ" SL° SZ° ooococoo Si. SZ° SL° SZ" SL° oooco SZ° Si” SZ° SL° SZ" ooocoo SL° SZ° SL° SZ" SL oooo°o SZ° SL° SZ° SL° SZ° (sanoy) uotzeang ooooo v-W STIeL 20/21/98 T0/2T/98 8Z/TT/98 LZ/TT/98 60/TT/98 OT/0T/98 LT/90/98 0/20/98 ZT/T0/98 TT/T0/98 OT/T0/98 S0/T0/98 ¥0/T0/98 €0/1T0/98 €Z/ZT/LB 61/ZT/L8 b0/ZT/L8 Z0/ZT/L8 O€/TT/L8 8Z/TT/L8 bT/TT/L8 €T/TT/L8 97/b0/L8 ST/E0/L8 Z0/Z0/L8 T0/Z0/L8 GT/TO/L8 bT/TO/L8 €T/TO/L8 OT/TO/L8 60/T0/L8 a qed VOLLIE peyesodicouy snd0q uolspeq Decision Focus Incorporated RI776A Date 87/03/18 87/03/18 87/04/01 87/04/01 87/04/01 87/04/02 87/04/02 87/05/14 87/05/14 87/05/26 87/06/18 87/06/21 87/06/21 87/06/21 87/07/01 87/07/01 87/07/01 87/07/01 87/07/01 87/07/01 87/07/01 87/07/02 87/07/02 87/07/03 87/07/03 87/07/10 87/07/25 87/07/26 87/07/26 87/07/26 87/07/26 87/07/29 87/07/29 87/07/29 87/07/29 Table A-5 FMUS Duration (hours) -28 -28 -08 -08 ee oooo°o -48 -47 -33 -50 +05 Sooo 66 siz 732 +20 233 a2) ooocoo 0.12 0.12 0.08 0.10 0.12 0.12 0.00 -00 -00 -00 ooo -70 -o -18 -07 ok? oooo°o -58 -07 ae a05 -05 ooooo° Feeder Affected Lathrop Garden Island Lathrop Garden Island Eagan Avenue WoW tori. Eagan Avenue WoW. TP Eagan Avenue Garden Island Lathrop Lathrop Garden Island Eagan Avenue Barnette 4th Avenue Lathrop West Gate Garden Island W.W.T.P. Eagan Avenue W.W.T.P. Eagan Avenue WOWeT2P. Eagan Avenue West Gate West Gate Lathrop Cowles Garden Island WoWotuP. Lathrop Garden Island Lathrop Garden Island INTERIM REPORT Decision Focus Incorporated R1776A Date 87/08/07 87/08/16 87/08/16 87/08/22 87/08/22 87/08/22 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/10/29 87/11/12 87/11/12 87/11/12 87/11/12 87/11/12 87/11/12 87/11/12 87/12/04 87/12/04 87/12/04 87/12/04 87/12/10 87/12/10 87/12/10 87/12/10 87/12/10 87/12/10 87/12/10 87/12/23 Table A-5 (Continued) Duration (hours) -00 -17 -18 -67 -70 SSOCDO0D CODCOD COD OHPH BRE BHE BPHERHPOH ooo oOH oooco A-6 -53 -59 43 +43 +43 +43 43 43 43 - 43 - 43 -08 -08 -10 -12 .22 32 32 ol7 oe +05 17 -02 -02 -02 03 07 - 48 - 48 +03 Feeder Affected Eagan Avenue Lathrop Garden Island Lathrop Garden Island Eagan Avenue First Avenue Barnette 4th Avenue Lathrop First Avenue West Gate Cowles Garden Island W.W.T.P. Eagan Avenue 4th Avenue Lathrop West Gate Cowles Garden Island WoW. T.. Pi. Eagan Avenue Lathrop West Gate Cowles Garden Island 4th Avenue Lathrop West Gate Cowles Garden Island W.W.T.P. Eagan Avenue Barnette INTERIM REPORT Decision Focus Incorporated Table A-5 (Continued) FMUS Duration Feeder Date (hours) Affected 87/12/23 0.03 4th Avenue 87/12/23 0.03 Lathrop 87/12/23 0.05 West Gate 87/12/23 0.28 Garden Island 86/01/17 0.28 Lathrop 86/01/17 0323 Garden Island 86/01/17 0327 West Gate 86/02/26 OR03) 4th Avenue 86/03/04 (ey) Lathrop 86/03/04 0.18 Garden Island 86/05/14 0.05 W.W.T.P. 86/05/26 0.85 W.W.T.P. 86/06/05 0.02 W.W.T.P. 86/06/27 0.33 West Gat 86/07/06 1-05 Garden Island 86/08/08 0.50 Garden Island 86/11/27 13:50) Garden Island 86/12/08 OL 75) Lathrop 86/03/04 0.35 Garden Island 86/12/11 OLaS W.W.T.P. 86/12/11 0-25) Eagan Avenue NOTES: ‘lie Individual outage information was not available for CEA, GVEA, and CVEA. GVEA provided necessary information on _ outage distributions. 2 1986 individual outage information was not available for HEA. RIT76A A-7 INTERIM REPORT Decision Focus Incorporated Appendix B CHARACTERIZATION OF COGENERATION TECHNICAL POTENTIAL The cogeneration technical potential is the total capacity that could be installed regardless of its economics. We performed some preliminary economic analyses to determine how electric customers in the commercial and industrial sectors would size their cogeneration systems. We found that customers would typically choose the system size to meet their peak electric demand or less. Thus, the cogeneration technical potential is equivalent to the peak demand for electricity. This appendix describes how we characterized the technical potential within each component of the commercial sector and the industrial sector. B.1 COMMERCIAL SECTOR We characterized the types of customers in the commercial sector according to the fifteen building types shown in Table B-1. Our main source of information was an ISER survey of commercial buildings in the Anchorage, Fairbanks, Kenai, and Matsu regions [1]. The survey covered buildings representing approximately 6 percent of total Railbelt commercial sector electricity consumption. We excluded the Miscellaneous, Vacant, and Assembly building types, for two reasons: First, the Anchorage International Airport (AIA), which is treated separately, makes up a major portion of these groups. Second, the buildings in these groups have diverse electric and thermal requirements making an average somewhat meaningless. The survey contained information on building type, annual electricity consumption, sources of space and water heating (electric, gas, or oil), floor area, and the fraction of the floor area served by these sources. Since only information on annual electricity consumption (kWh per year) is available, but not demand (kW), we derived the peak demand to estimate the technical potential in the commercial sector. RI776A B-1 INTERIM REPORT Decision Focus Incorporated Table B-1 COMMERCIAL SECTOR BUILDING TYPES 1. Small Office 6. Grocery 11. School 2. Large Office 7. Warehouse 12. College 3. Restaurant 8. Car Service 13. Assembly 4. Large Retail 9. Lodging 14. Miscellaneous 5. Small Retail 10. Medical 15. Vacant We also calculated the demand for space and water heating in each building type to determine thermal energy requirements.’ We refer to the demand for space and water heating as "thermal" demand in this appendix. We used a three-step approach to characterize the technical potential. First, we calculated the total annual electricity and thermal demand for each surveyed customer in each building type. We assumed that the average electric and thermal demand per square foot for each building type within the survey sample was representative of all commercial buildings within that type. Second, we developed annual load duration curves to represent the different levels of thermal and electricity demand over the year for each building type. This allowed us to determine the peak electric and thermal demand for each average customer.” Third, we estimated the total peak electric demand for the building type based on the peak electric demand for each average customer. The sum of the peak electric demands for each building type was the technical potential. B.1.1 Step 1: Estimating Total and Average Electric and Thermal Demand Given the ISER survey data on electricity consumption (kWh/yr) and floor area by customer, we calculated the average electricity consumption and floor area for each building type. We then calculated the total electricity consumption for each building type, by multiplying the average electricity consumption by the total number of average customers’ in that building type. We considered space cooling as a possible end use to be served by cogeneration, but cooling in the Railbelt is done primarily with compression chilling and outside air. Compression chilling usually precludes a switch to absorption chillers due to the fairly extensive change in equipment that would be involved. "The size of an “average customer" for each building type in the total market was assumed to be equal to the size of the average customer for each building type in the sample. ‘The total number of average customers in a building type is the ratio of the total floor area in that building type to the average customer’s floor area. RIT76A B-2 INTERIM REPORT Decision Focus Incorporated Thermal requirements for space and water heating were based on the ISER end- use analysis. Typical values for the twelve building types are shown in Table B-2. Space heating requirements in Fairbanks are around 10 percent higher than in Anchorage, Kenai, and Matsu_ regions. Note that, although thermal requirements are expressed in terms of kWh per square feet per year, this should not be interpreted as though electricity were the sole energy source for these applications. In this context, kWh is used merely to represent a unit of energy and is readily convertible to Btu requirements. Table B-2 ENERGY REQUIREMENTS (kWh/sq ft/yr-Existing Buildings) Space Heating Building Type Water Anchorage/ No. Name Heating Kenai/Matsu Fairbanks 1 Small Office 1.0 9.0 9.9 2 Large Office 7.0 9.0 9.9 3 Restaurant 7.0 9). 5) 1055 4 Large Retail 0.8 6.3 6.9 5 Small Retail 0.8 6.3 6.9 6 Grocery La 14.4 1538) 7 Warehouse 0.8 5.4 519) 8 Car Service 1.0 7.4 Ta) 9 Lodging 320 9.0 9.9 10 Medical 2.8 15.0 16.8 ge: School 1.8 10.4 25) 12 College 1.8 9.0 9.9 Source: Forecast of Electricity Demand in the Alaska Railbelt Region: 1988-2010, Draft Report, 19 November, 1988, pp. 3-16, 3-17. We applied these values to the survey data regarding the fraction of the floor area served by the different energy sources (electric, oil, or gas) to determine the demand by energy source. For example, if a small office building with a floor area of 5,000 square feet had 80 percent gas-fired and 20 percent electric space heating and 100 percent electric water heating, its total thermal demand would be: (80%) x (5,000 sq ft) x (9.0 kWh/sq ft/yr) {Gas-Fired} + (20%) x (5,000 sq ft) x (9.0 kWh/sq ft/yr) {Electric} + (100%) x (5,000 sq ft) x (1.0 kWh/sq ft/yr) {Electric} 36,000 (gas-fired) + 9,000 (electric) + 5,000 (electric) kWh/yr 50,000 kWh/yr, or 171 MBtuw/yr.* “To avoid double counting, we would subtract the 14,000 kWh/yr (electric) from the annual electric demand. RI776A B-3 INTERIM REPORT Decision Focus Incorporated We performed this type of calculation for each surveyed customer in each building type and then calculated the average thermal demand by building type. We determined the total thermal demand for each building type in the same way as for the total electric demand: we multiplied the average thermal demand by the total number of average customers in that building type. Table B-3 shows the results of these calculations by building type aggregated over the four regions. The table shows the average and total floor area, electric demand, thermal demand, and the percentage of total floor area, electric demand, and thermal demand for each building type. (Appendix D contains the detailed results by customer and by region.) B.1.2 Step 2: Developing the Load Duration Curves A load duration curve reflects the different levels of demand that occur over a period of time. We used a load duration curve with four periods during the course of a year: winter day, winter night, summer day, and summer night. We defined the peak season (called "winter") to be consistent with other assumptions in this study (refer to Section 6): November through February. The off-peak season (called "summer") is therefore March through October. The number of hours in the winter season is approximately one-third (four out of twelve months) the hours in the year (8760), or 3000 hours. The summer season then has 5760 hours. We divided each season into two periods: a peak period (called day) and an off- peak period (called night). Typically, lower levels of demand are found at night. For all building types except schools, we used 12 hours for the day and 12 hours for the night during a 24 hour cycle.® This 50/50 split between day and night hours meant there were 1500 hours in the winter day and winter night periods and 2880 hours in the summer day and summer night periods. For schools, we used the same definition of day and night during the winter season, but during the summer season we took into account the fact that most schools have lower energy requirements compared to the rest of the year. We included two of the eight summer months plus 50 percent of the hours in the other six months in the night period to represent the duration of lower energy requirements. This resulted in 2160 hours for the summer day period and 3600 hours for the summer night period. Table B-4 summarizes the assumptions for hours in each period for the commercial building types. 5We found that distinguishing between businesses that are open on the weekend versus those that are not was unimportant. RI776A B-4 INTERIM REPORT VOLLIE oa LYOdgu WINSLNI Building » CoCweerInauUewnPe BR NR Table B-3 AVERAGE AND TOTAL ELECTRIC AND THERMAL DEMAND Building Name Small Office Large Office Restaurant Large Retail Small Retail Grocery Warehouse Car Service Lodging Medical School College TOTALS Average Area 31,532 123,445 92,355 Average Electric Demand (MWh/yr) 112 2,401 158 1,050 217 729 110 72 2,331 562 1,177 2,349 Average Thermal Demand (MBtu/yr) Total Total Area Electric (million Demand sq -175 29,879 = Bp ft) (MWh/yr) -889 135,745 -424 323,657 -043 50,338 -560 47,300 -288 145,285 -185 90,204 -021 165,138 .144 37,333 -504 95,498 -497 26,694 .878 103,753 Total Thermal Demand (MBtu/yr) 305,823 219,251 115,796 86,821 224,855 116,822 405,132 89,932 143,943 92,478 455,445 43,311 -608 1,250,824 2,299,609 % of Total Electric Thermal peyesodicouy snd0q uorspaqy Decision Focus Incorporated Table B-4 HOURS BY LOAD DURATION CURVE PERIOD Winter Winter Summer Summer Day Night Day Night Total All Building Types Except Schools 1500 1500 2880 2880 8760 Schools 1500 1500 2160 3600 8760 Next, we determined the fraction of total electric and thermal energy requirements needed during the two defined seasons. For all building codes except schools, we used monthly electric billing information [2], ENSTAR total gas sales to the small commercial sector by month [8], the output of a small office building simulation performed by ISER [4], and information on daily load shapes provided by ISER [5]. For schools, we used available monthly billing information for electricity, gas, and oil for the Anchorage [6] and Fairbanks [7] areas. For all building types except schools, we found that electric demand is evenly distributed over the year. The fraction of demand occurring in the winter season was therefore equivalent to the fraction of the year represented by the winter season, or 35 percent. The fraction of electric demand in the summer was thus 65 percent. For schools, we found that 45 percent of electric demand occurs in the winter season, and 55 percent during the summer season. This is consistent with our observation that schools have lower energy requirements during the summer season. For the distribution of thermal demand over the year, we found that, for all building types, 55 percent of the demand occurs in the four selected winter months and 45 percent during the eight "summer" months. Table B-5 summarizes the fractions of electric and thermal demand by season for the commercial building types. Table B-5 ELECTRIC AND THERMAL DEMAND FRACTIONS BY SEASON Fraction of Fraction of Electric Demand Thermal Demand Winter Summer Winter Summer All Building Types Except Schools 0.35 0.65 0.55 0.45 Schools 0.45 2.n0 G.55 0.45 RIT76A B-6 INTERIM REPORT Decision Focus Incorporated We next determined the distribution of electric and thermal demand during the 24-hour daily cycle. We found that peak electric demand is approximately twice off- peak demand; thus, two-thirds of the demand occurs during the day and one-third at night. For thermal demand, we determined that approximately 50 percent of the demand occurs during the day and 50 percent at night [4]. We applied these fractions to all twelve building types. Table B-6 summarizes the fraction of electric and thermal demand by daily cycle. Table B-6 ELECTRIC AND THERMAL DEMAND FRACTIONS DURING DAILY CYCLE Fraction of Fraction of Electric Demand Thermal Demand Day Night Day Night All Building Types On67 Oas3 0.50 0.50 To calculate the load duration curve for the average customer in each building type, we applied the seasonal and daily fractions (shown in Tables 5-5 and 5-6) to the average electric and thermal demand values (shown in Table 5-3), and then divided by the appropriate number of hours (shown in Table 5-4). For example, the average small office customer has an electric demand of 112 MWh per year. During the winter, 35 percent of this demand occurs. Furthermore, 67 percent of the demand occurs during the day. Thus the average electric demand during the winter day period is: (112 MWh) X (35%) X (67%)/1500 hours = 17 kW We performed this type of calculation for each average customer’s electric and thermal demand for each period in the load duration curves. Table B-7 shows the building type, the electric and thermal demand for the average customer, the electric and thermal load duration curves, and the ratio of electric to thermal demand in each period (MWh/MBtu). We used this last calculation later in the analysis to help determine the appropriate cogeneration technology for each building type. B.1.3 Step 3: Determining Technical Potential. We sorted the building types according to their electric demand during the winter day period (called peak demand), and we identified four groups. The building types in the first group have peak demands less than 50 kW and consist of the car service, small office, warehouse, restaurant, and small retail building types. The building types in the second group have peak demands between 50 kW and 200 kW and include the medical, grocery, and RIT76A B-7 INTERIM REPORT VOLLIE Table B-7 LOAD DURATION CURVES FOR COMMERCIAL BUILDING TYPES Electric Demand (kW) Thermal Demand (MBtu) MWh/MBtu Building Building D Type Name Night Day Night Day Night Day Night Day Night oo 1 Small Office 9 17 8 0.046 0.046 0.020 0.020 0.38 0.19 2 = Large Office 187 361 181 0.298 «= 0.298 = 0.127, 0.127 1.25 0.63 3 =- Restaurant 12 24 12 0.067. 0.067. «0.028 ~=— 0.028 0.37 0.18 4 Large Retail 82 158 79 0.353 0.353 0.151 0.151 0.46 0.23 5 Small Retail 17 33 16 0.062 0.062 0.026 0.026 0.55 0.27 1.24 0.62 6 Grocery 57 110 55 0.173 0.173 0.074 0.074 0.66 0.33 1.49 0.74 7 ~~ ‘Warehouse 9 17 8 0.049 0.049 0.021 0.021 0.35 0.17 0.79 0.39 8 Car Service 6 11 5 0.032 0.032 0.014 0.014 0.35 0.18 0.80 0.40 9 edging 181 351 175 (0.644 «0.644 0.275. 0.275 0.56 0.28 1.28 0.64 10 = Medical “4 85 42 0.357. 0.357. 0.152. 0.152 0.24 0.12 0.56 0.28 11 School 118 200 60 0.948 0.948 0.538 0.323 0.25 0.12 0.37 0.19 12 College 183 353 177 0.624 (0.624 (0.266 = (0.266 0.59 0.29 1.33 0.66 LYOdgu WINSLNI payesodioouy sno0q woIspaqy z 3 s Table B-8 LOAD DURATION CURVES FOR AVERAGE CUSTOMER IN EACH GROUP Electric Demand (ki Average = Aver: -------- — Building Building Electric Thermal Winter Day 8 11 6 1 7 17 9 17 1 Small Office 112 251 17 ° 17 3 Restaurant ise 363 25 12 24 wo s Small Retail 217 336 34 17 33 © Averages 21 10 20 2 10 Medical 562 1,948 87 “4 85 42 0.357 0.357 0.152 0.152 0.24 0.12 0.56 0.28 6 Grocery 729 944 113 57 110 55 0.173 0.173 0.074 0.074 0.66 0.33 1.49 0.74 ‘4 Large Retail 1,050 1, 927 163 82 158 79 0.353 «0.3530 «0.151 0.151 0.46 0.23 1.05 0.52 Averages 121 61 117 59 0.44 0.22 1.00 0.50 3 u School 1,177 5,168 235 118 200 60 0.948 += 0.948 = 0.538 = 0.323 0.25 0.12 0.37 0.19 ‘ 9 Lodging 2,331 «3,514 363 181 351 175° (0.644 0.644 0.275 0.275 0.56 0.28 1.28 0.64 12 College 2,349 3,404 365 183 353 177 0.266 0.266 0.59 0.29 1.33 0.66 2 Large Office 2,401 1,626 373 361 1861 0.127 1.25 0.63 1 Averages 367 178 = 0.522, 0.522, 0.223 (0.223 0.66 0.33 LYOdaa WISLNI paywiodioouy sno0q uolspaq Decision Focus Incorporated large retail types. The third group consists solely of schools due to their unique load duration curve with a peak demand around 300 kW. The building types in the fourth group have peak demands greater than 300 kW and include the lodging, college, and large office types. Then we calculated the average electric and thermal load duration curves and the average electric to thermal ratio for each group. Table B-8 shows these results. To determine the technical potential for cogeneration in each group in each of the four regions, we multiplied the average peak electric demand for each group by the number of customers in that group in each region. The number of customers in each group in each region is the sum of the number of customers in each building type in the group. We estimated the regional number of customers in a building type by taking the ratio of the total regional floor area to the average customer’s floor area. The total technical potential for these four groups in the Railbelt is 204.7 MW. Table B-9 shows the technical potential for groups 1 through 4. Table B-9 COGENERATION TECHNICAL POTENTIAL OF GROUPS 1-4 (MW) uw Railbelt 88.6 26.2 20.7 69.2 Total 204.7 DWN FIO We included a fifth group in the commercial sector: hospitals. Although we did not have information on hospitals from the ISER survey, we had monthly electricity billing histories for the two hospitals in Anchorage [8]. Information was not publicly available for the hospital in Fairbanks. Since information on hospital thermal demand was not available, we used a ratio of electric to thermal demand of 0.2 MWh/MBtu to derive the thermal demand.° We used all of the same information to create the load duration curves for hospitals as we did for the other commercial building types (excluding schools). Table B-10 shows the electric and thermal demand, the load duration curves, and the electric to thermal ratios for the two hospitals in Anchorage. We used the average of these two hospitals to represent the hospital in Fairbanks, due to lack of better information. Thus, the technical potential for hospitals in the Railbelt is the sum of the peak electric demands for the three hospitals, or 6.6 MW. ®Personal communication with Science Applications International Corporation indicated a ratio of about 0.25 MWh/MBtu for the lower 48 states. We assumed that thermal requirements would be slightly higher in Alaska, hence the 0.20 MWh/MBtu. RIT76A B-10 INTERIM REPORT VOLLIE Il-d LaYOdaa WALNI Providence TOTALS Average (used for Fairbanks) Average Blectric (MWh/yr) 8,000 20,000 28,000 14,000 Average Thermal (MBtu/yr) 40,000 100,000 140,000 70,000 Winter Day 1,244 3,111 4,356 2,178 Table B-10 HOSPITAL TECHNICAL POTENTIAL Electric Demand (kW) Winter Night 622 1,556 2,178 1,089 602 1,505 2,106 1,053 Winter Day 7.333 18.333 25.667 12.833 Thermal Demand (MBtu) Winter Night Summer Winter Day 0.17 0.17 0.17 0.17 Winter Night 0.08 0.08 0.08 0.19 0.19 poyesodioouy sno0q uolsPpacy Decision Focus Incorporated The total technical potential was estimated at 2113 MW. Table B-11 summarizes the cogeneration technical potential results by group and region for the commercial sector. Table B-11 COMMERCIAL SECTOR COGENERATION TECHNICAL POTENTIAL Group Anchorage Fairbanks Kenai Railbelt 1 62.6 15.7 10.3 88.6 2 18.9 5.3 2.30) 26.2 3 1309 3.3 36% 20.7 4 63.1 3.9 mom 69.2 5 4.4 2.0 0.0 6.6 Total 162.9 30.3 18.1 2053 B.2 INDUSTRIAL SECTOR Information on historical electric demand was available for many industrial facilities, which formed the basis for calculating their cogeneration technical potential. We obtained data on peak electric demand and consumption by industry for almost all the companies in each industry [9]. This subsection describes how we chose the industries to include in the analysis, and the assumptions we used to determine the load duration curves for each industry. The major industries in the Railbelt are petroleum processing and transportation, manufacturing, fish processing, construction, and mining. Table B-12 shows electricity demand and consumption for each industry for the four Railbelt regions. We aggregated this information over the regions and sorted the industries by decreasing electric demand as shown in Table B-13. Petroleum processing is the largest industry, with a demand of 28 MW. This industry has substantial thermal demands and several refineries have or are considering building cogeneration systems. Because Tesoro has recently installed a 4 MW cogeneration unit, we subtracted 4 MW from the technical potential for a total of 24 MW. Although manufacturing is the second largest industry in the Railbelt in terms of electric demand, we excluded it because of the limited information and the assumption that their thermal requirements would be relatively small. RIT76A B-12 INTERIM REPORT Table B-12 Decision Focus Incorporated INDUSTRIAL SECTOR ELECTRICITY DEMAND Demand Energy Region Industry (MW) (GWh) Anchorage Manufacturing 14 a2 Fairbanks Petroleum Processing 7 50 Mining a 10 Petroleum Transportation al 4 Construction 1 * Kenai Petroleum Processing 21 129 Manufacturing 7 20 Fish Processing 6 9 Matsu Construction 2) 1 Totals 62 256 Table B-13 ELECTRICITY DEMAND BY TYPE OF INDUSTRY Demand Energy Industry (MW) (GWh) Petroleum Processing 28 179 Manufacturing 21 52 Fish Processing 6 9 Construction 3 2 Mining 3 10 Petroleum Transportation al 4 Totals 62 256 The fish processing industry is the third largest in terms of electric demand. This industry has substantial thermal demand during its summer processing season. Its technical potential is 6 MW. We excluded the construction and petroleum transportation industries because they are not major electric consumers in the Railbelt, and probably have little or no process heat needs. R1776A B-13 INTERIM REPORT Decision Focus Incorporated The mining industry consists of the Usibelli Coal Mine; its demand shown in Table B-13 includes only its present, utility-supplied consumption. A major cogeneration proposal now under consideration by Usibelli to supply thermal energy to a coal-drying process will be examined further during the next two months as part of this effort. Based on these considerations, we defined the technical potential in the industrial sector as the sum of the technical potential in the petroleum processing and fish processing industries, for a total of 30 MW (excluding the 3.0 MW of mining). The load duration curve calculations proceeded as follows. First, we used information on annual customer electric demands from ISER [10]. Table B-14 shows the customer name and annual electricity demand for these industries. We also obtained monthly billing histories from the appropriate electric utility Power Requirements Studies [11], [12], to calculate electricity requirements by season. Table B-14 ELECTRICITY DEMAND FOR PETROLEUM AND FISH PROCESSING 1987 Electricity Demand Industry Customer (MWh/yr) Petroleum Processing Petro Star Refinery 1,878 Chevron USA 8,162 ARCO Alaska 12,566 MAPCo Petroleum 47,964 Tesoro 89,266* Fish Processing Royal Pacific Fish 511 Cook Inlet S13 Allied Fish Processors 565 Columbia Wards 637 Salamotof Seafoods 807 Dragnet Fisheries 819 Inlet Salmon 928 Kenai Packers L,sl6 Seward Fisheries 37310 Seward Fish Processor 5,356 *Prior to 4 MW installation RI776A B-14 INTERIM REPORT Decision Focus Incorporated In the petroleum industry, we found that operation is year-round, resulting in an even distribution of electric demand over the year. In addition, we determined that operation is continuous, 24 hours a day. The thermal demand parallels the electric demand. Thus, electric and thermal demand are coincident and constant over the year and the daily cycle regardless of the number of hours in a period. This allowed us to use the same hours by period as we had for the commercial sector. The fraction of the demand in each period corresponds to the fraction of the year represented by that period. Table B-15 shows the hours by period and the fraction of electric and thermal demand occurring in each period for the petroleum processing industry. Table B-15 INFORMATION FOR LOAD DURATION CURVE FOR THE PETROLEUM PROCESSING INDUSTRY Winter Winter Total Summer Summer Total Day Night Winter Day Night Summer Total Hours 1500 1500 3000 2880 2880 5760 8760 Fraction of Electric Demand by Season == == On35 -=- <= 0.65 1210 Fraction of Thermal Demand by Season -- -- 6.55 _ -- 0.65 xa’ Fraction of Electric Demand During Daily Cycle 0. 50) || 10/-/50) | (055.0 0.50 1.0 -- Fraction of Thermal Demand During Daily Cycle 0.50 0.50 10 0.50 0.50 1.0 -- For the fish processing industry, however, the billing histories showed that approximately 90 percent of the electric demand occurs during four months of our eight-month summer season. So as not to underestimate the peak demands during the summer months, we changed the hours by season in the load duration curve for the fish processing industry. We defined the peak season to be four summer months, and the off-peak season to be the other eight months. The hours by period are thus simply reversed, 5760 hours in the winter and 3000 hours in the summer. Since operation is RIT76A B-15 INTERIM REPORT Decision Focus Incorporated around-the-clock (when the plants are operating), electric and thermal demand will be coincident and constant within a season and daily cycle. We used the 50/50 split between day and night hours. Table B-16 shows the hours by period and the fraction of electric and thermal demand occurring in each period for the fish processing industry. In contrast to the commercial sector, information on end use thermal demand was not available for the industrial sector. As a result, we estimated thermal demand from an assumed ratio of electric to thermal demand. For petroleum refining, we gathered gas consumption information from Tesoro [13] and from data provided by the Dun and Bradstreet Petroleum information service [14] and decided to use a ratio of 0.07 MWh/MBtu. For the fish processing industry, however, we had only partial information from Dun and Bradstreet. We used a ratio of 0.50 MWh/MBtu.’ Table B-16 INFORMATION FOR LOAD DURATION CURVE FOR THE FISH PROCESSING INDUSTRY Winter Winter Total Summer Summer Total Day Night Winter _Day Night Summer Total Hours 2800 2880 5760 1500 1500 3000 8760 Fraction of Electric Demand by Season -- -- 0.08 = -- 0.92 1.0 Fraction of Thermal Demand by Season -= co 0.08 = ar. 0.92 1.0 Fraction of Electric Demand During Daily Cycle 0.50 0.50 1.0 0.50 0.50 2 30 = Fraction of Thermal Demand During Daily Cycle 0.50 0.50 30 0/450) 0.50 (0) -- "We later performed a sensitivity analysis that showed that the choice of ratio within a reasonable range made less than a 1 MW difference in the economic potential. RI776A B-16 INTERIM REPORT Decision Focus Incorporated As with the commercial sector, we calculated the electric and thermal load duration curves for each customer as well as for the average customer in each industry. We applied the information on the load duration curves shown in Tables B-15 and B-16 and the electric to thermal demand ratios to the electric demand information in Table B-14. For the fish processing, we used the average of the small fish processors.® Tables B-17 and B-18 show the results for petroleum and fish processing industries, respectively. ®The two large Seward fish processors were excluded from the average. Due to the low cogeneration potential identified for fish processors, we assumed that the potential for the larger processors would not be appreciably better. RIT76A B-17 INTERIM REPORT VOLLIN 8T-a LaOdau WIYSLNI Table B-17 LOAD DURATION CURVES FOR THE PETROLEUM REFINING INDUSTRY Electric Demand (kw) ‘Thermal Demand (MBtu) doh feat u Electric Thermal Demand Demand Winter Winter Summer Summer Winter Winter Summer Sumer Winter Winter Sumer Sumer PETROLEUM PROCESSING* Qem/yr) (Qm@tu/yr) Day Might Day Might Day Might Day Wight Day Might Day Hight Region Petro star Refinery 1,878 219 219 212 212 3.130 3.130 3.028 = 3,028 0.07 0.07 0.07 0.07 Fairbanks Chevron USA 8,162 952 952 921 921 13.603. 13.603 13.158 13.158 0.07 0,07 0.07 0.07 Kenai ARCO Alaska 12,566 179,514 1466 1466 1418 1418 = 20,943 20.943 20.258 © 20.258 0.07 0.07 0.07 0.07 Kenai Tesoro 89,266 1,275,229 104146 10414 10073 10073 148.777 148.777 143.906 143.906 0.07 0.07 0.07 0.07 Kenai Average 3263 3263 3156 315647 a7 “5 4s 0.07 0.07 0.07 0.07 ‘marco is analysed separately (refer to Appendix &) Table B-18 LOAD DURATION CURVES FOR THE FISH PROCESSING INDUSTRY Electric Demand (km) Thermal Demand (Btu) oth fast u Rlectric ‘Thermal - == wae n nnn n nen ne cnn enna nena nnnnn nena nn ==- wennnanne- ---------- eacennnnnnnnn-n= Demand Demand Winter Winter Sumer Sumer Winter Winter Summer Sumer Winter Winter Sumer Sumer ISM PROCESSING Qem/yr) (@tu/yr) Day Might Day Right Day Day Hight Day Might Day Might Region Allied Fish Processors 565 1,130 8 8 173 173 0.016 0.347 = 0.347 0.50 0.50 0.50 0.50 Kenai Cook Inlet 513 1,026 7 7 157 157 0.014 0.315 0.315 0.50 0.50 0.50 0.50 Kenai Dragnet Fisheries 819 1,638 ql a 251 251 0.023 0.502 0.502 0.50 0.50 0.50 0.50 Kenai Royal Pacific rish 511 1,022 7 7 157 157 0.014 0.313 0.313 0.50 0.50 0.50 0.50 Kenai Inlet Salmon 928 1,856 13 13 285 285 0.026 0.569 0.569 0.50 0.50 0.50 0.50 Kenai Salamatof Seafoods 807 1,614 11 1. 247 247 0.022 0.495 0.50 0.50 0.50 0.50 Kenai Columbia Wards 637 1,274 9 9 195 195 0.018 0.391 0.50 0.50 0.50 0.50 Kenai Kenai Packers 1,376 2,752 19 19 422 422 0.038 0.844 0.50 0.50 0.50 0.50 Kenai Seward Fish Processor 5,356 Seward Fisheries 3,310 27,910 _ _— AVERAGE W/O SEWARD 11 1 236 236 0.021 0.021 0.472 0.472 0.50 0.50 0.50 0.50 peyeiodioouy sno0q uolspaq B3 {1] {2} [3] [4] [5] [6] (7] [8] [9] [10] {11] {12} {13] (14] RI776A Decision Focus Incorporated REFERENCES Commercial End Use Survey, Institute for Social and Economic Research, 1988. Ibid. Personal communication with Dan Dieckgraf, November, 1988. Provided by Alan Mitchell, ISER, November 1988. Anchorage Municipal Power and Light 1987 daily load shapes provided by Steve Colt, November, 1988. Provided by Steve Colt, ISER, November 1988. Provided by Dave Ferree, Maintenance Manager, Fairbanks North Star Borough School District, November, 1988. Provided by Steve Colt, ISER, November, 1988. ISER industrial sector database, provided by Steve Colt, November 1988. Ibid. Power Requirements Study, Prepared for Golden Valley Electric Association, Prepared by CH2M Hill, August 1987. Homer Electric Association, Power Requirements Study, Attachment A: System Consumer and Sales Data, Large Commercial Load Data, and Commercial Class Load Summary, 1988. Personal communication with David Brown, Tesoro, December, 1988. Major Industrial Plant Database, State of Alaska, Dun & Bradstreet Petroleum Information, 1988. B-19 INTERIM REPORT Decision Focus Incorporated Appendix C SUMMARY OF ISER END USE SURVEY This appendix will be included in the final report. RI776A C-1 INTERIM REPORT Decision Focus Incorporated Appendix D CHARACTERIZATION OF POTENTIAL COGENERATION TECHNOLOGIES Because of the small-scale potential cogeneration applications in the Railbelt, we focused on small (less than 5 MW) cogeneration technologies. Small cogeneration technologies typically use an internal combustion engine (ICE) or a combustion turbine as the prime mover. Following is a brief description of the two technologies. D.1 INTERNAL COMBUSTION COGENERATION SYSTEM Energy conversion is achieved with good efficiency in internal combustion engines because of low heat rejection. Typical heat rates of ICEs are between 10,000 Btu/kWh and 12,000 Btu/kWh. ICEs with high compression ratios (15:1) maintain good performance (low heat rates) at low loadings (around 13,000 Btu/kWh at 25 percent loading). Low compression ratio ICEs (10:1) have higher heat rates (16,000 Btu/kWh at 25 percent loading). Heat recovery from ICE systems is usually achieved through heat exchangers by using the available heat in both (or either) the engine exhaust and the water jacket. Steam generation is not always feasible from the water jacket. Smaller high speed engines (1200 rpm) can operate at jacket temperatures of up to 250°F, which makes steam generation at 20 to 30 psia possible. Larger lower speed engines (900 rpm) generally operate at jacket temperatures of about 180°F, which eliminates jacket heat as a source of steam. Steam generation at pressures of 30 to 165 psia must be derived from exhaust heat recovery only. ICE cogeneration systems usually have a high electric/thermal output ratio (up to 0.6 MW/MBtu/hr). Higher pressure steam recovery lowers the amount of heat recovery without affecting electricity generation and therefore increases the electric/thermal output (to as high as 2). Thermal energy recovery is usually achieved without compromises in electricity production in most ICE applications. RIT76A D-1 INTERIM REPORT Decision Focus Incorporated D.2_ COMBUSTION TURBINE COGENERATION SYSTEMS Heat rates of combustion turbine systems vary by engine type, size, and manufacturer. Typical heat rates are 12,000 Btu/kWh to 14,000 Btu/kWh. Heat rates can increase by as much as 5,000 Btu/kWh when the turbine is operated below 50 percent of rated capacity. Heat recovery in combustion turbine systems is achieved from the turbine exhaust by using a heat recovery steam generator. The temperature of the recovered heat is usually limited by the turbine exhaust temperature. Typical turbine exhaust temperatures vary between 760°F and 1100°F., Supplementary firing may be used to increase this temperature up to 1400°F and a wide range of steam pressures (from 30 psia to over 300 psia). Heat recovery combustion turbine systems have electric/thermal output ratios between 0.04 and 0.11. Lower ratios can be achieved through supplementary firing of the turbine exhaust. Heat recovery is achieved without compromises in electricity production. The choice and size of prime mover can be a function of many factors. In this analysis, we focused on peak electric demand and the ratio of electric to thermal demand as the main determinants of the cogeneration system. We chose cogeneration systems that best matched the thermal and electric needs of the average customer in each group. Table D-1 shows the cogeneration unit cost and performance information used for the commercial and industrial sector analyses [1], [2]. Table D-1 COGENERATION UNIT COST AND PERFORMANCE (1987 dollars) Fixed Capital Initial O&M Heat Power to Size* Cost Investment Cost Rate Heat Ratio Group (kW) ($/kW) ($M) ($/kW/yr) (Btu/kWh) (MW/MBtu/hr) 1 50 1800 0 90 12,000 O72 2 200 1100 0.06 70 11,000 0.45 3 300 800 0.522 60 11,000 022 4 500 550 G.476 45 11,000 0.4 5 2500 565 0.3 so 10,000 0.50 Petro- leum 4300 1072 2.027 65 12,000 0.09 Fish 200 1100 0.06 70 11,000 0.50 *All are internal combustion engines, except the 4300 kW system which is a combustion turbine with heat recovery. RIT76A D-2 INTERIM REPORT Decision Focus Incorporated D.3 [1] [2] RIT76A REFERENCES Reference Guide to Small Cogeneration Systems for Utilities, RMR Associates, Electric Power Research Institute (EPRI), EM-4371, Final Report, February 1986. Small Cogeneration System Costs and Performance, Science Applications International Corporation, EPRI, EM-5954, Final Report, August 1988. D-3 INTERIM REPORT Decision Focus Incorporated Appendix E SUMMARY OF OTHER POTENTIAL COGENERATION PROJECTS This appendix summarizes the information we collected on various potential cogeneration projects discussed during our meetings with Railbelt electric utility staff in the fall of 1988. The project descriptions are organized by the four main Railbelt regions: Anchorage, Kenai, Fairbanks, and Copper Valley. Table E-1 at the end of this appendix summarizes the total potential capacity of each project and its estimated likelihood of being developed. Since detailed information necessary to perform economic evaluation was not available, our judgments of the likelihood of project development were based only on the general information that was available. E.1 ANCHORAGE REGION E.1.1 Chugach Alaska Corporation This has been proposed as a 5 MW project fired by hog fuel—the refuse from logging operations. The online date could be 1990. We were not able to gather much information on this project since the main contact, Paul Tweinten, of Chugach Alaska Corporation, was out of the country during the study period. Their on-site electric demand is 3.25 MW, currently served by Chugach Electric Association. We have insufficient information to make a judgment on the likelihood of project development E.1.2 Anchorage International Airport The State of Alaska, Department of Transportation and Public Facilities (DOTPF) contracted with USKH Engineers to study the technical and economic feasibility of cogeneration at the Anchorage International Airport (AIA). We received the Phase II report (December 7, 1988) from DOTPF. The study analyzed the economics of a conventional thermal plant versus gas cogeneration systems in providing the electric and thermal needs of the Domestic and International Terminals. The study recommended the installation of a 1.05 MW gas engine cogeneration unit to provide electricity, heating, and cooling to the Domestic Terminal only. RIT76A E-1 INTERIM REPORT Decision Focus Incorporated The report stated that the 1.05 MW gas engine generator has a Savings to Investment Ratio (SIR) of 1.12 relative to a conventional plant. Its lifecycle cost was $14.39 million versus $14.5 million for the conventional plant. The study found that the best operating strategy for the cogeneration plant was continuous operation. In this way, most of the electricity and thermal energy could be utilized. USKH studied the sensitivity of the SIR to various input parameters and found it was most sensitive to electricity rates, natural gas costs, and discount rate. We believe that the discount rate is the most significant parameter since electricity and gas rates will change together. The USKH sensitivity analysis examined the impacts of electricity and gas rates separately. The report concluded by recommending the implementation of the 1.05 MW facility and stated that "the future would need to consistently favor negative influences in order for the recommended cogeneration system not to be economically superior to a totally conventional plant" (p. 39). We conclude that the 1.05 MW system is very likely to be developed. Note that the USKH analysis, like all of the cogeneration forecasts developed in Section 5, was performed with respect to retail electricity rates from the perspective of the investor, not the cost of generation to the electric utility. E.1.3 Post Office/Federal Express We tried to determine whether plans for cogeneration exist for the Post Office and the new Federal Express building in Anchorage. We had heard that the Post Office had spent $250,000 on a study of cogeneration. We talked with Jim Janneck of USKH about these possible projects. He said he had also heard of the Post Office study and was aware that Federal Express was building a new facility, but was not aware of any plans for cogeneration. We were not able to get in touch with the Post Office and Federal Express. The APUC may have more information. We do not include this potential project due to the lack of evidence indicating cogeneration plans. E.1.4 Valley Energy Wood Plant This proposed 15 MW, wood-fired project is owned by Hydro Development Inc. (HDI) of Littleton, Colorado. We spoke with Don Pope who said he would send us the engineering and economic information on the project. We have not yet received this as of this draft (January 1989). RIT76A E-2 INTERIM REPORT Decision Focus Incorporated This project and another peat-fired project were previously owned by Matsu Energy, which was then bought out by HDI. Subsequently, the APUC approved an agreement that requires MEA to purchase all its power from Chugach Electric Association. Thus potential QFs in the MEA service area will only be entitled to CEA’s avoided costs. The low CEA avoided costs cast doubt on the future of the wood plant; the peat-fired plant has been abandoned. HDI told us they had spent several thousand dollars arguing this arrangement without success. Because HDI can not commit to building the project, they cannot attract steam users to locate near the project. We believe this project is very unlikely to be built. E.1.5 Hydro We were unsuccessful in contacting either Earl Ausmen (run-of-river, 600 kW) or Jo Jenkie (run-of-river, 1000 kW) about their proposed projects. E.1.6 Anchorage Alaska Project Originally, this was to be a 50 MW, waste-coal fired cogeneration facility potentially coming on-line in 1996 and owned by SGI International, Inc. We spoke with John Cooley at Anchorage Municipal Light & Power (MLP) who had serious doubts that the project would ever come on-line. He said they had not heard from the developer in over five months. We spoke with the developer’s lawyer, Oren Orndorff of Boise, Idaho, and were told that there is a controversy over the avoided costs to be available to this project. SGI filed a complaint against MLP. We recently received a copy of a "Memorandum on How to Proceed" (dated 11/29/88) to the APUC from the developer’s lawyer stating that SGI has transferred its rights to develop the Anchorage Alaska Project to Rosebud Enterprises, Inc. The memorandum states that Rosebud Enterprises is negotiating with steam hosts in the Anchorage area and requests a six-month stay in the proceedings to finalize site selection and steam demand. Although there is renewed effort to determine steam hosts, the history and current status of the project suggests that its development is unlikely. E.1.7 Water and Waste Water We asked John Cooley at MLP what he knew about potential cogeneration projects having to do with water and/or waste water. He said that the local water utility has a contract to divert water from Eklutna for its water supplies with the condition that they replace the power that the water could have generated through the hydro project. Currently, the water utility makes payments to MLP in lieu of replacing the power. It RI776A E-3 INTERIM REPORT Decision Focus Incorporated is possible that the water utility is considering building a cogeneration facility to provide this power. The MLP forecast of the needed replacement power is 4.5 GWh in 1994, and 16 GW in 2025 with the caveat that these numbers may be high due to assumptions regarding population growth. We do not include this potential project due to the lack of evidence indicating cogeneration plans. E.2. KENAI REGION E.2.1 Tesoro Refinery Tesoro has installed two 4 MW gas turbine cogeneration units with heat recovery steam generators. One is for backup purposes only. Tesoro leases the units and the facility from Solar. The facility has the capacity to hold 5 turbines. The unit that will operate was being tested during the fall of 1988 and was expected on-line by the end of 1988. The unit will operate at a 100 percent capacity factor and produces 55,000 pounds per hour of steam with the supplemental firing. David Brown at Tesoro told us that their power costs go down to 3 cents per kWh, assuming they utilize 100 percent of the steam output. Homer Electric Association (HEA) is limiting Tesoro’s electricity generation to 38 GWh per year or less (the 4 MW unit would produce about 35 GWh per year at 100 percent capacity factor) until 1990. At that time, if Tesoro’s load has grown, it can increase its electricity generation to up to one-half of HEA’s load in excess of 396 GWh per year. Since the 4 MW unit is already installed and operating, we did not include it in the assessment of economic cogeneration potential. We have inadequate information to judge the likelihood of additional cogeneration in the future. E.3. FAIRBANKS REGION E.3.1 Alaska Energy Management (AEM) This project was originally conceived of as a 25 MW cogeneration facility, but now the plans are for a 15 MW waste coal-fired fluidized bed plant. Mike Tavella at APUC estimated it could come on-line between 1992 and 1994. The developer has an order from the APUC and is negotiating with Golden Valley Electric Association (GVEA). We found out that AEM and SGI (now Rosebud—see Section E.1.6) have the same corporate officers and lawyer, Oren Orndorff of Boise, Idaho. We spoke with Oren Orndorff who told us the project had qualified as a QF and that they were actively negotiating with GVEA. He would not reveal any additional information about the project. Again, we have insufficient information to judge the likelihood of development. RIT76A E-4 INTERIM REPORT Decision Focus Incorporated E.3.2 MAPCo We spoke with Jerry Fritz at the refinery who told us they were thinking about a 5 to 6 MW cogeneration unit. They plan to do an inhouse study in 1989. They stated that their prime motivation for cogeneration is to increase the reliability of their electricity supply. We have insufficient information to judge the likelihood of development. E.3.3 Usibelli—The Healy Project This has been proposed as a 50 to 150 MW cogeneration facility located at the mine mouth to come on-line in 1993 at the earliest. It is proposed as a coal-fired atmospheric fluidized bed combustion plant with the waste heat being used by the coal processor to dry the high moisture coal. This would increase the heat content of the coal by 10,000 to 11,000 Btu per pound. The fuel for the plant would be either waste coal (6000 Btw/lb) or the standard product (8000 Btw/lb). The initial capacity of the drying facility is estimated to be 500,000 tons from 650,000 tons of raw feed. The economics of the plant will depend largely on the market for this coal. As of this draft (January 1989), the details of the plant’s economics are not known. Further examination of this proposal is planned as part of this study effort. E.4 COPPER VALLEY REGION E.4.1 Alaska Pacific Refining Inc. (APRI) According to Richard Schuller of the APRI parent company, they are considering building a refinery in Valdez and plan to build a cogeneration facility with supplemental firing to supply the refinery’s electric and process heat needs as well as 30 MW of additional electric demand. He stated the total capacity of the plant would be 105 MW (2 gas turbines and 1 steam turbine). APRI believes its cost-of-service from the plant would be the lowest in the Railbelt due to the availability of cheap gas from the propane separation operation to be conducted in Valdez. Richard Schuller told us APRI was negotiating power purchase rates with Copper Valley Electric Association (CVEA). Construction of the APRI refinery was included only in the high demand forecast prepared by APA for the Copper Valley area. Given that judgment, and the somewhat sketchy information on this project, project development appears unlikely. RI776A E-5 INTERIM REPORT Decision Focus Incorporated Table E-1 SUMMARY OF OTHER POTENTIAL COGENERATION PROJECTS Region Anchorage Chugach Alaska Corporation Anchorage International Airport Post Office/Federal Express Valley Energy Wood Plant Hydro Anchorage Alaska Project Water & Waste Water TOTAL Kenai Tesoro Fairbanks Alaska Energy Management MAPCo Usibelli TOTAL Copper Valley Alaska Pacific Refining Inc. RAILBELT TOTAL *NI: No information **4 MW unit already installed ***To be examined further RI776A E-6 Total Potential MW Likelihood 5.0 == 2.05) Very Likely NI* -- 1550 Very Unlikely 1.6 - 50.0 Unlikely NI* -- 72.65 kk 15.0 5.0 50) 00%*™* == 70.0 30.0 Unlikely 172.65 INTERIM REPORT Decision Focus Incorporated Appendix F POWER PLANTS DATA NOTES TO TABLES a Many of the retirements shown are likely to be extended by repowering. Appropriate modifications to this schedule will be covered in the final report. 2. Fuels are defined as follows: Gasl : Chugach natural gas (Beluga/Bernice Lake plants) Gas2 : Enstar natural gas (Anchorage area only) Gas3 : Soldotna plant natural gas (including delivery charge) Oil2 : Fuel Oil #2 (Diesel)—Mainly ICEs Oil4 : Fuel Oil #4—CTs of GVEA Coall : Healy coal delivered to Fairbanks Coal2 : Minemouth Healy coal. 3. FMUS Chena CT #4 cannot be operated (EPA restriction) and is not shown. 4, The Soldotna plant, owned by the Alaska Electric Generation and Transmission Cooperative, is listed as belonging to HEA. 5. Variable operating costs include both fuel costs and variable O&M costs. 6. All costs are in 1987 dollars. RIT76A F-1 INTERIM REPORT VOLLIa oa LaOdga WIYALNI UNIT NAME AMLPCT#1 AMLPCT#2 AMLPCT#3 AMLPCT#4 AMLPCC56 AMLPCC76 AMLPCT#S CHENST#1 CHENST#2 CHENST#3 CHENST#5S CHENCT#6 FMUSIC#1 FMUSIC#2 FMUSIC#3 BELCT#1 BELCT#2 BELCT#3 BELCT#4 BELCT#5 BELCC68 BELCC78 BERNCT#1 BERNCT#2 BERNCT#3 BERNCT#4 INTCT#1 INTCT#2 INTCT#3 COOPER SESIC#1 SESIC#2 SESIC#3 SESIC#HA SESIC#S FMUS FMUS FMUS FMUS FMUS FMUS FMUS FMUS BSSERRBRRRRBERE SES SES SES SES SES Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Total Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Total Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Kenai Kenai Kenai Kenai Anchorage Anchorage Anchorage Kenai Total Kenai Kenai Kenai Kenai Kenai Total Table F-1 RAILBELT UNITS BY UTILITY CAPACITY PLANT FUEL (MW) TYPE TYPE 16.2 CT Gas2 16.2 cT Gas2 19.4 CT Gas2 33.2 CT Gas2 47.4 CC Gas2 109.3 cc Gas2 86.5 CT Gas2 328.2 5.1 ST Coall 2.0 st Coall 1.5 st Coall 20.0 ST Coall 23.2 CT 0i12 2.8 Ic 0i12 2.8 Ic 0i12 2.8 IC 0112 60.2 15.7 CT 15.7 cT 55.0 CT 8.7 CT 66.0 CT 100.6 Cc 100.6 CC 7.8 CT 18.0 CT 25.0 CT 25.0 CT 15.0 CT 15.0 cT 18.0 CT 17.4 HYDRO 503.5 1.5 Ic 0i12 1.5 Ic 0i12 2.5 Ic oi12 2.5 Ic 0i12 2.5 Ic 0i12 10.5 HEAT RATE (Btu/kWh) 15000 15000 15000 15000 15000 (year) (year) (yr) 57 49 43 36 31 26 27 28 25 26 28 19 20 21 20 49 41 34 13 33 30 29 105 26 26 31 21 20 FORCED PLANNED YEAR YEAR UNIT OUTAGE OUTAGE EQUIV, ON-LINE RETIRED LIFE RATE (1/yr) RATE AVAIL. (1/yr) 0.852 0.865 0.828 0.841 0.828 0.832 0.832 0.865 0.865 0.852 0.828 0.877 0.877 0.804 1.000 0.941 0.941 0.941 0.941 0.941 VAR O&M ($/MWh) 1.29 1.29 1.29 0.68 0.61 24.12 24.12 24.12 1.48 1.48 1.48 1.48 1.48 1.48 1.48 2.31 2.31 2.31 2.31 14.24 14.24 14.24 0.00 FIXED O&M ($/kW/yr) 54.03 54.03 54.03 77.76 9.26 0.92 0.92 0.92 11.85 11.85 11.85 11.85 11.85 11.85 11.85 10.60 10.60 10.60 10.60 20.50 20.50 20.50 5.11 0.62 0.62 0.62 0.62 0.62 peyesodicouy sno0q uoIstaqy VOLLIN ed LYOdgau WIYGLNI Table F-1 (continued) FORCED PLANNED HEAT YEAR YEAR UNIT OUTAGE OUTAGE EQUIV. VAR =~ FIXED UNIT CAPACITY PLANT FUEL RATE ON-LINE RETIRED LIFE RATE RATE AVAIL, OEM OsM NAME OWNER AREA (MW) TYPE TYPE (Btu/kWh) (year) (year) (yr) (1/yr) (1/yr) (1/yr) ($/MWh) ($/kW/yr) SELDIC#1 HEA Kenai 0.3 Ic 0112 14998 1952 2000 49 0.050 0.040 0.912 41.01 2. 97 SELDIC#2 HEA Kenai 0.6 Ic 0i12 12006 1964 2010 47 0.050 0.040 0.912 41.01 2.97 SELDIC#3 HEA Kenai 0.6 Ic 0i12 12006 1970 2011 42 0.050 0.040 0.912 41.01 2.97 SELDIC#4 HEA Kenai 0.6 Ic 0i12 12006 1982 2012 31 0.050 0.040 0.912 41.01 2.97 SOLDOTCT HEA Kenai 39.0 CT Gas3 11900 1985 2010 26 0.050 0.120 0.836 66.67 51.28 Total 41.1 HEALST#1 GVEA Fairbanks 25.0 ST Coal2 12750 1967 2002 36 0.018 0.070 0.913 4.34 73.90 HEALIC#2 GVEA Fairbanks 2.6 Ic 0i12 11210 1967 1997 31 0.010 0.200 0.792 6.05 0.62 NOPOCT#1 GVEA Fairbanks 60.9 CT oi14 10900 1976 2006 Si. 0.010 0.150 0.842 1.51 7.84 NOPOCT#2 GVEA Fairbanks 60.9 CT oi14 10900 1977 2007 31 0.010 0.150 0.842 1.51 7.84 ZENGT#1 GVEA Fairbanks 18.0 CT oil4 14869 1971 2001 31 0.010 0.150 0.842 0.62 9.29 ZENGT#2 GVEA Fairbanks 18.0 cr oi14 14869 1972 2002 31 0.010 0.150 0.842 0.62 9.29 DSLIC#1 GVEA Fairbanks 1.9 Ic 0i12 11209 1961 1991 31 0.050 0.200 0.760 6.05 0.62 DSLIC#2 GVEA Fairbanks 1.9 Ic 0i12 11209 1961 1991 31 0.050 0.200 0.760 6.05 0.62 DSLIC#3 GVEA Fairbanks 1.9 Ic 0i12 11209 1961 1991 31 0.050 0.200 0.760 6.05 0.62 DSLIC#5 GVEA Fairbanks 2.6 Ic 0i12 11210 1970 2000 31 0.050 0.200 0.760 6.05 0.62 DSLIC#HE GVEA Fairbanks 2.6 Ic 0112 11210 1970 2000 31 0.050 0.200 0.760 6.05 0.62 UAFIC#7 GVEA Fairbanks 1.9 Ic 0112 11209 1970 1996 27 0.050 0.200 0.760 6.05 0.62 UAFIC#S GVEA Fairbanks 1.9 Ic 0112 11209 1970 1996 27 0.050 0.200 0.760 6.05 0.62 Total 200.1 SOLGCH#1 CVEA Copper Valley 6.0 HYDRO HYDRO 2054 0.000 0.000 1.000 0.00 0.00 SOLGCH#2 CVEA Copper Valley 6.0 HYDRO HYDRO 2054 0.000 0.000 1.000 0.00 0.00 GLNDSL#1 CVEA Copper Valley 0.3 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#2 CVEA Copper Valley 0.3 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#3 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#4 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#5 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#6 CVEA Copper Valley 2.6 IC 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#7 CVEA Copper Valley 2.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#8 CVEA Copper Valley 2.8 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#1 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#2 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#3 CVEA Copper Valley 0.6 IC 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#4 CVEA Copper Valley 1.8 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#5 CVEA Copper Valley 2.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#6 CVEA Copper Valley 1.0 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 Total 29.6 BRADLEY APA Kenai 119.0 HYDRO HYDRO 1992 2055 64 0.000 0.000 1.000 0.00 24.60 EKLUTNA APA Anchorage 30.0 HYDRO HYDRO 1955 2054 100 0.000 0.000 1.000 0.00 99.48 Total 149.0 payerodioouy sno0q uoIspaq VOLLIE vA LaO0dau WIYSLNI UNIT NAME EKLUTNA BELCC78 BELCC68 AMLPCC76 AMLPCCS5S6 BELCT#5S BELCT#3 AMLPCT#8 BELCT#1 BELCT#2 BELCT#4 AMLPCT#4 AMLPCT#2 AMLPCT#1 AMLPCT#3 INTCT#3 INTCT#2 INTCT#1 COOPER BRADLEY BERNCT#3 BERNCT#4 BERNCT#2 BERNCT#1 SESIC#H2 SESIC#1 SESIC#3 SESIC#S SESIC#HA SOLDOTCT SELDIC#4 SELDIC#3 SELDIC#2 SELDIC#1 Table F-2 UNITS SORTED BY AREA & VARIABLE OPERATING COST PLANT TYPE FUEL TYPE BRRGEG GGRREREQOREOGEEG ES SES SES SES SES SES BBEEE Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai Kenai SCOOCOMONNNEHBA WANNACUUUUUME HYDRO HYDRO ct cr cr cr Ic Ic Ic Ic Ic cT Ic Ic Ic Ic 1994 FUEL cost ($/MBtu) 0.00 1.64 1.64 1.90 1.90 1.64 1.64 1.90 1.64 1.64 1.64 1.90 1.90 1.90 1.90 1.90 1.90 1.90 0.00 0.00 1.64 1.64 1.64 1.64 5.26 5.26 5.26 5.26 5.26 2.59 5.26 5.26 5.26 5.26 HEAT RATE VAR O&M (Btu/kWh) ($/MWh) 9250 9250 8625 10400 12600 12800 11732 15600 17300 18900 13541 14703 14808 17807 14400 15700 15700 13300 13500 15000 20000 15000 15000 15000 15000 15000 11900 12006 12006 12006 14998 0.00 1.48 1.48 0.81 0.81 1.48 1.48 0.81 1.48 1.48 1.48 6.75 6.75 6.75 6.75 14.24 14.24 14.24 0.00 0.00 2.31 2.31 2.31 2.31 6.05 6.05 6.05 6.05 6.05 66.67 41.01 41.01 41.01 41.01 FIXED O&M ($/kW/yr) 99.48 11.85 11.85 9.54 9.54 11.85 11.85 9.54 11.85 11.85 11.85 10.41 10.41 10.41 10.41 20.50 20.50 20.50 5.11 24.60 10.60 10.60 10.60 10.60 0.62 0.62 0.62 0.62 0.62 51.28 2.97 2.97 2.97 2.97 VARIABLE OPERATING costs ($/MWh) 0.00 16.65 16.65 17.20 20.57 22.14 22.47 23.10 27.06 29.85 32.48 32.48 34.69 34.89 40.58 41.60 44.07 44.07 0.00 0.00 24.12 24.45 26.91 35.11 84.95 84.95 84.95 84.95 84.95 97.48 104.16 104.16 104.16 119.90 payeiodioouy snd0q uoIsteq VOLLIN GA LaOdau WIMALNI Table F-2 (continued) 1994 VARIABLE UNIT FUEL HEAT VAR FIXED OPERATING UNIT CAPACITY PLANT FUEL cost RATE O&M O&M costs NAME OWNER AREA (MW) TYPE TYPE ($/MBtu) (Btu/kWh) ($/MWh) ($/kW/yr) ($/MWh) HEALST#1 GVEA Fairbanks 25.0 ST Coal2 1.30 12750 4.34 73.90 20.92 CHENST#5 FMUS Fairbanks 20.0 st Coall 2.52 14236 0.68 77.76 36.55 NOPOCT#1 GVEA Fairbanks 60.9 CT oi14 3.58 10900 1.51 7.84 40.53 NOPOCT#2 GVEA Fairbanks 60.9 CT 0i14 3.58 10900 1.51 7.84 40.53 CHENST#1 FMUS Fairbanks 5.1 ST Coall 2.52 15968 1.29 54.03 41.53 CHENST#2 FMUS Fairbanks 2.0 ST Coall 2.52 18049 1.29 54.03 46.77 CHENST#3 FMUS Fairbanks 1.5 ST Coall 2.52 18091 1.29 54.03 46.88 ZENGT#1 GVEA Fairbanks 18.0 CT oi14 3.58 14869 0.62 9.29 53.85 ZENGT#H2 GVEA Fairbanks 18.0 CT oi14 3.58 14869 0.62 9.29 53.85 UAFIC#7 GVEA Fairbanks 1.9 Ic 0112 5.26 11209 6.05 0.62 65.01 UAFIC#S GVEA Fairbanks 1.9 Ic 0112 5.26 11209 6.05 0.62 65.01 DSLIC#2 GVEA Fairbanks 1.9 Ic 0112 5.26 11209 6.05 0.62 65.01 DSLIC#1 GVEA Fairbanks 1.9 Ic 0112 5.26 11209 6.05 0.62 65.01 DSLIC#3 GVEA Fairbanks 1.9 Ic 0112 5.26 11209 6.05 0.62 65.01 DSLIC#S GVEA Fairbanks 2.6 Ic 0112 5.26 11210 6.05 0.62 65.01 DSLIC#HE GVEA Fairbanks 2.6 IC 0112 5.26 11210 6.05 0.62 65.01 HEALIC#2 GVEA Fairbanks 2.6 IC 0i12 5.26 11210 6.05 0.62 65.01 CHENCT#6 FMUS Fairbanks 23.2 CT oi12 5.26 12733 0.61 9.26 67.59 FMUSIC#2 FMUS Fairbanks 2.8 Ic o0i12 5.26 12128 824.12 0.92 87.91 FMUSIC#1 FMUS Fairbanks 2.8 Ic 0112 5.26 12128 24.12 0.92 87.91 FMUSIC#3 FMUS Fairbanks 2.8 IC 0112 5.26 12128 24.12 0.92 87.91 SOLGCH#2 CVEA Copper Valley 6.0 HYDRO HYDRO 0.00 0.00 0.00 0.00 SOLGCH#1 CVEA Copper Valley 6.0 HYDRO HYDRO 0.00 0.00 0.00 0.00 VALDSL#4 CVEA Copper Valley 1.8 Ic 0112 5.26 13403 107.03 29.73 177.53 VALDSL#5 CVEA Copper Valley 2.6 IC 0112 5.26 13403 107.03 29.73 177.53 VALDSL#1 CVEA Copper Valley 0.6 Ic 0112 5.26 13403 107.03 29.73 177.53 VALDSL#2 CVEA Copper Valley 0.6 Ic 0112 5.26 13403 107.03 29.73 177.53 VALDSL#3 CVEA Copper Valley 0.6 Ic 0i12 5.26 13403 107.03 29.73 177.53 GLNDSL#7 CVEA Copper Valley 2.6 IC 0i12 5.26 13403 107.03 29.73 177.53 GLNDSL#3 CVEA Copper Valley 0.6 IC 0112 5.26 13403 107.03 29.73 177.53 GLNDSL#6 CVEA Copper 2.6 IC 0112 5.26 13403 107.03 29.73 177.53 VALDSL#6 CVEA Copper 1.0 Ic 0i12 5.26 13403 107.03 29.73 177.53 GLNDSL#1 CVEA Copper 0.3 Ic 0112 5.26 13403 107.03 29.73 177.53 GLNDSL#2 CVEA Copper 0.3 Ic 0112 5.26 13403 107.03 29.73 177.53 GLNDSL#8 CVEA Copper 2.8 Ic 0i12 5.26 13403 107.03 29.73 177.53 GLNDSL#5 CVEA Copper 0.6 Ic 0i12 5.26 13403 107.03 29.73 177.53 GLNDSL#4 CVEA Copper 0.6 IC 0112 5.26 13403 107.03 29.73 177.53 peyesodioouy snd0q wolsteq VOLLIE LaOdga WALNI accl aCT1 ACT2 ACC2 ACT3 ACT4 ACTS FSsTl FST2 FcT1l UNIT NAME BELCC78 BELCC68 BELCT#5 BELCT#3 BELCT#1 BELCT#2 BELCT#4 AMLPCC76 AMLPCC56 AMLPCT#8 AMLPCT#4 AMLPCT#2 AMLPCT#1 AMLPCT#3 INTCT#3 INTCT#1 INTCT#2 CHENST#5 CHENST#1 CHENST#2 CHENST#3 HEALST#1 NOPOCT#1 NOPOCT#2 Table F-3 OVER/UNDER MODEL TECHNOLOGIES FORCED PLANNED YEAR OUTAGE OUTAGE UNIT CAPACITY (MW) B02 EER E EE S88 98 g8 3 FMUS FMUS 3 BB OE Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Anchorage Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks Fairbanks 121.8 PLANT FUEL TYPE TYPE cc Gasl ce Gasl cr Gasl cr Gasl Gasl cr cr ct cc Gas2 cc Gas2 ct Gas2 cr Gas2 cT Gas2 ctr Gas2 cr Gas2 cr Gas2 ct Gas2 cr Gas2 st Coall st Coall st Coall st Coall st Coal2 cr 0i14 cr oi14 HEAT RATE (Btu/kWh) 9250 9250 9250 12600 12800 12691 15600 17300 18900 16982 8625 10400 9162 11732 13541 14703 14808 17807 14978 14400 15700 15700 15213 14236 15968 18049 18091 15014 12750 10900 10900 10900 RETIRED (year) 1995 1996 1994 1999 1992 1993 1994 1999 1999 2009 1995 1993 1993 1994 1998 1997 1997 2005 2010 2000 1995 2002 2006 2007 RATE (1/yr) 0.060 0.060 0.060 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.020 0.020 0.020 0.040 0.214 0.214 0.214 0.214 0.214 0.050 0.050 0.050 0.050 0.060 0.060 0.060 0.060 0.060 0.018 0.010 0.010 0.010 RATE (1/yr) 0.115 0.115 0.128 0.128 0.103 0.090 0.115 0.144 0.238 0.225 0.154 0.154 0.154 0.154 0.154 0.077 0.077 0.060 0.060 0.060 0.060 0.070 0.150 0.150 EQUIV. AVAIL. (1/yr) 0.832 0.832 0.832 0.828 0.828 0.828 0.852 0.865 0.841 0.855 0.839 0.747 0.811 0.744 0.665 0.665 0.665 0.665 0.665 0.804 0.877 0.877 0.849 0.884 0.884 0.884 0.884 0.884 0.913 0.842 0.842 0.842 VAR O&M ($/MWh) ($/kW/yr 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 0.81 0.81 0.81 6.75 6.75 6.75 6.75 6.75 14.24 14.24 14.24 14.24 0.68 1.29 1.29 1.29 0.86 4.34 1.51 1.51 1.51 FIXED O&M 11.85 11.85 11.85 11.85 11.85 11.85 11.85 9.54 9.54 9.54 10.41 10.41 10.41 10.41 20.50 20.50 20.50 77.76 54.03 54.03 54.03 payeiodioouy snd04q uolspaqy VOLLIN Ld LaOdgau WIYSLNI FcT3 KCT1 KCT2 KCT3 HYDRO FICE1 FICE2 KICE1 UNIT NAME CHENCT#E6 ZENCT#1 ZENCT#2 BERNCT#3 BERNCT#4 BERNCT#2 BERNCT#1 SOLDOTCT BRADLEY COOPER EKLUTNA SOLGCH#2 SOLGCH#1 DSLIC#3 UAFIC#7 DSLIC#1 DSLIC#2 UAFIC#S DSLIC#6 HEALIC#2 DSLIC#5 FMUSIC#1 FMUSIC#2 FMUSIC#3 SESIC#HA SESIC#2 SESIC#5 SESIC#3 SESIC#1 Table F-3 (continued) PLANT TYPE FUEL TYPE peeeeees pees BB BEE BBE FMUS FMUS FMUS SES SES SES SES UNIT CAPACITY (MW) Fairbanks 23.2 Fairbanks 18.0 Fairbanks 18.0 36.0 Kenai 25.0 Kenai 25.0 Kenai 18.0 68.0 Kenai 7.8 Kenai 39.0 Kenai 119.0 Kenai 17.4 Anchorage 30.0 Copper Valley 6.0 Copper Valley 6.0 175.4 Fairbanks 1.9 Fairbanks 1.9 Fairbanks 1.9 Fairbanks 129 Fairbanks 1.9 Fairbanks 2.6 Fairbanks 2.6 Fairbanks 2.6 17.3 Fairbanks 2.8 Fairbanks 2.8 Fairbanks 2.8 8.4 Kenai 2.5 Kenai 1.5 Kenai 2.5 Kenai 2.5 Kenai 1.5 10.5 ct cr cr cr cr cr cr cr HYDRO HYDRO HYDRO HYDRO HYDRO Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic Ic 0112 oil4 oi14 Gasl Gasl Gasl Gasl Gas3 HYDRO HYDRO HYDRO HYDRO HYDRO 0i12 0112 0112 0112 0112 0112 0112 0i12 0112 0112 0112 0112 0112 0i12 0i12 0i12 HEAT RATE (Btu/kWh) 12733 14869 14869 14869 13300 13500 15000 13824 20000 11900 11209 11209 11209 11209 11209 11210 11210 11210 11209 12128 12128 12128 12128 15000 15000 15000 15000 15000 15000 RETIRED (year) 2006 2001 2002 2011 1993 2011 2011 2010 2055 2064 2054 2054 2054 1991 1996 1991 1991 1996 2000 1997 2000 1992 1994 1996 2005 1990 2005 1995 1990 FORCED PLANNED YEAR OUTAGE OUTAGE RATE (1/yr) 0.080 0.010 0.010 0.010 0.050 0.050 0.050 0.050 0.050 0.050 0.000 0.000 0.000 0.000 0.000 0.050 0.050 0.050 0.050 0.050 0.050 0.010 0.050 0.044 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 RATE (1/yr) 0.030 0.150 0.150 0.103 0.128 0.090 0.090 0.120 0.000 0.000 0.000 0.000 0.000 0.200 0.200 0.200 0.200 0.200 0.200 0.200 0.200 0.020 0.020 0.020 0.010 0.010 0.010 0.010 0.010 EQUIV. AVAIL. (1/yr) 0.892 0.842 0.842 0.842 0.852 0.828 0.865 0.847 0.865 0.836 1.000 1.000 1.000 1.000 1.000 0.760 0.760 0.760 0.760 0.760 0.760 0.792 0.760 0.765 0.931 0.931 0.931 0.931 0.941 0.941 0.941 0.941 0.941 0.941 VAR O&M ($/MWh) ($/kW/yr 0.61 0.62 0.62 0.62 2.31 2.31 2.31 2.31 2.31 66.67 0.00 0.00 0.00 0.00 0.00 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 24.12 24.12 24.12 24.12 6.05 6.05 6.05 6.05 6.05 6.05 FIXED O&M 9.26 9.29 9.29 10.60 10.60 10.60 10.60 51.28 24.60 5.11 99.48 0.00 0.00 0.62 0.62 0.62 0.62 0.62 0.62 0.62 0.62 poyeiodioouy snd0q uoIspeq VOLLIE Table F-3 (continued) FORCED PLANNED UNIT HEAT YEAR OUTAGE OUTAGE EQUIV. VAR FIXED TECHNOLOGY UNIT CAPACITY PLANT FUEL RATE RETIRED RATE RATE AVAIL. O&M O&M NAME NAME OWNER AREA (MW) TYPE TYPE (Btu/kWh) (year) (1/yr) (l/yr) (1/yr) ($/MWh) ($/kW/yr SELDIC#4 HEA Kenai 0.6 Ic 0112 12006 2012 0.050 0.040 0.912 41.01 2.97 SELDIC#3 HEA Kenai 0.6 Ic 0i12 12006 2011 0.050 0.040 0.912 41.01 2.97 SELDIC#2 HEA Kenai 0.6 Ic 0i12 12006 2010 0.050 0.040 0.912 41.01 2.97 SELDIC#1 HEA Kenai 0.3 Ic 0i12 14998 2000 0.050 0.040 0.912 41.01 2.97 KICE2 2.1 12433 0.050 0.912 41.01 SELDIC#4 HEA Kenai 0.6 Ic 0112 12006 2012 0.050 0.040 0.912 41.01 2.97 SELDIC#3 HEA Kenai 0.6 Ic oi12 12006 2011 0.050 0.040 0.912 41.01 2.97 SELDIC#2 HEA Kenai 0.6 Ic 0112 12006 2010 0.050 0.040 0.912 41.01 2.97 3 SELDIC#1 HEA Kenai 0.3 Ic 0112 14998 2000 0.050 0.040 0.912 41.01 2.97 oo KICE2 2.1 12433 0.050 0.912 41.01 GLNDSL#6 CVEA Copper Valley 2.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#2 CVEA Copper Valley 0.3 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#2 CVEA Copper Valley 0.6 IC 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#1 CVEA Copper Valley 0.3 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#7 CVEA Copper Valley 2.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#8 CVEA Copper Valley 2.8 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#4 CVEA Copper Valley 1.8 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#5 CVEA Copper Valley 2.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#1 CVEA Copper Valley 0.6 Ic 0i12 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#3 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#3 CVEA Copper Valley 0.6 Ic 0i12 13403 2000 0.018 0.014 0.968 107.03 29.73 VALDSL#6 CVEA Copper Valley 1.0 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#5 CVEA Copper Valley 0.6 Ic 0112 13403 2000 0.018 0.014 0.968 107.03 29.73 GLNDSL#4 CVEA Copper Valley 0.6 Ic 0i12 13403 2000 0.018 0.014 0.968 107.03 29.73 CICE1 17.6 13403 0.018 0.968 107.03 LadOdga WIYALNI peyeiodioouy snoog worse