EEEcon 458 LongTerm Planning J Mc Calley A

EE/Econ 458 Long-Term Planning J. Mc. Calley A complementary treatment focused on resource adequacy is given in “Maintaining reliability in the modern power system, ” US Dept. of Energy, Dec. , 2016, www. energy. gov/sites/prod/files/2017/01/f 34/Maintaining%20 Reliability%20 in%20 the%20 Modern%20 Power%20 System. pdf 1

What I will do in this class 1. Describe the need for adequacy; 2. Identify who ensures adequacy for a. Traditionally regulation b. Competitive electricity markets 3. Distinguish capacity markets from long-term planning; 4. Describe the long-term planning process; 5. Describe generation expansion planning (GEP) problem; a. Give formulation and relate it to capacity markets b. Describe several critical features of GEP 6. Describe two other tools: a. Transmission expansion planning (TEP) b. Cooptimization expansion planning (CEP) 2

Adequacy “Adequacy is the ability of the electric system to supply the aggregate electric power and energy requirements of the electricity consumers at all times, taking into account scheduled and reasonably expected unscheduled outages of system components. ” “Definition of ‘Adequate Level of Reliability’, ” North American Electric Reliability Corporation (NERC), approved by the NERC Operating Committee and the NERC Planning Committee, December 2007, available at http: //www. nerc. com/docs/pc/Definition-of-ALR-approved-at-Dec-07 -OC-PC-mtgs. pdf. The single most influential way to maintain adequacy is to ensure the area’s generating capacity will almost always exceed the area’s peak requirement for each year in a given planning horizon. Two qualifications to the last statement… 3

Adequacy Generator failures: The “almost always” part of this last statement recognizes that generation may fail, and so no matter how much generation is built, it is impossible to ensure the area’s generation capacity will definitely exceed the area’s peak requirement for each year in a given planning horizon (there is always the possibility, albeit very small, that enough generation may fail simultaneously to cause remaining generation to be unable to meet the demand). 4

Adequacy Non-generator means of ensuring adequacy: Adding generation capacity is not the only way to improve adequacy. • Other node-based options: Alternatively, one may consider purchasing capacity from neighboring utilities, demandside means (e. g. , conservation programs and/or load control during peak periods), and/or storage. • Transmission: There may be situations where the total generation is in fact sufficient, but the transmission between that generation and the load is insufficient to carry all of the generation that can be produced. Although building generation close to the load center may be an option, it might be an expensive one. In such a case, adding transmission may be the most cost-effective means 5 of achieving adequacy.

Who Ensures Adequacy in Traditional Regulation In traditional regulation, for any given service area, there is a single entity having responsibility, actually an obligation, to maintain adequacy for the region’s customers - the vertically integrated utility. The vertically integrated utility accepts this obligation in exchange for a guarantee that they be allowed to earn a “fair rate of return” on their investments through energy rates charged to customers. Four observations to make about this “exchange”: 1. Compact: This “exchange” is an explicit agreement between the electric utility and its customers, known as the “regulatory compact. ” 2. Monopoly: It is usually provided that the vertically integrated utility will be the only organization having the regulatory compact within a given service area, thus, the regulatory compact ensures the vertically integrated utility operates as a monopoly. 6

Who Ensures Adequacy in Traditional Regulation 3. Regulator: It is awkward for the utility to make an agreement with all of its customers. Thus, the customers it serves are represented by the state regulator (the “utility board, ” the “public service commission, ” or the “public utilities commission”). The state regulator has sole authority to regulate the rates the electric utility charges the people of the state. The state regulator also perceives oversight of resource adequacy within their state to be its responsibility, e. g. , • Iowa Utilities Board: “Performance Reports for the Iowa Utilities Board (IUB) highlight the services the IUB provided to Iowans, along with results achieved to ensure utility service reliability and to improve and expand utility service infrastructure in Iowa. ” (see https: //iub. iowa. gov/iowa-utilities-board-performance-reports-0) • Organization of MISO States: “Resource Adequacy: • Monitor resource adequacy implications of solutions proposed through the Resource Availability and Need (RAN) Initiative. • Coordinate demand-side resource interactions between retail and wholesale tariffs. • Ensure the generator interconnection process is not acting as a barrier to maintaining resource adequacy. • Facilitate continued improvement of the OMS-MISO Resource Adequacy Survey. ” (see https: //www. misostates. org/index. php/about/strategic-priorities ) 7

Who Ensures Adequacy in Traditional Regulation 4. Energy rates: It is mainly through energy rates that revenues are provided to the utility to maintain adequacy. However, the maintenance of adequacy is more of a capacity issue rather than an energy issue. The difference between capacity and energy is that whereas capacity determines whether demand can be met at a singular point in time; energy is the integration of that demand (or the generation that supplies it) over a time interval. Thus, because energy rates are used to pay for capacity needs, there needs to be an “adder” to the short-term cost of energy to pay for the longer-term cost of capacity. Regulators are aware of this need and understand they must be willing to allow the electric utility to reflect this adder in their energy rates. 8

Who Ensures Adequacy in CEM? When some regions in the US introduced competitive electricity markets (CEM), the regulatory compact had to be revisited. The reason for this was that because CEM necessarily relied on having many competitive market participants, the vertically integrated utility • could no longer be granted monopolistic status; • neither could a rate of return be guaranteed, at least not for the parts of its organization which participated in the CEM. In CEM, because the regulator no longer provides these two guarantees to the utility, the utility could no longer be expected to accept the obligation to ensure the reliability of the service area. This left open the question: what is the revenue source to maintain adequacy? Early answer was “The day-ahead and realtime (DART) electricity markets. ” 9

Who Ensures Adequacy in CEM? The thinking: At the heart of the CEM were the day-ahead/real-time (DART) markets where market participants provided offers to sell energy and ancillary services. • Because revenues received by energy suppliers was dictated by the market clearing price, and because the market clearing price would by definition be equal to (for the marginal units) or higher than (for all other units) the offers to sell energy made by the market participants, there would be a difference between what market participants were paid and what they were willing to be paid (their costs), and this difference, integrated over time, would be sufficient to fund development of additional generation capacity. • An important feature of this thinking is that the price of energy would become particularly high during periods of peak demand, and extremely high during the rare moments when generation is just sufficient (or insufficient) to meet demand. This is illustrated on the next slide. 10

Who Ensures Adequacy in CEM? Baseload units are making much money when price is established by emergency peakers. These short-duration bursts of revenue are very useful to incent suppliers to build additional capacity. 3 comments: 1. True vs. offered: The figure is based on suppliers offering true marginal costs, which they do when capacity significantly exceeds demand. But when capacity is just sufficient (or insufficient) for supplying demand, suppliers increase their offers to push the clearing price up based on the realization that if the market needs all the supply it can get, it will be impossible to offer too high a price, a behavior that is consistent with the following supplier thinking: “Even if I offer far above the next highest offer, I will be selected (and will become the marginal unit, i. e. , the price-setting unit). ” Thus, the price can soar to very high levels during such a condition. $300/MWhr Solid line shows normal supply curve. Dashed line shows supply curve when some units (in this case, the expensive peakers) fail. Expensive peakers $160/MWhr Efficient peakers $130/MWhr Price-responsive demand reduction Mid-range units $90/MWhr $50/MWh Emergency peakers Baseload units Peak condition MW This is how a market operates when the commodity is scarce. There exists a large literature on the topic of scarcity pricing which argues that such price increases are not bad in and of themselves and in fact provide the very signal that suppliers need to increase their capacity. P. Cramton, A. Ockenfels, and S. Stoft, “Capacity market fundamentals, ” May 26, 2013, unpublished paper, available at 11 http: //www. cramton. umd. edu/papers 2010 -2014/cramton-ockenfels-stoft-capacity-market-fundamentals. pdf.

Who Ensures Adequacy in CEM? 2. Price caps: Most electricity markets provide price caps to limit the energy price that can be reached during scarcity conditions. (PJM imposes a price cap of $3. 7 k/MWhr; MISO $3. 5 k/MWhr, ERCOT $9 k/MWhr) to protect customers against extremely volatile and very high prices, see www. aeso. ca/assets/Uploads/3 -Price-and-Offer-Cap-and-Floor-in-Other-Jurisdictions-2017 -11 -08. pdf). It has been argued that such price caps inappropriately skew the need for capacity investment by limiting the additional revenues that can be obtained during scarcity conditions and therefore should not be allowed. However, some argue that the basic problem cannot be addressed by lifting the price cap because the price of blackouts are not known and cannot be known. 3. Blackouts: The extreme form of an event caused by a lack of adequacy is a blackout whereby a large number of customers are interrupted for a significant duration. The cost of such sustained load interruptions has been referred to as the value of lost load (VOLL), and estimates vary greatly with region and with type of customer, as indicated in the figure There is uncertainty with VOLL. But VOLL is much higher (note units are $/k. Wh, to be than the price caps imposed on most markets today. And so multiplied by 1000 to get $/MWh). it may follow that price caps of ~ $3700/MWhr do distort the market. Therefore, lifting those price caps would provide a 12 market signal to incentivize capacity investment. However…

Who Ensures Adequacy in CEM? However, reference [1] argues that blackouts cannot be priced simply because, by definition, the commodity (energy) cannot be bought and sold during a blackout and thus it is not possible to establish a scarcity price during a blackout. That is, “electricity markets cannot optimize blackouts, ” and “the price that is being paid to generators during blackouts must be set by administrative rules”. [1] P. Cramton, A. Ockenfels, and S. Stoft, “Capacity market fundamentals, ” May 26, 2013, unpublished paper, available at http: //www. cramton. umd. edu/papers 2010 -2014/cramton-ockenfels-stoft-capacity-market-fundamentals. pdf This reasoning has led to the concept of the “missing money, ” which refers to the fact that the energy markets do not provide revenue sufficient to induce capacity investment, a concept that is consistent with experience of several years of real-time and day-ahead market operation, where it was observed that construction of new generation capacity was not keeping pace with generation retirements and demand growth, so that reserve margins were declining. Historical installed reserve margins (IRMs) since 1999 are indicated on the next slide for PJM and for NYISO. 13

Who Ensures Adequacy in CEM? PJM NYISO PJM Historical Installed Reserve Margins, available at www. pjm. com/~/media/planning/res-adeq/historical -pjm-installed-reserve-margins. ashx. 14

Who Ensures Adequacy in CEM? So the answer: “DART markets can provide the revenue to maintain adequacy in CEM” has some problems as evidenced by the fact that it has not worked so far (or, some would argue, it has not been allowed to work so far). Alternative answer: Capacity markets. 15

The closest thing we have to planning markets today is the capacity market. The capacity market has been motivated by the “missing money” problem, where • The real-time market price is capped so that during (rare) very high-stress time periods, prices (and the system) avoids sociallyunacceptable performance. • This results in suppliers not seeing the signal (and money) to build more capacity. • So “tight” real-time market price -caps are generally coupled with capacity markets to supply that “missing money. ” Capacity markets today, where they exist, only address generation capacity. They do not address transmission capacity. Transmission capacity, despite its close interlinkage with generation capacity, is addressed in a separate process. 16

Today’s Capacity Markets ISO Cap mrkt Number of auctions, time before delivery period, and delivery period duration [1] and other info; OR Why they don’t have cap market. Participants Recent prices ($/MWday) [1] MISO Yes Single auction 2 mnths before 1 -yr delivery period. OMS says resource Existing power plant owners 2 NYISO Yes Seasonal auction, monthly auction, final auction; from 6 mnths to few days before 1 -mnth delivery period Existing power plant owners 73 -328 PJM Yes Single auction 3 yrs before 1 -yr delivery period Existing power plant owners and project developers 77 -188 ISONE Yes Single auction 3 yrs before 1 -yr delivery period Existing power plant owners and project developers 234 CAISO No Bringing cap mrkt in spurred by high wind/solar increase+concern for uneconomic gas units [4], but Cal legislatively-mandated long-term capacity procurement plan [4]. Cal lets gas self-schedule to keep their capacity [5]. NA NA ERCOT No Enrgy capped $9 k/mwh instead of ~$2 k in other energy mrkts: scarcity price NA NA SPP No SPP lets coal self-schedule to keep their capacity [5]. NA NA adequacy within MISO is state/local responsibility; unlike other Eastern Interconnection RTOs, MISO is composed of traditional vertically-integrated utilities subject to state/local regulation; OMS members have jurisdiction over type/amount of gen constructed within their boundaries by utilities they regulate & costs recovered by those utilities [2]. Also see MISO BPM 011 [3]. [1] US Government Accountability Office, GAO-18 -313, “Electricity Markets: Four Regions Use Capacity Markets to Help Ensure Adequate Resources but FERC Has Not Fully Assessed Their Performance, ” Dec. , 2017, www. gao. gov/assets/690/688811. pdf. [2] G. Bade, “FERC rejects generator proposal for CAISO capacity market” Utility Dive, Nov. 21, 2018, www. utilitydive. com/news/ferc-rejects-generator-proposal-for-caiso-capacity-market/542833/. [3] MISO Business Practice Manual BPM 11, “Resource Adequacy. ” See section 5. 5, “Planning Resource Auction. ” https: //cdn. misoenergy. org//BPM%20011%20 -%20 Resource%20 Adequacy 110405. zip. [4] Organization of MISO States, “State Regulatory Sector Response September Hot Topic on Resource Adequacy, ” Sept, 2016, www. misostates. org/images/stories/Filings/Hot. Topics/2016/Item_7_OMS_Hot_Topic_Comments_FINAL. pdf. [5] J. Gheorghiu, “Capacity pricing changes: how each power market plans to account for resource adequacy, ” Deep Dive, Dec. , 2018, https: //www. utilitydive. com/news/capacity-pricing-changes-how-each-power-market-plans-to-account-for-resour/542449/. 17

View of Today’s Electricity Market Systems FINANCIAL MARKETS (a side comment): “Like other commodities, wholesale electricity is transacted both physically and traded financially. And like other financially traded commodities, specialized environments, such as exchanges and electronic trading platforms, have evolved to facilitate financial trading. For instance, financial electricity is traded on the New York Mercantile Exchange (“NYMEX”), the Intercontinental Exchange (“ICE”) and Nodal Exchange. These exchanges offer futures, options and swaps to trade electricity specific to PJM and at multiple locations (or nodes) on the PJM system. These so-called “secondary markets” in PJM electricity are not regulated by the FERC. They are separate from PJM’s FERC-regulated markets and affect PJM’s markets only very indirectly. While these secondary financial markets are not the subject of today’s hearing, I raise them only to clarify that highly developed, highly liquid and specialized forums exist for those that wish to hedge or speculate on PJM electricity prices outside of the PJM market itself. PJM’s markets are fundamentally designed to facilitate the dispatch, purchase, sale and delivery of physical electricity from power plants to wholesale electricity buyers, who in turn sell retail electricity to homes and businesses. ” - V. Duane, VP Compliance 7 External Relations, PJM, “Examining the role of financial trading in the electricity markets, ” Nov. 29, 2017, in testimony to the US House of Representatives Committee on Energy& Commerce/Subcommittee on Energy. www. pjm. com/-/media/library/reports-notices/special-reports/20171129 -duane-testimony-to-house-energy-subcommittee-on-financial-trading. ashx. Capacity Market Process & Resource Adequacy Evaluation The picture to the left is for PJM, but other ISOs look the same, with exception of the capacity market. Transmission Planning Process (RTEP in PJM; MTEP in MISO) Generatio n Queue Proposed generation projects 18

Capacity Market Called “Planning Resource Auction” in MISO 19

Capacity Market Called “Planning Resource Auction” in MISO Sys. New. Cap. Z Existing Capacity Bidding In CONE=cost of new entry Zone. New. Cap. Z New Capacity How does MISO know what “Sys. Required. Cap” and “Required. Capz” should be? Sys. Cleared. Cap=Σz Cleared. Capz Sys. New. Cap=Σz Sys. New. Capz Zone. New. Capz, is capacity specific to zone z, that is not subject to export. 20

Capacity Market Called “Planning Resource Auction” in MISO Answer: How does MISO know what “Sys. Required. Cap” and They evaluate reliability indices “Required. Capz” should be? using resource adequacy software. LOLE (loss of load expectation) is the amount of time during a planning period the system can expect to interrupt load. Industry norm: LOLERequired≤ 1 day in 10 years So MISO identifies Sys. Required. Cap and Required. Capz to satisfy this requirement. See http: //home. engineering. iastate. edu/~jdm/ee 653 schedule. htm for more info on reliability eval. 21

Capacity Market Called “Planning Resource Auction” in MISO Where do the revenues come from to fund a capacity market? An auction is conducted for suppliers using a specified demand curve, illustrated below. The demand curve is determined based administratively, based on the cost of new entry (CONE). T. Jenkin, P. Beiter, and R. Margolis, “Capacity payments in restructured markets under low and high penetration levels of renewable energy, ” NREL Technical Report NREL/TP 6 A 20 -65491, Feb, 2016, available https: //www. nrel. gov/docs/fy 16 osti/65491. pdf. Capacity obligations are determined by a LSE’s peak load contribution (PLC) during a certain timeframe. In MISO, an LSE PLC is determined by their usage during the peak hour from the previous year. The peak hour is the hour during which the usage was the highest across the ISO. The LSE is charged the market clearing price × PLC. 22 https: //business. directenergy. com/understanding-energy/managing-energy-costs/deregulation-and-energy-pricing/capacity-markets

Three additional questions that capacity markets do not answer… • • • How do vertically-integrated rate-regulated utilities, under traditional regulation, know what kind of technologies to build? How do market participants know what “Offer Price” to submit and how do they know what kind of generation to build? How can ISO’s forecast generation builds beyond what is in the interconnection queue? Solutions to the Generation Expansion Planning (GEP) problem can contribute to answering these questions. 23

Generation Expansion Planning (GEP, e. g. , EGEAS or Electric Generation Expansion Analysis System) i: bus number j: technology t: time s: load conditions T: final time subject to Updates existing capacity after each yr Initializes year 0 capacity Updates retirements after each yr Power balance at each node MW at each gen limited by capacity Non-negativity on decision variables System reserve constraint Constraint on energy generated Transmission limitations Salvage Value calculations, used in objective function (manipulation eliminates nonlinearity) 24

Generation Expansion Planning (GEP) What input data does this problem require? • Existing generation fleet: • Demand data • • Cost data: Fuel cost, Heat rate, O&M Capacity credit Capacity factor Age, expected life, salvage value • Data for investment options: • • Data required by existing fleet Investment cost • • • Growth rate Load duration curve & slices/yr Reserve requirements Transmission system data Discount rate Simulation time What does the solution to this problem give you? • • • 400 MW 2000 MW 100 MW What to build NGCC, Wind, Nuke, Solar, Where bus 34 bus 98 bus 62 bus 12 When How much Net present value 1 2 3 4 5 6 7 8 20 21 22 23 24 25 of total cost. • Net present value or levelized cost for specific capacities. 25

Net Present Worth The objective function includes all costs of building, retiring, and operating the generation resources over the time period t=1, …, T. Discount rate i: annual income as a percentage of amount invested Discount factor: The discount factor is given by ζ=1/(1+i) where i is the discount rate. Thus we have that We assume the investments made in year 1 are already present value, and so it is not until year 2 that we need to discount to present worth; therefore we utilize ζt-1 as the discount factor. 26

Salvage Value A key issue for multi-period planning is what is referred to as end effects. End effects refer to the difficulty of appropriately representing the influence of investment costs and operational costs at the end of the planning period. There are two problems to address: • Remnant investment value: The retirement year for some facilities occurs after the final year. For these facilities, there is a value to the facility because it has remaining life, i. e. , some of the investment amount paid has not yet been depreciated. • Remnant operational cost: Because the simulation must be truncated at a particular final year, operational costs after that final year are not included since those years are not simulated. Both of the above effects tend to bias decisions in favor of alternatives that have low investment costs, since ignoring the remnant investment value and the remnant operational cost means the optimization sees only the investment cost, and because it is minimizing costs, it chooses the investments with the lowest investment costs. 27

Salvage Value These issues are addressed by doing two things: Extend the simulation time beyond the planning horizon: A multiperiod formulation with extended planning horizon is based on the observation that end effects increase their influence as the final year gets closer. For example, if the planning horizon is 20 years, end effects have more influence in years 5 -20 than they do in years 1 -5. This observation leads to a very natural solution: extend the final year well-beyond the planning horizon. Therefore if the planning horizon is 20 years, we may run the optimization problem to 50 years; but the decisions for years 20 -50 will be ignored. A general guide for how far to extend the simulation beyond the planning horizon is that the final year of the simulation should exceed the final year of the planning horizon by the lifetime of the facility with the longest life. Model salvage values to facilities: Salvage value is the net sum to be realized from the disposal of an asset (net of disposal costs) at the time of its replacement or resale, or 28 at the end of the study period.

Capacity Factor Constraint on energy generated The capacity factor of a resource identifies the actual annual energy production as a percentage of annual energy production at the resource’s capacity. The capacity factor for wind is usually 0. 4 -0. 5 in areas of very high wind conditions, 0. 3 -0. 4 in areas of medium wind conditions, and less than 0. 3 in areas of low wind conditions. Capacity credits for solar are lower, from 15 -40%, since we do not get any solar energy during about half the time (when it is night), and we can get greatly reduced solar energy even during the day if it is cloudy. 29

Capacity Credit MW at each gen limited by capacity System reserve constraint The capacity credit (or value) of a resource identifies the percentage of the resource’s capacity to be identified for reliability calculations at peak load. The capacity credit indicates the availability at peak load. The capacity credit for wind is usually fairly low because wind generation during daytime hours is usually lower. For example, MISO was using 13. 3% capacity credit for wind [1]. Capacity credits for solar are higher, from 40 -95%, because peak loads usually occur during the daytime when solar insolation is highest. [1] See “Planning year 2013 -2014: Wind Capacity Credit, ” https: //www. midwestiso. org/Library/Repository/Study/LOLE/2013%20 Wind%20 Capacity%20 Report. pdf 30

Demand Representation One might consider to choose the demand d to be either peak load or average load. However… • If demand d is chosen to be the peak load, then the GEP will build the right amount of capacity but will over-estimate the energy requirements and corresponding generation production. • If demand d is chosen to be the average demand, then the GEP will identify the right energy requirements and corresponding generation production but will underestimate the capacity. In both cases, the solution will identify the one technology that minimizes the objective function. An improvement on this formulation is to increase the number of load levels, or load blocks, so that we use ds, s=1… instead of just d. In doing this, we must also identify the appropriate duration hs for each load block. 31

Demand Representation i: bus number j: technology t: time s: load conditions T: final time Power balance at each node MW at each gen limited by capacity Constraint on energy generated Transmission limitations These constraints represent operating conditions for a given demand level. The load duration curve provides on the abscissa the number of hrs the load is expected to be greater than or equal to the corresponding load given as the ordinate. Only 3 load blocks are shown, but the formulation generalizes to any desired number of load blocks. 32

What is coordinated expansion planning (CEP)? If G only, then it is GEP. If T only, then it is TEP. If G&T or G&T&D, then it is CEP. 33

Adequacy We could add a constraint on adequacy in our formulation, e. g. , LOLE≤ 0. 1 day/year. However, computing LOLE is computationally intensive. A better way is to iterate between an adequacy evaluation program (e. g. , GE-MARS) and a GEP. Expansion Planning Optimization Adequacy Evaluation e. g. , Multiarea Reliability Analysis Adjust (e. g. , increase reserve requirements) LOLE≤ 0. 1? YES Stop NO 34

Definition: Long-Term Planning Process “The planning process is the systematic assembly and analysis of information about electric energy supply, transport, and demand, and the presentation of this information to decisionmakers who must choose an appropriate course of action. ” “The plan is a statement of the choices made by decisionmakers at any one point in time in order to meet specific goals and objectives. ” REF: International Atomic Energy Agency, “Expansion planning for electrical generation systems, ” Technical Reports Series No 241, 1984. 35

Long-Term Planning Process: Illustration GEP Vertically-integrated utilities develop data input, perform analysis, and make decisions. RTOs obtain the data from stakeholders, then perform the analysis. But they do not make the investment decisions; they do not build/own equipment. A. Gaikwad and K. Carden, “Probabilistic Risk Analysis: A consideration of risks in transmission planning, ” presentation slides, Eastern Interconnection States Planning Council (EISPC) Meeting, October 8, 2013. 36

The Long-Term Planning Process at MISO RTOs coordinate the regional planning processes. Gen. Cos solve the GEP problem (see previous slide) and then enter their plans into the RTO’s generation interconnection queue. 37

The Long-Term Planning Process at MISO 38

High-Fidelity Production Cost Simulation (PROMOD or PLEXOS) hr 1 Must initialize hour 1, day k with hour 24, day k-1 A 24 hour period Run SCUC on 24 hr period to decide unit commitment Run SCED on each hour of 24 hr period to decide dispatch, flows, and LMPs hr 8760 39

The Long-Term Planning Process at MISO SERVM Capabilities: • • Probabilistic Production Cost Forecasts: Produce S Curve Analysis for future production cost years for any number of zones Probabilistic Market Price Forecasts: Produce S Curve Analysis for future energy price forecasts for any number of zones Probabilistic Energy Margin Forecasts: Produce S Curve Analysis for future energy margins by resource dispatch price for any number of zones Physical Reliability Metrics: Produce Loss of Load Expectation (LOLE), Loss of Load Hours (LOLH), Expected Unserved Energy (EUE) for any number of zones Intermittent Penetration Studies: Understand reliability impact of increased intermittent generation Intermittent ELCC Studies: Calculate the Effective Load Carrying Capability of intermittent resources on the system System Flexibility Requirement Studies: Calculate the flexible resource needs for a system under varying 40 intermittent penetration levels

Financial Transmission Rights (FTRs) Capacity Market Process & Resource Adequacy Evaluation The picture to the left is for PJM, but other ISOs look the same, with exception of the capacity market. Transmission Planning Process (RTEP in PJM; MTEP in MISO) Generatio n Queue Proposed generation projects FTR Market 41

Financial Transmission Rights (FTRs) Recall from slide 42 of LPOPF slides: The load pays into the market an amount exceeding the amount gens are paid by the congestion charges. These charges are the sum of products of line constraint dual variables and the RHS of the constraint. Let’s check that in the example we just did (next slide). These charges are used to fund the FTR market. 42

Financial Transmission Rights (FTRs) Financial transmission rights (FTRs), defined between any two nodes in a network, entitle their holder to a revenue equal to the product of the amount of transmission rights bought and the congestion price differential between the two nodes. FTRs isolate their holders from the risk associated with congestion in the transmission network when the holders make bilateral transactions. They are sold via an commodity market. 43

Financial Transmission Rights (FTRs) 44

Financial Transmission Rights (FTRs) 45