Western Interconnection Flexibility Assessment Update to TEPPC May 7, 2015 Arne Olson, Partner.

Western Interconnection Flexibility Assessment Update to TEPPC May 7, 2015 Arne Olson, Partner

22 About the Study WECC and WIEB are interested in understanding power system reliability and flexibility needs under higher renewable penetration There have been many stakeholder requests for studies to help understand the operational issues associated with higher renewables WECC is interested in understanding and gaining experience with new reliability planning tools WECC and WIEB engaged E3 and NREL to study operational needs Funding for E3 work from WECC and WIEB (through ARRA) Funding for NREL work from DOE

33 Study Goals Assess the ability of the fleet of resources in the Western Interconnection to accommodate high renewable penetration while maintaining reliable operations Quantify the size, magnitude and duration of operating challenges resulting from high renewable penetration Investigate potential flexibility solutions, including: Renewable dispatch as an operational strategy Regional coordination Flexible supply and demand-side resources Energy storage Transmission Learn about how to do flexibility modeling and planning Institutional solutions Physical solutions

44 Project Team Partnership between WECC, WIEB, NREL and E3 WECC and WIEB provide general project oversight role E3 directs technical work NREL provides data, HPC resources, technical support, and insights from previous experience E3 Arne Olson, Partner Nick Schlag, Project Manager Dr. Elaine Hart Ryan Jones Ana Mileva E3 Arne Olson, Partner Nick Schlag, Project Manager Dr. Elaine Hart Ryan Jones Ana Mileva NREL Bri-Mathias Hodge, Project Manager Carlo Brancucci Martinez-Anido Greg Brinkman NREL Bri-Mathias Hodge, Project Manager Carlo Brancucci Martinez-Anido Greg Brinkman WECC Dan Beckstead Vijay Satyal Donald Davies Branden Suddeth WECC Dan Beckstead Vijay Satyal Donald Davies Branden Suddeth WIEB Tom Carr Maury Galbraith WIEB Tom Carr Maury Galbraith

55 Project Elements 1.Renewable energy capacity value analysis Effective load-carrying capability (ELCC) method using E3s Renewable Energy Capacity Planning Model (RECAP) 2.Power system flexibility analysis Regional model using E3s Renewable Energy Flexibility (REFLEX) model for PLEXOS run on NRELs High Performance Computing environment 3.Stakeholder participation opportunities 1.Technical review committee 2.Executive advisory group

66 Cases Studied 2024 Common Case Few reliability or flexibility issues anticipated Primary purpose of case is calibration 2024 High Renewables Case Renewable penetration that is high enough to show interesting operational challenges Composition of case now being determined in consultation with technical review committee

77 Regional Focus Study focuses on five regions Explicitly control interactions between regions through modeling of interties Most regions share characteristics appropriate for a resource planning study: Similar weather and load patterns across the region Limited internal transmission constraints Some degree of regional coordination already Limited reliance on other regions

8 RECAP MODEL RESULTS

99 E3 Renewable Energy Capacity Planning Model (RECAP) Flexibility Assessment utilizes RECAP, E3s non-proprietary model for evaluating power system reliability and resource capacity value under high renewable penetration Initially developed to support CAISO renewable integration modeling Used by a number of utilities and state commissions Will be transferred to WECC as part of study process

10 Calculating LOLP LOLP is determined by comparing the distributions of potential load and resource states and calculating the probably that load exceeds generation Gross load distribution Net thermal generation distribution LOLP comes from the chance that net load exceeds net thermal generation Gross load Net thermal generation LOLP

11 Adding Renewables After adding renewables to the system, net loads are reduceddistribution shifts to left LOLP decreases in every hour (nearly) Gross load distribution Net thermal generation distribution Net load distribution with renewables Renewable net load Gross load Thermal generation Reduction in LOLP with increase in renewables

12 Calculating ELCC Since LOLE has decreased with the addition of renewables, adding load will return the system to the original LOLE The amount of load that can be added to the system is the effective load carrying capability (ELCC) Original system LOLE LOLE after renewables Additional load to return to original system LOLE = ELCC

13 Portfolio vs. Marginal ELCC Values 1.The cumulative portfolio capacity value is used for resource adequacy planning Due to the complementarity of different resources the portfolio value will be higher than the sum of each individual resource measured alone May need to attribute the capacity value of the portfolio to individual resources There are many options, but no standard or rigorous way to do this 2.The marginal capacity value, given the existing portfolio, is used in procurement Provides a measure of the value of the next resource to be procured This value will change over time with the mix of system needs & resources Individual Solar Capacity Value Individual Wind Capacity Value Combined Capacity Value

14 Reliability Metrics The RECAP model calculates conventional power system reliability metrics: Loss of Load Probability (LOLP) Loss of Load Expectation (LOLE) Loss of Load Frequency (LOLF) Expected Unserved Energy (EUE) RECAP also calculates effective capacity of renewables, demand response, and other dispatch- limited resources: Effective Load Carrying Capability (ELCC) LOLE Marginal ELCC Cumulative ELCC

15 Renewable penetration in the Common Case is approximately 20% of load (U.S. portion); wind and solar serve approximately 13% of load: E3 and NREL have developed production profiles to reflect the operational characteristics of these resources Common Case Renewable Mix

16 Target Planning Reserve Margins RECAP estimates reserve margins needed to achieve a target reliability threshold LOLF = 1 event in 10 years Target PRM needed to meet standard varies by region Common Case above Target PRM for all regions TypeTarget PRM Common Case PRM Basin14%17% California13%25% Northwest15%32% Rockies17%19% Southwest15%32%

17 Marginal ELCC Curves by Technology and Region SouthwestCalifornia Marginal ELCC = capacity contribution of next increment of capacity of a given type Curves are illustrative they assume a single technology

18 Marginal ELCC Curves by Technology and Region (Cont.) BasinRockies Northwest

19 Observations on ELCC Values Marginal ELCC of solar PV at low penetrations is 50-60% of nameplate capacity (except in NW) Aligns well with commonly used heuristics At low penetrations, marginal ELCC values for wind range from 15-30% of nameplate capacity Slightly higher than common heuristics ELCC values exhibit significant diminishing returns to scale, particularly solar PV which shifts net load peak into the evening As penetration increases, heuristics become increasingly inaccurate

20 Effect of Diversity on ELCC Values For a diverse portfolio, ELCC of combined portfolio is higher than individual ELCC values At 20% of load: W = 3041, S = 6172, W+S = 12,861

21 Ongoing uses for RECAP by WECC The need for ELCC in WECCs planning studies will increase as the penetration of variable generation increases As part of the flexibility assessment project, the RECAP model will be transferred to WECC staff to help support modeling efforts TEPPC Common Case development Summer load assessments Section 111(d) impacts As an open-source tool, RECAP can also be shared with or modified by stakeholders

22 REFLEX MODEL STATUS

23 E3s Renewable Energy Flexibility (REFLEX) Model REFLEX answers critical questions about flexibility need through stochastic production simulation Captures wide distribution of operating conditions through Monte Carlo draws of operating days Illuminates the significance of the operational challenges by calculating the likelihood, magnitude, duration & cost of flexibility violations Assesses the benefits and costs of investment to avoid flexibility violations Implemented as an add-on to Plexos for Power Systems

24 WECC Flexibility Assessment Distinguishing Characteristics The use of REFLEX for PLEXOS in this study is different from conventional production cost modeling of the WECC in several important respects: 1.Economic tradeoff between upward (loss of load) and downward (curtailment) flexibility violations 2.Endogenous determination of load following reserves as function of expected within-hour flexibility deficiencies 3.Stochastic sampling of load, wind, solar, and hydro conditions 4.Sub-regional study footprints with specified boundary conditions Import/export limitations and maximum ramp rates

25 Renewable Dispatch is Used to Solve Upward Ramping Shortages Model needs robust information on cost of upward vs. downward shortages Cost of unserved energy due to ramping shortfall: very high ($5,000-50,000/MWh) Cost of renewable dispatch: replace the lost production ($50-$150/MWh) Limited Ramping Capability Unserved Energy Limited Ramping Capability Renewable Curtailment Strategy to Minimize Downward ViolationsStrategy to Minimize Upward Violations

26 Stochastic Sampling From a Range of Conditions In order to ensure robust sampling results, RECAP and REFLEX sample from a broad range of load, wind, & solar conditions Historical data matched up based on month of year, day type (i.e. load level) Range of available data

27 Capturing Transmission in Flexibility Assessment Multiple options for representing interregional power flows have been tested 123 Original project plan Single-Zone Models Each region modeled independently with no internal transmission Imports and exports captured through supply curves Offers simplest modeling framework, but difficult to represent interregional power exchange Single-Zone Models Each region modeled independently with no internal transmission Imports and exports captured through supply curves Offers simplest modeling framework, but difficult to represent interregional power exchange Zonal Model Loads and resources grouped together by region Regions linked together by transport model Provides macro level view of interregional power exchange, but ignores individual line and path flow limits Zonal Model Loads and resources grouped together by region Regions linked together by transport model Provides macro level view of interregional power exchange, but ignores individual line and path flow limits Nodal Model All nodes in WECC (25,000) represented Dispatch solution is constrained by DC OPF and enforced line limits Provides greatest fidelity of transmission system, but requires significant development and is computationally intensive Nodal Model All nodes in WECC (25,000) represented Dispatch solution is constrained by DC OPF and enforced line limits Provides greatest fidelity of transmission system, but requires significant development and is computationally intensive Model Complexity Final project plan

28 Zonal Topology Zonal topology chosen based on aggregations of interregional WECC paths California Northwest Basin Rocky Mountain Southwest NW to CA P65 P66 CA to BS P24 P28 P29 SW to CA P46 RM to SW P31 BS to SW P35 P78 P79 All Paths P14: ID to NW P18: NT - ID P24: PG&E - Sierra P28: Intermtn - Mona P29: Intermtn Gonder P30: TOT 1A P31: TOT 2A P35: TOT 2C P38: TOT 4B P46: WOR P65: COI P66: PDCI P76: Alturas Project P78: TOT 2B1 P79: TOT 2B2 P80: MT SE BS to NW P14 P18 P76 P80 BS to RM P30 P38

29 Regional Ramping Constraints Production simulation models tend to overstate ramps on interties compared to historical levels Constrained case limits intertie ramps based on historical levels Example: Historical vs. Modeled Flows over Path 46 Upward Ramps Downward Ramps Source: http://www.wecc.biz/Lists/Calendar/Attachments/5076/130124_DWGMeeting_E3_Presentation-PCM.pdf Duration (hrs)

30 Strawman High Renewables Case High renewables case should have enough wind and solar generation to illuminate significant flexibility constraints Test simulations of this Strawman case, while preliminary, provide some useful insights into challenges at higher penetrations to be shared today

31 Strawman Limited-Draw Results: California Curtailment: ~6% RG Overgeneration occurs regularly and periodically, especially in the spring months Overgeneration is solar-driven and occurs in the middle of the day Hydro, pumped storage, and imports help to meet nighttime load April Day Strawman High Renewables All dispatchable plants reduce output to minimum levels Curtailment due to solar oversupply

32 Strawman Limited-Draw Results: Northwest Curtailment: ~3% of RG Overgeneration conditions occur during high hydro and/or high wind conditions Curtailment may be concentrated during nighttime or could persist through day if wind output remains high More day-to-day variability in conditions within seasons compared to regions with high solar penetration April Day Strawman High Renewables Curtailment due to simultaneous high wind & hydro conditions

33 Strawman Limited-Draw Results: Northwest Curtailment: ~3% of RG Overgeneration conditions occur during high hydro and/or high wind conditions Curtailment may be concentrated during nighttime or could persist through day if wind output remains high More day-to-day variability in conditions within seasons compared to regions with high solar penetration April Day #2 Strawman High Renewables During lower hydro conditions, high wind may not result in curtailment

34 Strawman Limited-Draw Results: Southwest Curtailment: ~3% of RG Coal plants are cycled down the middle of the day to accommodate solar, but curtailment still occurs Steep morning down-ramp and evening up-ramp of coal, hydro, and gas are challenging operational conditions April Day Strawman High Renewables Large coal ramps require further investigation Curtailment due to solar oversupply

35 Strawman Limited-Draw Results: Rocky Mountains Curtailment:

Western Interconnection Flexibility Assessment Update to TEPPC May 7, 2015 Arne Olson, Partner.

Documents

e3 wecc

study wecc

high renewable penetration

partner slide

e3 renewable energy

e3 work

renewable dispatch

higher renewable penetration