Presentation: Achieving 50 Percent Renewable Electricity in California

Achieving 50 percent renewable electricity in California: The role of non-­‐fossil flexibility in a cleaner electricity grid August 2015 Jimmy Nelson, Ph.D. Energy Modeler jnelson@ucsusa.org Laura Wisland Senior Energy Analyst lwisland@ucsusa.org Full report available at: www.ucsusa.org/California50RPSanalysis 2 “Curtailment is a reducOon in the output of a generator from what it could otherwise produce given available resources” Source: Bird, L., J. Cochran, and X. Wang. 2014. Wind and solar energy curtailment: Experience and pracOces in the United States. Golden, CO: NaOonal Renewable Energy Laboratory. NREL/TP-­‐6A20-­‐60983. Online at hcp://www.nrel.gov/docs/fy14osO/60983.pdf. • 
Curtailment is a default opOon for maintaining reliability when an electricity grid relies more heavily on renewable sources of power • 
Curtailment can be expensive because most renewables have low or zero marginal cost • 
In limited quanOOes, curtailment can make important contribuOons to grid operaOonal flexibility, reducing the need to make other investments in flexibility 3 Avoid coincident gas
generation and renewable
curtailment
4 The UCS “Duck Chart” Fact Sheet: Renewables and Reliability, Union of Concerned ScienOsts, March 2015. Available at: hcp://www.ucsusa.org/sites/default/files/acach/2015/03/california-­‐renewables-­‐and-­‐reliability.pdf 5 It would be very difficult to meet long-­‐term climate goals if gas is online when renewables are being curtailed Williams, J.H., et al., 2012. The technology path to deep greenhouse gas emissions cuts by 2050: the pivotal role of electricity. Science 335, 2012, 53-­‐59. 6 Modeling 7 Acknowledgements •  Ryan Jones, Arne Olson, and Energy and Environmental Economics (E3) •  Hourly renewable, reserve, outage, and demand data •  ProducOon cost model framework •  Discussions •  UCS for support, fellowship, and discussions •  NREL for discussions •  Energy Exemplar / PLEXOS for sojware support •  Report reviewers 8 Scope of analysis •  Renewables Porkolio Standard (RPS) target: • 
33%, 40%, 50% •  Geographic scope: the California Independent System Operator (CAISO) footprint • 
~80% of California •  Timeframe: 2024 •  Economy-­‐wide decarbonizaOon might necessitate very quick renewable deployment •  Results relevant to 2030 50% RPS policy discussion •  Modeling tool: PLEXOS producOon cost model • 
• 
Focuses on electricity operaOons NOT an investment model 9 What our analysis does and does not address Our analysis does: Our analysis does not: Focus on the year 2024 Simulate years other than 2024 Use industry standard sojware to simulate the CAISO electricity system Simulate other sectors of the economy Include electricity demand from 2 million electric vehicles within the CAISO footprint Determine an opOmal amount of vehicle electrificaOon Include a diverse porkolio of renewables OpOmize a porkolio of renewables or focus on a single renewable technology Simulate detailed grid operaOons and constraints on generators and resources inside the CAISO footprint Simulate detailed grid operaOons and constraints on generators and resources outside the CAISO footprint Include generaOon requirements in specific regions Model transmission constraints inside the CAISO inside the CAISO footprint footprint Explore a variety of ways to reduce renewable curtailment Find an opOmal amount of renewable curtailment or grid flexibility QuanOfy renewable curtailment, producOon costs, and GHG emissions from electricity producOon QuanOfy capital and fixed costs of renewable generators or addiOonal grid flexibility QuanOfy shorkalls in generaOon capacity and reserves Perform an analysis of system capacity need or loss of load expectaOon 10 Electricity Produc>on (TWh/Yr) A diverse porkolio of renewables is simulated 120 PV Behind-­‐the-­‐Meter PV Large Out-­‐of-­‐State PV Large In-­‐State Solar Thermal Wind Out-­‐of-­‐State Wind In-­‐State Small Hydroelectric Biogas Biomass Geothermal 100 80 60 40 20 0 33% RPS 40% RPS 50% RPS •  Heavily weighted towards in-­‐state renewables •  Behind-­‐the-­‐meter photovoltaics (PV) don’t directly count towards RPS 11 As the RPS is increased… [without addiOonal operaOonal flexibility] ProducOon Costs 50 40 30 20 10 0 33 40 50 RPS Target (%) 7 6 5 4 3 2 1 0 Renewable Curtailment 6 % Curtailment 60 Produc>on Costs ($B/Yr) CO2 Emissions (MMTCO2/Yr) GHG Emissions 33 40 50 RPS Target (%) 4.8 5 4 3 2 1 0 33 40 50 RPS Target (%) Note: producOon costs do not include capital and maintenance costs, and are therefore only one important part of the total electricity cost 12 Genera>on (GWh) OperaOonal constraints keep gas online during hours that experience renewable curtailment Coincident gas generaOon and renewable curtailment 35 30 25 20 15 10 5 0 -­‐5 -­‐10 -­‐15 -­‐20 Demand Adjusted Input Supply Renewable Storage Supply Exports Gas Peaker Gas Combined Cycle Hydroelectric Combined Heat and Power Nuclear 0 2 4 6 8 10 12 14 16 18 20 22 Hour of Day Downward Balancing Renewable Curtailment Storage Demand Exports Example day depicted here is a near worst-­‐case weekend day in May 13 3.5 33% RPS 3 40% RPS 2.5 50% RPS Base 2 Non-­‐Spinning Load Following Up 1.5 1 July Sept July Sept 17 18 17 17 18 19 16 17 16 17 18 19 15 16 17 18 19 0 15 16 17 18 19 0.5 14 15 16 17 18 19 Reserve ShorPall (GW) Renewables can provide some electricity during Omes of greatest need for system capacity Hour of Day Sept •  However, even at a 50 percent RPS, renewables cannot produce enough power to ensure a reliable electricity system during some peak periods, resulOng in reserve shorkalls •  Our study suggests that the projected 2024 CAISO fleet of natural gas power plants will be important in ensuring that the grid has enough capacity to meet demand during peak periods 14 Reliability requirements cause renewable curtailment 15 EssenOal reliability services needed in grid operaOons [But some reliability requirements cause curtailment at a 50% RPS] Curtailment (%) 6 5 4.8 4 Load Following and Regula>on DOWN 3 1.5 2 1 1.0 0 50% RRPS PS 50% Base Remove Downward Reserves Remove Downward Reserves and Regional GeneraOon Requirements 16 Load following bridges between hourly and five minute schedules Net demand (MW)
Load following up
Five-minute
schedule
Five
minutes
Load following down
Hourly schedule
One hour
Figure based on: Makarov, Y. V., et al., 2009. OperaOonal Impacts of Wind GeneraOon on California Power Systems. IEEE Transac/ons on Power Systems 24 (2), 2009, 1039-­‐1050. 17 RegulaOon bridges between five minute schedules and actual generaOon Net demand (MW)
Regulation up
Five-minute
schedule
Five
minutes
Regulation down
Actual
generation
One hour
Figure based on: Makarov, Y. V., et al., 2009. OperaOonal Impacts of Wind GeneraOon on California Power Systems. IEEE Transac/ons on Power Systems 24 (2), 2009, 1039-­‐1050. 18 Regional generaOon requirements are a proxy for many essenOal reliability services These reliability consideraOons could cause coincident renewable curtailment and gas generaOon: ² 
² 
² 
² 
² 
voltage support and reacOve power inerOa and primary frequency response/governor response local and system-­‐wide requirements for operaOng reserves transmission congesOon transmission conOngencies Grid planners and operators should encourage non-­‐fossil resources (including renewable generators) to provide essenOal reliability services 19 Increasing operaOonal flexibility at a 50% RPS: What works and what doesn’t 20 We modeled three different flexibility strategies [1] Operate renewables more flexibly [2] Increase natural gas power plant fleet flexibility [3] Add non-­‐generaOon flexibility: Storage Advanced Demand Response Exports 21 Strategy [1]: Operating renewables flexibly is
particularly effective
•  Curtailment: nimble is much becer than blocky •  Renewables should be incenOvized to parOcipate in CAISO markets and to install or uOlize control equipment Curtailment (%) 6 5 4.8 4 2.7 3 2 Hourly Curtailment Sub-­‐Hourly Curtailment 1 0 50% RRPS PS 50% Base Renewables Provide Reserves 22 Renewables frequently provide downward
reserves
Upward 5 4 3 2 1 0 -­‐1 -­‐2 -­‐3 -­‐4 Downward Average Online Reserve Provision (GW) •  Renewables don’t have to be pre-­‐curtailed to provide downward reserves 50% RPS Base Renewable Upward Storage Upward Hydro Upward Gas Upward Renewables Provide Reserves Online reserves: Load following, regulaOon, and spinning 23 Strategy [2]: Non-­‐generaOon flexibility reduces curtailment and GHG emissions 41.1 5 50 38.8 37.6 37.0 4 40 30 3 20 2 1 10 0 0 0 3 6 CO2 Emissions (MMTCO2/Yr) % Curtailment 6 9 AddiOonal Non-­‐GeneraOon Flexibility (GW) = storage + advanced demand response + exports •  More analysis needed on cost tradeoffs 24 37% 56% 67% 75% % Non-­‐Fossil Upward 5 4 3 2 1 0 -­‐1 -­‐2 -­‐3 -­‐4 Downward Average Online Reserve Provision (GW) Reserves from non-fossil resources are valuable 0 3 6 Demand Response Upward Storage Upward Hydro Upward Gas Upward 9 Addi>onal Non-­‐Genera>on Flexibility (GW) •  Downward reserves are parOcularly valuable 25 Strategy [3]: Gas power plant flexibility
has important limits
•  Providing some reliability services with gas requires electricity producOon, which “crowds out” renewables 6 % Curtailment 5 4 3 4.8 4.7 3.2 3.0 2 1 0 Default Gas Double Combined Flexible Gas Flexibility Ramp Rate Cycle Half Minimum Power Level Gas fleet changes in the Flexible Gas run: •  Double ramp rate •  Half minimum power level •  1 hour start and stop Omes •  2 hour minimum upOme and downOme •  Duck Chart: “belly” is much more important than the “neck” 26 Renewable flexibility competes with gas flexibility
•  When renewables can provide reserves, the amount of operaOonal flexibility on the grid increases. Consequently, the amount of curtailment that can be avoided by increasing the flexibility of natural gas power plants decreases 6 % Curtailment 5 4.8 Without Renewable Reserves With Renewable Reserves 4.7 4 3 2 2.7 2.6 3.2 3.0 2.1 2.0 1 0 Default Gas Double CCGT Half Flexible Gas Flexibility Ramp Rate Minimum Power Level 27 CO2 Emissions (MMTCO2 /Yr) % Curtailment OperaOonal flexibility: Non-­‐fossil can go further than gas 6 5 4 3 2 1 0 50 40 30 20 10 0 4.8 3.0 1.1 Flexible 50% 50% RRPS PS Non-­‐Fossil Gas SoluOons Base 39.0 Flexible Gas 41.1 37.4 50% RPS Non-­‐Fossil SoluOons Base Non-­‐Fossil SoluOons: •  Renewables can provide reserves •  1 GW addiOonal electricity storage •  1 GW advanced demand response •  1 GW net exports allowed 28 Genera>on (GWh) Non-­‐fossil soluOons can turn gas down or off when ample renewable energy is available Demand Adjusted Input 35 30 25 20 15 10 5 0 -­‐5 -­‐10 -­‐15 -­‐20 Supply Renewable Storage Supply Demand Response Supply Gas Peaker Gas Combined Cycle Hydroelectric Combined Heat and Power Nuclear 0 2 4 6 8 10 12 14 16 18 20 22 Hour of Day Downward Balancing Renewable Curtailment Storage Demand Demand Response Demand Exports Example day depicted here is a near worst-­‐case weekend day in May 29 At a 50% RPS, renewables should help to integrate themselves by being dispatchable during criOcal Omes 33% RPS 50% RPS Renewable curtailment Renewable dispatchability 30 Key findings •  Increasing the RPS from 33% to 50% reduces electricity GHG
emissions by 22 - 27%
•  Coincident gas generation and renewable curtailment should
be avoided
•  Reliability requirements cause renewable curtailment
•  Operating renewables flexibly is particularly effective
•  Non-generation flexibility can reduce GHG emissions
•  Natural gas power plant flexibility has important limits
31