Integrating Renewables into the Power Grid

Transcription

1 Integrating Renewables into the Power Grid An Overview by Barney Speckman November 2010 Energy Solutions

2 Overview Why do we have to integrate renewables What is meant by integrating renewables How does one think about renewable integration needs What are some of the challenges What are the options to meet the system s integration needs Q&A 1

3 Why do we have to integrate renewables? Before renewables, generation was added for Low cost (coal, CCGT, hydro, geothermal) Cost/Performance Ratio (GT, Pumped Hydro), and Controllability and predictability was normally a given With the most common renewables (wind and solar) Production is variable and Production is uncertain For the first time, significant amounts of generation are being added that are not controllable and by their nature (variable and uncertain) they require a higher level of controllability and because of potential forecast error they require more resources be held in reserve 2

4 What is meant be Integrating Renewables? Integration can be thought of as the collection of steps or measures that are needed to operate the power system reliably with relatively large levels of renewables A simple way to think about it is the electric power system must still Serve customer needs and Serve them reliably Operate the system to meet the control requirements of the NERC and WECC -control the power system from the seconds to hours to day time frames Maintain efficient dispatch through control and market mechanisms to result in low costs 3

5 How does one think about integration needs Lets look at California as an example to show the integration problem and one type of analysis to find potential solutions California Legislature passed AB32 in 1996 and the voters supported its implementation with the defeat of Prop 23 in Nov Lets look at what it does and how one might think about what steps are needed to integrate the renewables envisioned by AB 32 4

6 Assembly Bill 32 AB 32 (2006) Requires California to reduce GHG emissions to 1990 levels by 2020 California Air Resources Board (CARB) has been assigned task of developing a plan to implement AB 32 CARB s proposed plan includes establishing a 33% Renewable Portfolio Standard for the California electric industry Currently a mandatory 20% RPS (by 2010) is in place for the states investor owned utilities (PG&E, SCE, SDG&E) with many Municipal Utilities voluntarily implementing RPS targets 5

7 How the 33% RPS works Would require 33% of the amount of energy sold to customers to be produced by eligible renewable resources Applies to all companies selling energy at retail in CA Sets 2020 as date to meet the 33% Standard Establishes penalties for non compliance 6

8 How might the 33% RPS be met No none knows for certain!!! Many degrees of freedom in the implementation thus many uncertainties will have to be dealt with What technologies will utilized? Where, when and what plants will be built? How will the power be delivered to customers? What other infrastructure will be needed? How will integration needs be met? Several studies have/are being conducted 7

9 Possible 33% Renewable Futures Recent CPUC study used to consider several possible futures as a way to bracket the future implementation of the 33% RPS Examines cost, difficulty, GHG reductions for several mixes of technology, infrastructure requirements and integration requirements Looks at: 33% Reference 27.5% Reference High Wind High Imports High Distributed Generation 8

10 Possible 33% Futures (Cont d) Energy and Capacity Requirements based upon CPUC Forecast compared to 2007 Renewable Portfolio Standard Additional Energy Required Additional Renewable Capacity Required (Approx) 20% RPS 35 TWh 10,000 MWs 33% RPS 75 TWh 22,000 MWs 2007 Renewable Energy was 27 TWh TWh = Watt-hours 9

11 Lets Look at the Technologies in California Major contributors (potentially) to new production Wind Solar Thermal Solar PV (utility and customer) Geothermal Lesser contributors to new production Biomass Biogas Small Hydro Large percentage of new renewables in California are in Southern part of the state 10

12 Renewable Portfolios: Incremental (MW) and Existing Renewables (MWh) for Cases Studied - Preliminary Biogas Biomass Geothermal Small Hydro Solar Thermal Solar PV 20% Reference , ,024 33% Reference , ,513 3,165 8,338 Wind Out-of-State , ,753 (534 Outside CA) High Distributed Generation , ,095 15,959 (15,098 DG) ,870 (6,290 Outside CA) 27.5% , ,868 2,864 5,977 Low Load , ,907 2,867 7,091 5,067 Existing (MWhrs) Biogas Biomass Geothermal Small Hydro Solar Thermal Solar PV Wind 0 6,256 13, ,229 ISO 33% RPS Study of Operational Requirements and Market Impacts 11 Slide 11

13 Nexant Study - Assumed Locations of Incremental Renewables in California - 33% RPS, 2020 Preliminary Siskiyou Geothermal 135 Humboldt Biomass 100 Biomass Wind Bay Area Excluded areas with high population density, national parks, landmarks, sensitive areas, etc., to exclude areas not practical for renewable project development San Luis Obispo Solar 177 San Bernardino Solar 2303 Wind 750 LA/Kern Biomass 100 Solar 2250 Wind 2610 San Diego/ Imperial Wind 513 Riverside Solar 1250 Wind 500 Imperial Solar

14 Lets Look at the Technologies in California Wind Generation 13

15 Wind Generation Characteristics Low cost technology on an energy basis Production is Variable Uncertain Often Remotely Located Not highly correlated in time with system load Capacity credit 8-30% of nameplate for long range planning purposes Thus is considered a source of energy but not a significant source of capacity 14

16 Tehachapi Wind Generation in April 2005 Variable and Uncertain Each Day is a different color. Day Day 9 Megawatts Day 5 Day 26 Average Hour CAISO Data

17 Wind generation tends to be inversely correlated to daily system load CAISO Load -- Fall 2006 MW 32,000 31,000 30,000 29,000 28,000 27,000 26,000 25,000 24,000 23,000 22,000 21,000 20, Total Wind Hours -- Fall 2006 MW Hours Load Total Wind 16 CAISO Data

18 Wind vs. Actual Load on a Typical Hot Day in ,200 1,000 CAISO Wind Generation July 2006 Heat Wave /16/06 07/17/06 07/17/06 07/18/06 07/18/06 07/19/06 07/19/06 07/20/06 07/20/06 07/21/06 07/21/06 07/22/06 07/22/06 07/23/06 07/23/06 07/24/06 07/24/06 07/25/06 07/25/06 07/26/06 07/26/06 Total Wind Generation Installed Capacity = 2,648 MW Wind Generation at Peak MW Wind Generation Wind Generation at Peak 17 CAISO Data

19 Lets Look at the Technologies in California Solar Characteristics Higher cost on an energy basis Several technologies Thermal (central tower, trough, Sterling, etc) Thermal with storage or supplemental gas firing PV roof top PV large scale (> 1MW) Irradiation is variable but absent clouds relatively certain With clouds, production is less certain Solar production correlates better with system load Technologies with larger thermal mass tend to filter out short term variability (e.g. solar thermal) Solar Capacity Credit 60% 95% (depending upon technology) for the purpose of long range planning 18

20 Lets Look at the Technologies in California Solar Thermal Generation Trough Design Solar II Solar Central Receiver esolar s Modular Solar Power Plant Concept 19

21 Lets Look at the Technologies in California Solar PV 20

22 Solar Irradiation Examples Sacramento Area Spring 2009 data collected at SMUD PV site Data is collected at 1 minute intervals Several days shown in first week of March and May Indicative of Solar production, especially PV Source: 21

23 Solar Irradiation Examples Sacramento March 7 March 7, :30 6:48 7:06 7:24 7:42 8:00 8:18 8:36 8:54 9:12 9:30 9:48 10:06 10:24 10:42 11:00 11:18 11:36 11:54 12:12 12:30 12:48 13:06 13:24 13:42 14:00 14:18 Global Horizontal [W/m^2] 14:36 14:54 15:12 15:30 15:48 16:06 16:24 16:42 17:00 17:18 17:36 17:54 Time 22

24 Solar Irradiation Examples Sacramento March 5 March 5, :33 6:51 7:09 7:27 7:45 8:03 8:21 8:39 8:57 9:15 9:33 9:51 10:09 10:27 10:45 11:03 11:21 11:39 11:57 12:15 12:33 12:51 13:09 13:27 13:45 14:03 14:21 Global Horizontal [W/m^2] 14:39 14:57 15:15 15:33 15:51 16:09 16:27 16:45 17:03 17:21 17:39 17:57 Time 23

25 Solar Irradiation Examples Sacramento March 2 March 2, :38 6:56 7:14 7:32 7:50 8:08 8:26 8:44 9:02 9:20 9:38 9:56 10:14 10:32 10:50 11:08 11:26 11:44 12:02 12:20 12:38 12:56 13:14 13:32 13:50 14:08 Global Horizontal [W/m^2] 14:26 14:44 15:02 15:20 15:38 15:56 16:14 16:32 16:50 17:08 17:26 17:44 Time 24

26 Solar Irradiation Examples Sacramento May 5 May 5, :05 5:23 5:41 5:59 6:17 6:35 6:53 7:11 7:29 7:47 8:05 8:23 8:41 8:59 9:17 9:35 9:53 10:11 10:29 10:47 11:05 11:23 11:41 11:59 12:17 12:35 12:53 Global Horizontal [W/m^2] 13:11 13:29 13:47 14:05 14:23 14:41 14:59 15:17 15:35 15:53 16:11 16:29 Time 25

27 Typical Daily Wind vs. Solar Generation Pattern Shows Complimentary Nature 1390 Wind vs Solar May 25, :00 00:45 01:30 02:15 03:00 03:45 04:30 05:15 06:00 06:45 07:30 08:15 09:00 09:45 10:30 11:15 12:00 12:45 13:30 14:15 15:00 15:45 16:30 17:15 18:00 18:45 19:30 20:15 MW 21:00 21:45 22:30 23:15 00:00 Wind Solar 26 CAISO Data

28 System Operations Renewable Integration Operating the Power System Reliably Requires: Sufficient Regulation (second to second Auto Generation Control of generators or other resources) and Within-the-hour Net Load Following (ramping generators or other resources minute to minute) and Inter-hour Net Load Following (ramping over hour to hour) and Unit commitment to cover the peak plus reserves and Increased unit commitment requirements due to Variability and Uncertainty (forecast error) All of these are a function of the mix of renewables 27

29 Load-Following Up requirements under alternative, Summer 33% RPS Reference Case - Preliminary 12,000 Load following up vs. forecast errors 10,000 8,000 6,000 4,000 2, %Ref 20%HiLD HiDG OOS AllGas 27.5%HiLD All Forecast Error Improved Forecast Error No Forecast Error ISO 33% RPS Study of Operational Requirements and Market Impacts 28 Slide 28

30 System Operations Renewable Integration Over-generation potential increases with renewables Overgen occurs when inflexible generation exceeds load plus planned exports Energy Dump occurs when over-gen can not be sold to willing buyers in neighboring areas 2008 Nexant studies indicate Technology dependent With high solar penetration can occur during high load hours 29

31 System Operations Renewable Integration Overgen/Dump Energy Nexant Results Most likely to happen in March-May period when hydro, wind and solar production can all be high and on weekends when load is low In 2008 simulations, more than 90% of dump occurs in SCE territory Simulation understates dump due to simplified transmission model used in production simulation and normal hydro conditions assumed May require changes in current and future contract structure to allow more frequent curtailment, as well as needing to reduce minimum generation levels 30

32 Results Dump Energy for 50% RPS (2008 Study) Storage Technologies reduce dump energy MW 60,000 55,000 50,000 45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 Without Storage Hour MW With Storage (6 hours) 60,000 55,000 50,000 45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5, Wind Solar Traditional Load+Exports Load Hour Conditions on a day with solar and wind conditions both high 50% RPS 31

33 Analysis of generation fleet flexibility in 2020 with varying levels of renewables - Preliminary 6,500 6,000 5,500 5,000 4,500 Reg. Up & Load Following Up Req. (MW ) Summer Max 4,000 All Gas 20% 27.50% 33% 32

34 Analysis of generation fleet flexibility in Preliminary 58,000 56,000 54,000 52,000 50,000 48,000 46,000 44,000 42,000 40,000 Total Resources (MWs) at 17% PRM That Provide Regulation & Load Following All Gas 20% 27.50% 33% 33

35 33% Implementation Challenges - Transmission Long lead time to build transmission Uncertain which generation projects will develop Uncertain where transmission upgrades will be needed 2009 CPUC Study shows need for up to 7 new major high voltage transmission projects Indicative Nexant results 34

36 Nexant Results (2008) Transmission Expansion Bulk Transmission Upgrades For 33% RPS

37 33% Implementation Challenges - Transmission Potential Source of Multi-Year Delay for HV Transmission Typical lead-time 6-11 years Typical process involves: Project study and approval by CAISO (1-2 years) CPUC approval (2-3 years) Can add significant delays, e.g., delay in approval of Sunrise Project Other litigation post CPUC approval can also delay the project Engineering/Procurement (1-3 years) Construction/Environmental Mitigation (2-3) years Delays could be based on the route and degree of environmental mitigation 36

38 33% Implementation Challenges - Resources Need for resources to integrate renewables dependent upon renewable mix For example a high wind case would require more capacity to meet Planning Margin than a high solar 3000 MW needed for Wind (10,000 MW) and Solar (2,000 MW) 300 MW needed for Wind (4000 MW) and Solar (8000 MW) Assuming wind at 15% and solar at 60% capacity credit Regulation and ramping needs are dependent upon renewable mix Not well understood at this time for full mix of renewables CAISO 33% RPS Integration studies underway to clarify requirements 37

39 What are some of the options to address these challenges Focused geographical development to streamline transmission siting to reduce the time to build transmission Improved longer term analytical tools to help narrow the uncertainty as 2020 approaches - more probabilistic Planning Reserve Margin (PRM) may have to be expanded to include integration requirements Potential increased use of distributed generation to reduce the transmission delays associated with larger scale remote wind and solar 38

40 Resources can potentially meet the system integration needs In addition to traditional CCGTs and GTs Improved control over new wind and solar to deal with severe ramps and over-gen events Potential increased role for demand side responses to address regulation and ramps Potential increased role for storage to address regulation, load following and over-gen (full range of options from PP Hydro, CAES, batteries etc.) Improved wind forecasting to reduce daily uncertainty and thus reduce regulation and load following requirements Improved solar generation forecasting through better cloud cover forecasting to reduce daily uncertainty 39

41 What potentially can meet integration needs (2) In addition Solar thermal generation with integrated storage or supplemental firing to reduce Reg and LF requirements Increased contribution from existing hydro and pumped storage may result in increased maintenance and costs Increased contribution from existing thermal generation may require capital improvements to achieve increased level of flexibility Fast regulation from high speed batteries or inertial storage Electric Vehicle battery management Increased reliance on RECs for out of state renewables 40

42 Questions Questions? 41