ERCOT Historic Synchronous Inertia (Kinetic Energy) and ... … · of all online synchronous generators on hourly basis; • Jan-May, Nov-Dec data for 2010-2013 and January-May data

ERCOT Historic Synchronous Inertia (Kinetic Energy) and Future Projections

Julia Matevosyan, PhDSr. Planning EngineerResource AdequacyERCOT

Historic Analysis Assumptions

• Kinetic energy is calculated as a sum of H*MVA of all online synchronous generators on hourly basis;

• Jan-May, Nov-Dec data for 2010-2013 and January-May data for 2014 are included in the analysis;

• Peak wind penetration hour (i.e. max Pwind/Pload) in each year is analyzed in details (load, wind, wind capacity factor)

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

3.2

3.4x 10

5

2010 2011 2012 2013 2014

Kin

etic

ene

rgy,

MW

s

2010 2011 2012 2013 2014Installed

Capacity, MW 9,116 9,452 10,034 10,570 11,066

Max Pwind/Pload 25.5% 27.4% 29.8% 35.8% 39.4%

Pwind, MW 6,483 6,772 7,247 8,773 9,699

Capacity Factor 71% 72% 72% 83% 88%

Pload, MW 25,427 24,745 24,328 24,488 24,617

Historic Kinetic Energy

Expected installed wind capacity in a future year (with SGIA or SGIA&FC)

Project an hour with highest instantaneous

penetration of wind based on historical trends

Projected peak wind penetration hour:

Pwind=CapacityFactorhist*P wind capacity

Pload= Avg. historical load at wind penetration peaks

Penetration = Pload/Pwind

Net Load=Pload – Pwind

Expected wind capacity in a future year X:

P wind capacity

Project system inertia for this hour, based on

historical inertia trendline

Synchronous inertia at projected wind penetration peak:

SI = a*Net Load +b

Future Inertia Projection Methodology

Kinetic Energy Trend used for Future Projections

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

3.2

3.4x 10

5

2010 2011 2012 2013 2014 2015 2016 2017

Kin

etic

ene

rgy,

MW

s

2010 2011 2012 2013 2014 2015 2016 2017Installed

Capacity, MW 9,116 9,452 10,034 10,570 11,066 19,443 20,630 21,130

Max Pwind/Pload 25.5% 27.4% 29.8% 35.8% 39.4% 69% 73.2% 75%

Pwind, MW 6,483 6,772 7,247 8,773 9,699 17,041 18,082 18,520

Capacity Factor 71% 72% 72% 83% 88% 88% 88% 88%

Pload, MW 25,427 24,745 24,328 24,488 24,617 24,700 24,700 24,700

Historic Kinetic Energy and Future Projections

at max wind penetration, historic

at max wind penetration, projected based on SGIAs

∗ at max wind penetration, projected based on SGIA&FCs

Frequency deviation after 2750 MW trip (in 0.5 s)

Additional metrics

• Maximum permissible RoCoF, Hz/s: based on frequency deviation to under-frequency load shed (UFLS) and time until first fast frequency response (FFR) in a system is fully deployed (e.g. 0.5 seconds for Responsive Reserve provided by Loads in ERCOT).

RoCoFmax=ΔfUFLS/tFFR,

From this metric based on largest contingency and load damping constant, minimum inertia requirement can be calculated for a system.

• Hz/MW metric: this is frequency nadir per MW generation trip, this metric does not only consider inertia but also includes governor response, load damping and fast frequency response. This metric can be tracked based on historic events and projected for the future (in use in ERCOT).

Frequency nadir for system load conditions ≤ 35000 MW (interpolated based on historical events)

Appendix: Supporting data for slide 6

2010 2011 2012 2013 2014 2015 (w. FC) 2015 2016 2017Installed

Capacity, MW 9,116 9,452 10,034 10,570 11,066 17179 19,443 20,630 21,130

Max Pwind/Pload 25.5% 27.4% 29.8% 35.8% 39.4% 61% 69% 73.2% 75%

Pwind, MW 6,483 6,772 7,247 8,773 9,699 15057 17,041 18,082 18,520

Capacity Factor 71% 72% 72% 83% 88% 88% 88% 88% 88%

Net Load, MW 18944 17973 17082 15716 14918 9643 7659 6618 6180

Inertia, MWs 161741 147081 133675 120030 119604 89469 80020 75066 72979

EstimatedRoCoF, Hz/s 0.501 0.551 0.605 0.672 0.674 0.89 0.996 1.059 1.088

20000

30000

40000

50000

60000

70000

80000

5/1

05/

1 1

05/

1 2

05/

2 6

5/2

16

5/3

25/

3 1

25/

3 2

25/

4 8

5/4

18

5/5

45/

5 1

45/

6 0

5/6

10

5/6

20

5/7

65/

7 1

65/

8 2

5/8

12

5/8

22

5/9

85/

9 1

85/

10

45/

10

145/

11

05/

11

105/

11

205/

12

65/

12

165/

13

25/

13

125/

13

225/

14

85/

14

185/

15

45/

15

14

SCIT

Iner

tia

Hourly IMay 2

SC

30000

35000

40000

45000

50000

55000

60000

65000

70000

75000

80000

MW

/Sec

Inertia Southern

5/15

14

5/16

0

5/16

10

5/16

20

5/17

6

5/17

16

5/18

2

5/18

12

5/18

22

5/19

8

5/19

18

5/20

4

5/20

14

5/21

0

5/21

10

5/21

20

5/22

6

5/22

16

5/23

2

5/23

12

5/23

22

5/24

8

5/24

18

5/25

4

5/25

14

5/26

0

5/26

10

5/26

20

5/27

6

5/27

16

5/28

2

5/28

12

5/28

22

5/29

8

5/29

18

5/30

4

5/30

14

5/31

0

5/31

10

5/31

20

Inertia 2014

SCIT Inertia (MW/Sec)

a Max and Minn California May 2014

Max Min

-8000-7500-7000-6500-6000-5500-5000-4500-4000-3500-3000-2500-2000-1500-1000-5000

/

Net

Impo

rts

NERC’ Essential Reliability Service Task Force

Clyde Loutan, Senior Advisor Renewable Energy Integration

October 29, 2014NERC Office Atlanta

Take away from Vancouver

Key focus for our group is to develop the process of predicting future ramping needs, specifically the three hour ramping need as illustrated by CAISO’s example

• Develop hourly upward ramping needs

• Develop hourly downward ramping needs

• Develop three-hours upward ramping needs

• Develop three-hours downward ramping needs

Slide 2

Loads and Resources Team

Subgroup Lead Company Email

Clyde Loutan CAISO [email protected]

Subgroup Members

Amir Najafzadeh NERC [email protected]

Brendan Kirby Kirby Consulting [email protected]

Dave Devereaux IESO [email protected]

Ed Scott Duke Energy [email protected]

Jay Ruberto First Energy [email protected]

Layne Brown WECC [email protected]

Michael McMullen MISO [email protected]

Michael Milligan NREL [email protected]

Noha Abdel-Karim NERC [email protected]

Pooja Shah NERC [email protected]

Ron Carlsen Southern Company [email protected]

Todd Lucas Southern company [email protected] Siegrist Brickfield, Burchette, Ritts & Stone, P.C. [email protected] Dariush Shirmohammadi California Wind Energy Association [email protected] Tuohy EPRI [email protected]

Slide 3

2014 Monthly load vs. net-load profiles

18,00019,00020,00021,00022,00023,00024,00025,00026,00027,00028,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

January 2014 --- Load vs. Net Load

18,000

19,000

20,000

21,000

22,000

23,000

24,000

25,000

26,000

27,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

February 2014 --- Load vs. Net Load

16,000

18,000

20,000

22,000

24,000

26,000

28,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

March 2014 --- Load vs. Net Load

16,000

18,000

20,000

22,000

24,000

26,000

28,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

April 2014 --- Load vs. Net Load

16,000

18,000

20,000

22,000

24,000

26,000

28,000

30,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

May 2014 --- Load vs. Net Load

18,000

20,000

22,000

24,000

26,000

28,000

30,000

32,000

34,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

June 2014 --- Load vs. Net Load

18,00020,00022,00024,00026,00028,00030,00032,00034,00036,00038,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

July 2014 --- Load vs. Net Load

18,00020,00022,00024,00026,00028,00030,00032,00034,00036,00038,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

August 2014 --- Load vs. Net Load

18,000

20,000

22,000

24,000

26,000

28,000

30,000

32,000

34,000

36,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

September 2014 --- Load vs. Net Load

Slide 4

Option 1 – CAISO’s determination of flexible capacity for future years

• NREL’s 2005 VERs profiles were used to develop the wind profiles and Clean Power data was used to develop the solar profiles

• Obtain the latest PTOs assumption of VERs build-out and CPUCs RPS calculator• Calculate 1-minute net-load

Load

• Use 2005 1-minute actual load data

Wind

• Develop 1-minute wind production profiles for CREZs based on their geographic location using NREL’s 2005 wind profiles

Solar installation

• Develop 1-minute solar production profiles for CREZs based on their geographic location and technology using NREL’s 2005 solar profiles (i.e. solar thermal, solar PV tracking & solar PV fixed)

• Aggregate all new solar 1-minute production data by technology

• New CREZs does not have the load/solar correlation but the maximum 3-hour ramps during the non-summer months are highly influenced by sunset which is consistent with existing solar data

Slide 5

Option 2 - The monthly flexibility capacity requirement is calculated using the most recent full year of the CAISO’s load, wind, and solar 1-minute data

• Use 2013 actual CAISO’s load, wind and solar 1-minute data

• For new VERs installation, use NREL’s simulated production data for CREZs located in close geographic proximity to develop minute-by-minute production profiles

• Solar profiles were created using both technology type and location of the new resources

• Generate net-load profiles for 2020

– Generate 1-minute load profiles for 2020

– Generate 1-minute solar profiles for 2020

– Generate 1-minute wind profiles for 2020

Slide 6

Option 2 - Wind growth assumptions

• Use actual 1-minute wind production data for the most recent year e.g. 2013 actual 1-minute data was used to build 2020 1-minute data

• 1-minute wind profiles for projects installed in 2020 were created using 2013 actual data for the months the projects were not in-service (i.e. profiles for projects installed in May 2013 were created for January through April)

• Wind 1-minute profiles for 2020 were created by scaling the 1-minute wind data for 2013 based on installed capacity

2020 W1-min = 2013WActual_1-min * 2020WInstalled Capacity/2013WInstalled Capacity

Slide 7

Option 2 - Solar growth assumptionsExisting solar

• Use actual solar 1-minute production data for the most recent year (e.g. 2013 actual 1-minute solar data was used to develop 2020 profile

• 1-minute solar profiles for projects installed in 2013 were created for the months the projects were not in-service using 2013 actual solar data

New solar installation

• Develop 1-minute solar production profiles for CREZs based on their geographic location and technology using NREL’s 2005 solar profiles (i.e. solar thermal, solar PV tracking & solar PV fixed)

• Aggregate all new solar 1-minute production data by technology

• New CREZs does not have the load/solar correlation but the maximum 3-hour ramps during the non-summer months are highly influenced by sunset which is consistent with existing solar data

• Sum the actual 1-minute existing solar production data with the aggregated simulated solar data for new installation

Total solar 20201-min = 2013Actual_1-min + 2020Simulated_1-min data

Slide 8

Calculating 1-hour and 3-hour upward/downward ramp capacity using net load (NL)

Option 1 – One minute moving window• 1-Hour Ramp: NL61-NL1, NL62-NL2, NL63-NL3….NLn+61-NLn+1 ….n ≥0• 3-Hour Ramp: NL181-NL1, NL182-NL2, NL183-NL3….NLn+180-NLn

Option 2 – Five minute moving window• 1-Hour Ramp: NL61-NL1, NL66-NL6, NL71-NL11….NL5n+61-NL5n+1 …. n ≥0• 3-Hour Ramp: NL181-NL1, NL186-NL6, NL191-NL11….NL5n+181-NL5n+1

Option 3 – Average of one minute moving window• 1-Hour Ramp or 3-Hour

Up RampIf(Avg(NLn+6+NLn+7+NLn+8+NLn+9+NLn+10) - Avg(NLn+1+NLn+2+NLn+3+NLn+4+NLn+5)) >0

Down RampIf(Avg(NLn+6+NLn+7+NLn+8+NLn+9+NLn+10) - Avg(NLn+1+NLn+2+NLn+3+NLn+4+NLn+5)) <0

Slide 9

Maximum monthly 1-Hour upward ramping capacity for 2014 and expected 1-hour ramping capacity for 2020

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec2014 4,084 3,844 3,279 3,641 2,308 3,063 2,340 2,464 2,8802020 6,152 6,944 6,688 5,852 4,830 4,517 4,283 5,131 6,362 6,499 7,199 7,777

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

MW

Monthly 1-Hour Up Ramp Capacity --- 2014 & 2020

2014 2020

Slide 10

Maximum monthly 1-hour upward ramping capacity for 2014 and expected 1-hour ramping capacity for 2020

Maximum monthly 1-hour downward ramping capacity for 2014 and expected 1-hour ramping capacity for 2020

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec2014 -2,943 -2,671 -3,389 -3,822 -4,131 -4,180 -3,674 -3,992 -4,2482020 -4,355 -4,443 -5,821 -4,989 -4,956 -4,736 -5,009 -5,375 -5,721 -5,452 -5,084 -4,946

-7,000

-6,000

-5,000

-4,000

-3,000

-2,000

-1,000

0

MW

Monthly 1-Hour Down Ramp Capacity --- 2014 & 2020

2014 2020

Slide 11

Maximum monthly 3-hour upward ramping capacity for 2014 and expected 3-hour ramping capacity for 2020

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec2014 7,288 7,078 6,843 6,613 5,574 5,150 5,750 5,577 7,8852020 13,788 15,830 14,917 12,071 11,800 11,516 11,127 12,867 15,053 13,194 15,899 17,956

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000

MW

Monthly 3-Hour Up Ramp Capacity --- 2014 & 2020

2014 2020

Slide 12

Maximum monthly 3-hour downward ramping capacity for 2014 and expected 3-hour ramping capacity for 2020

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec2014 -6,490 -6,734 -7,625 -8,461 -10,615 -9,831 -9,729 -10,449 -10,8052020 -9,228 -9,882 -9,651 -9,565 -12,205 -10,837 -11,830 -13,042 -13,310 -10,362 -8,657 -9,847

-14,000

-12,000

-10,000

-8,000

-6,000

-4,000

-2,000

0

MW

Monthly 3-Hour Down Ramp Capacity --- 2014 & 2020

2014 2020

Slide 13

1

WWSIS III: Western Frequency Response and Transient Stability Study – System Inertia Discussion GE Energy Nicholas W. Miller (PM)

Miaolei Shao Slobodan Pajic Rob D’Aquila

NREL Kara Clark (PM)

Presented by Jason MacDowell for ERSTF – Frequency Response TF Atlanta, GA October 29-30, 2014

The draft report is under review by the TRC and by DOE. Therefore, all of the results and statements in this presentation MUST be regarded as preliminary and subject to further review and modification.

Preliminary Results of Western Wind and Solar Integration Study, Phase III : Transient Stability and Frequency Response. Subject to Final Review and Approval by DOE

WECC CALIFORNIA DSW NORTHEAST NORTHWEST

Wind (GW) 27.2 4.7 7.0 5.4 8.4

PV (GW) 10.2 5.8 3.3 0.8 0.3

CSP (GW) 8.4 1.5 7.0 0.0 0.0

DG (GW) 7.0 3.7 2.9 0.4 0.2

Others (GW) 65.7 15.1 11.4 5.4 11.7

total (GW) 118.6 30.8 31.6 12.0 20.5 2

Light Spring Base and High Renewable Cases (‘22)

~ 52.8 GW of wind and solar total

~ 25 GW of wind and solar total in

reference case

US only

~25% of US

generation

~54% of US

generation

Preliminary Results of Western Wind and Solar Integration Study, Phase III : Transient Stability and Frequency Response. Subject to Final Review and Approval by DOE

Inertial Impact of Increasing VG on ROCOF (rate of change of frequency)

Trip 2 Palo Verde units (~2,750MW)

Light Spring Base

Light Spring Hi-Mix

Light Spring Hi-Mix

CSP changed to PV ROCOF (1 sec – 3 sec):

• Blue (Base): 0.096 Hz/Sec • Red (+25GW Wind + Solar):

0.113Hz/Sec = +18% • Green (7GW CSP -> PV):

0.118 Hz/Sec =+22%

Initial ROCOF proportional to

1/Inertia

Preliminary Results of Western Wind and Solar Integration Study, Phase III : Transient Stability and Frequency Response. Subject to Final Review and Approval by DOE

Thank you!