wholesale market operations · pdf filewholesale market operations update john dumas director...

Wholesale Market Operations Update

John DumasDirector of Wholesale Market Operations

December 12-13, 2011

Board of Directors Meeting

Day-Ahead Schedule (October )

• On average the DAM net transmission flow (defined below) was greater than the real• On average the DAM net transmission flow (defined below) was greater than the real-time system load for all 24 hours

MW50000

30000

40000

10000

20000

0

10000

Hour

Acronym : TPO - Three Part Offer; EOO – Energy Only Offer; DAM Net Flow = Combined market transmission flow of Energy purchased/sold in Day Ahead Market plus

Average DAM Schedule 01OCT2011-31OCT2011

PLOT EOO_Scheduled TPO_Scheduled DAM_Energy_PurchaseDAM_Net_Flow RT_System_Load

2

DAM_Net_Flow = Combined market transmission flow of Energy purchased/sold in Day-Ahead Market plus Point-to-Point Obligations and NOIE CRR Options carried forward to real-time.

December 12-13, 2011 ERCOT Public

Day-Ahead Schedule (November)

• On average the DAM net transmission flow (defined below) was greater than the real• On average the DAM net transmission flow (defined below) was greater than the real-time system load for all 24 hours

MW50000

30000

40000

10000

20000

0

10000

Hour

Acronym : TPO - Three Part Offer; EOO – Energy Only Offer; DAM Net Flow = Combined market transmission flow of Energy purchased/sold in Day Ahead Market plus

Average DAM Schedule 01NOV2011-30NOV2011

PLOT EOO_Scheduled TPO_Scheduled DAM_Energy_PurchaseDAM_Net_Flow RT_System_Load

3

DAM_Net_Flow = Combined market transmission flow of Energy purchased/sold in Day-Ahead Market plus Point-to-Point Obligations and NOIE CRR Options carried forward to real-time.


Day-Ahead Electricity And Ancillary Service Hourly Average Prices (October )

50$/

MW

h) 30

40

Pric

e ($

10

20

0

Hour

• Both Energy and AS prices followed the trend of load profile on average.

Day-Ahead Electricity and Ancillary Service Hourly Average Prices 01OCT2011-31OCT2011

PLOT System_lambda Reg_Up Reg_DownResponsive_Reserve Non_Spin

4

gy p p g• In some hours, Responsive Reserve prices were higher than Regulation Up prices.

ERCOT PublicDecember 12-13, 2011

Day-Ahead Electricity And Ancillary Service Hourly Average Prices (November)

40$/

MW

h)

20

30

Pric

e ($

10

20

0

Hour

• Both Energy and AS prices followed the trend of load profile on average.

Day-Ahead Electricity and Ancillary Service Hourly Average Prices 01NOV2011-30NOV2011

PLOT System_lambda Reg_Up Reg_DownResponsive_Reserve Non_Spin

5

gy p p g• In all hours, Responsive Reserve prices were lower than or close to Regulation Up prices.


Day-Ahead Vs Real-Time Load Zone SPP(Hourly Average) (October )

8090

100

8090

100Load Zone West Prices Load Zone North Prices

Pric

e ($

/MW

h)

01020304050607080

Pric

e ($

/MW

h)

01020304050607080

0

Hour

0

Hour

8090

100

8090

100

Average_of_DASPP Average_of_RTSPP Average_of_DASPP Average_of_RTSPPLoad Zone South Prices Load Zone Houston Prices

Pric

e ($

/MW

h)

1020304050607080

Pric

e ($

/MW

h)1020304050607080

0

Hour

0

HourDA Vs RT Hourly Average SPP Load Zone Summary 01OCT2011-31OCT2011

6 ERCOT PublicDecember 12-13, 2011

Day-Ahead Vs Real-Time Load Zone SPP(Hourly Average) (November)

8090

100

8090

100Load Zone West Prices Load Zone North Prices

Pric

e ($

/MW

h)

01020304050607080

Pric

e ($

/MW

h)

01020304050607080

0

Hour

0

Hour

8090

100

8090

100

Average_of_DASPP Average_of_RTSPP Average_of_DASPP Average_of_RTSPPLoad Zone South Prices Load Zone Houston Prices

Pric

e ($

/MW

h)

1020304050607080

Pric

e ($

/MW

h)1020304050607080

0

Hour

0

HourDA Vs RT Hourly Average SPP Load Zone Summary 01NOV2011-30NOV2011


Day-Ahead vs Real-Time HUB SPP (Hourly Average) (October )

8090

100

8090

100Hub West Prices Hub North Prices

Pric

e ($

/MW

h)

01020304050607080

Pric

e ($

/MW

h)

01020304050607080

0

Hour

0

Hour

8090

100

8090

100

Average_of_DASPP Average_of_RTSPP Average_of_DASPP Average_of_RTSPPHub South Prices Hub Houston Prices

Pric

e ($

/MW

h)

1020304050607080

Pric

e ($

/MW

h)1020304050607080

0

Hour

0

HourDA Vs RT Hourly Average SPP Hub Summary 01OCT2011-31OCT2011


Day-Ahead vs Real-Time HUB SPP (Hourly Average) (November)

8090

100

8090

100Hub West Prices Hub North Prices

Pric

e ($

/MW

h)

01020304050607080

Pric

e ($

/MW

h)

01020304050607080

0

Hour

0

Hour

8090

100

8090

100

Average_of_DASPP Average_of_RTSPP Average_of_DASPP Average_of_RTSPPHub South Prices Hub Houston Prices

Pric

e ($

/MW

h)

1020304050607080

Pric

e ($

/MW

h)1020304050607080

0

Hour

0

HourDA Vs RT Hourly Average SPP Hub Summary 01NOV2011-30NOV2011


Day-Ahead Vs Real-Time Hub Average SPP (Hourly Average) (October )

90

100M

Wh) 60

70

80

Pric

e ($

/M

20

30

40

50

0

10

20

Hour

• Day Ahead prices were close to Real-Time prices, especially during off-peak hours.• Day Ahead prices were on average higher than Real-Time prices for some peak hours

Hour

Day-Ahead Vs Real-Time Hourly Average SPP of Hub Hub-Average Prices 01OCT2011-31OCT2011

PLOT Average_of_DASPP Average_of_RTSPP

10 ERCOT Public

• Day Ahead prices were on average higher than Real-Time prices for some peak hours.


Day-Ahead Vs Real-Time Hub Average SPP (Hourly Average) (November)

90

100M

Wh) 60

70

80

Pric

e ($

/M

20

30

40

50

0

10

20

Hour

• Day Ahead prices were close to Real-Time prices, especially during off-peak hours.• Day Ahead prices were on average higher than Real-Time prices for most peak hours

Hour

Day-Ahead Vs Real-Time Hourly Average SPP of Hub Hub-Average Prices 01NOV2011-30NOV2011

PLOT Average_of_DASPP Average_of_RTSPP

11 ERCOT Public

• Day Ahead prices were on average higher than Real-Time prices for most peak hours.


Day-Ahead Vs Real-Time Cumulative Average SPP (October )

40M

Wh)

20

30

SP

P ($

/M

0

10

-20

-10

Date

• The cumulative Real Time prices were typically more volatile than the Day-Ahead prices i O t b b t th diff d i th th

Date

Day-Ahead Vs Real-Time Cumulative Average SPP for Simple Average Prices 01OCT2011-31OCT2011

PLOT AVG_DAM AVG_RT AVG_DAM-AVG_RT

12

in October but the difference was decreasing over the month


Day-Ahead Vs Real-Time Cumulative Average SPP (November)

40M

Wh)

20

30

SP

P ($

/M

0

10

-20

-10

Date

• The cumulative Real Time prices were more volatile than the Day-Ahead prices

Date

Day-Ahead Vs Real-Time Cumulative Average SPP for Simple Average Prices 01NOV2011-30NOV2011

PLOT AVG_DAM AVG_RT AVG_DAM-AVG_RT


Load Weighted Average SPP (October )

Day-AheadReal-Time

40

33 5

37.9

35.132 9

e SP

P ($

/MW

h)

30

31.829.9

33.531.8

30.6 30.328.8

30.232.4

30.328.6

32.9

27.826.2

eigh

ted

Ave

rage

20

Load

W

0

10

• The load weighted average RT SPPs were slightly higher than the load weighted average

Load Weighted Average SPP for Each Zone 01OCT2011-31OCT2011

0LZ_1

AEN2

CPS3

LCRA4

RAYBN5

HOUSTON6

NORTH7

SOUTH8

WEST

14 ERCOT Public

DAM SPPs in 2 load zones and lower in the other 6 load zones.


Load Weighted Average SPP (November)

D Ah dDay-AheadReal-Time

30

26.9

29.027.3

29.1

26.9

29.1

26.6

28.827.1 26.8

29.3 29.1

27.4

29.529.6

age

SPP

($/M

Wh)

20

22.3

Wei

ghte

d A

vera

10

Load

0LZ1 2 3 4 5 6 7 8

• The load weighted average RT SPPs were slightly higher than the load weighted average DAM SPPs in 7 load zones.

Load Weighted Average SPP for Each Zone 01NOV2011-30NOV2011

LZ_1AEN

2CPS

3LCRA

4RAYBN

5HOUSTON

6NORTH

7SOUTH

8WEST

15 ERCOT Public

DAM SPPs in 7 load zones.• The load weighted average RTSPP was significantly higher than that of DASPP in

LZ_WEST .


DRUC Monthly Summary (October )

45000

31 Executions (0 Missed) 9.90-Min Average Execution Time1 Published after 1600; 0 Published after 1800 0 MWh Committed (0 Resources for 0 Hours)

Fore

cast

(MW

)

3700038000390004000041000420004300044000

ge S

ched

ule/

Load

290003000031000320003300034000350003600037000

DR

UC

Ave

rag

2500026000270002800029000

Hour

DRUC Average QSE Scheduled Capacity/Load Forecast 01OCT2011-31OCT2011

PLOT LD_FCST_MW DRUC_INPUT_HSL_MWSCHED_HASL_MW DRUC_OUTPUT_HSL_MW


DRUC didn’t commit any Resource and hence input and output HSL MW is the same

DRUC Monthly Summary (November)

30 Executions (0 Missed)0 Published after 1600; 0 Published after 1800Note: Colors Indicate Individual Resources

41000

6.50-Min Average Execution Time71940 MWh Committed (5 Resources for 555 Hours)

acity

(MW

)

3000

4000

Fore

cast

(MW

)

34000350003600037000380003900040000

et C

omm

itted

Cap

a

2000

ge S

ched

ule/

Load

28000290003000031000320003300034000

DR

UC

Ne

0

1000

DR

UC

Ave

rag

24000250002600027000

Hour

Net Committed Capacity in DRUC 01NOV2011-30NOV2011 DRUC Average QSE Scheduled Capacity/Load Forecast 01NOV2011-30NOV2011

PLOT LD_FCST_MW DRUC_INPUT_HSL_MWSCHED_HASL_MW DRUC_OUTPUT_HSL_MW


HRUC Monthly Summary (October )

742 Executions (2 Missed)7.70-Min Average Execution TimeNote: Colors Indicate Individual Resources

p. (M

W)

20000

30000

Net

Com

mitt

ed C

ap

10000

20000

HR

UC

0

10000

Net Committed Capacity in HRUC 01OCT2011-31OCT2011


HRUC Monthly Summary (November)

719 Executions (1 Missed)5.20-Min Average Execution TimeNote: Colors Indicate Individual Resources

p. (M

W)

1000

1100

1200

1300

1400

Net

Com

mitt

ed C

ap

500

600

700

800

900

HR

UC

0

100

200

300

400

Net Committed Capacity in HRUC 01NOV2011-30NOV2011


Supplemental Ancillary Service Market (SASM) Summary(October )

17 SASMs in 01OCT2011-31OCT2011For AS Failure to Provide

8

0 SASMs in 01OCT2011-31OCT2011For AS Undeliverability

8

o P

rovi

de

6

7

8

erab

ility

6

7

8

ASM

s fo

r Fai

lure

to

3

4

5

ASM

s fo

r Und

eliv

e

3

4

5

Cou

nt o

f SA

1

2

3

Cou

nt o

f SA

1

2

3

0Hour

0Hour


Supplemental Ancillary Service Market (SASM) Summary(November)

7 SASMs in 01NOV2011-30NOV2011For AS Failure to Provide

6

0 SASMs in 01NOV2011-30NOV2011For AS Undeliverability

6

o P

rovi

de

5

6

erab

ility 5

6

ASM

s fo

r Fai

lure

to

3

4

ASM

s fo

r Und

eliv

e

3

4

Cou

nt o

f SA

1

2

Cou

nt o

f SA

1

2

0Hour

0Hour


Non-Spinning Reserve Service Deployment

Deployment St t

Deployment E d

Deployment D ti H

Max Deployment Reason

Start End Duration Hoursp y

MW

10/15/2011 15:08

10/15/2011 21:29

6.34 67 Congestion

10/16/2011 15:51

10/16/2011 18:59

3.14 372.6 Congestion

10/17/2011 14:30

10/17/2011 20:00

5.49 93 Congestion14:30 20:00

10/21/2011 14:43

10/21/2011 22:00

7.27 96 Congestion

10/22/2011 10/22/2011 1 63 140 Congestion

11:32 13:101.63 140 Congestion

11/1/2011 6:5311/1/2011

19:1112.29 147 Congestion


CRR Auction for Operating Month November 2011

127 442 Bid /Off• 127,442 Bids/Offers• 13,284 Auction Awards

• 136 144 MW Total• 136,144 MW Total• 49,747 Peak WD• 47,022 Peak WE• 39,373 Off-peak

• Total Auction/Allocation Revenue = $ 8.35 M

23December 12-13, 2011 ERCOT Public

CRR Auction for Operating Month December 2011

118 448 Bid /Off• 118,448 Bids/Offers• 11,594 Auction Awards

• 140 441 MW total• 140,441 MW total• 51,675 Peak WD• 47,166 Peak WE• 41,600 Off-peak



CRR Price Convergence*

*C t l l ti th d f litti t b t th i lti th ti P i th d th l ith th CRR A ti R


*Cost calculation uses a new method of splitting costs between months in multi-month auctions. Previous months used the same algorithm as the CRR Auction Revenue Distribution.

Annual CRR Auction for 2012/2013

2012 Annual Auction (55% of system capacity)2012 Annual Auction (55% of system capacity)• 90,620 Bids/Offers• 25,514 Auction Awards

• 1,008,660 MW• 370,880 Peak WD• 350,418 Peak WE• 287,362 Off-peak

• Total Auction/Allocation Revenue = $ 167 5 MTotal Auction/Allocation Revenue $ 167.5 M• Credit Allocation = $1,210,000,000

2013 Annual Auction (15% of system capacity)• 190,359 Bids/Offers• 27,292 Auction Awards

• 309,475 MW114 096 P k WD• 114,096 Peak WD

• 105,745 Peak WE• 89,634 Off-peak


26

• Credit Allocation = $442,000,000


Both years of Auction published on time (completed execution between Oct31 Nov16)

Annual CRR Auction for 2012/2013

– Both years of Auction published on time (completed execution between Oct31- Nov16)– Lessons learned:

• Credit- Significant market credit hold of $1.6 billion for 2 weeks, so team to propose changes to timing and sequence to alleviate coinciding credit collateral posting (next bullet).

• Sequence/timing considering with market changes to execute 2 year auction in separate bid and• Sequence/timing- considering with market changes to execute 2-year auction in separate bid and credit windows for each year to relieve market credit and improve analysis of results.

• Technical- 2013 PeakWeekDay auction executed for 194 hours, so evaluating performance improvement options for next year’s auction (software, servers, configuration).

• Bid Limit of 200 000- Market was oversubscribed for 2012 so looking at alternative minimum bidsBid Limit of 200,000- Market was oversubscribed for 2012 so looking at alternative minimum bids prices or curtailment of least value bids prior to running auctions.


Market Enhancements Under Consideration

• Three NPRRs were submitted on Nov 9th and approved by TAC on Dec 1st to cover discussion at PUCT Open Meeting on October 27

– NPRR426 Standing Non-Spin Deployment in the Operating Hour for Resources Providing On-Line Non-Spin– NPRR427 Energy Offer Curve Requirements for Generation Resources Assigned Reg-Up and RRS– NPRR428 Energy Offer Curve Requirements for Generation Resources Assigned Non-Spin Responsibility

• Evaluating market design improvement proposals– Moving 500MW from Non-spin to RRS– Putting offer floor for RUC committed resources– Changing Power Balance penalty cost.

• Evaluating feasibility of implementing Pilots for– Fast response regulation service– 30 min EILS service– Active deployment of Load Resources participating in Non-Spin based on forecasted pricep y p p g p p

• Demand Response– EILS Proposed Rule Enhancements

• Option to renew contract after deployment obligations are met early in a contract period• Adding Distributed Generation could potentially add an additional 50 – 100MW by summer• 30 minute EILS response could potentially add an additional 80 – 100 MW by summer

• Look-Ahead SCED functions framed for market consideration – Workshop held on November 28th to begin detail discussions of whitepaper and implementation– NPRR351 will represent Phase I of Implementation (indicative future prices and ramping)

28

p p ( p p g)


Look-Ahead SCED

• Key milestonesKey milestones– Jan –Oct 2012: Work with market on Look-Ahead SCED whitepaper and scope– Summer 2012: Implement non-binding Phase I / NPRR351 (future prices)– October 2012 : Market approval of protocols for subsequent Phase(s) – Summer 2013: Potential implementation of Phase 2 (binding prices and commitment)

Phase Timeline Feature Implemented Comment

Phase 1 Summer Short Term future Advisory/Indicative Base Will initially run in Open LoopPhase -1 Summer 2012

Short-Term future Advisory/Indicative Base Points and LMPs using basic version of RTD.

Will initially run in Open-Loop i.e outputs (Base Points,LMPs) are Advisory/Indicative. i.e. non-binding.

Phase-2 Summer 2013

Commitment for QSGR and Load Resources with intra-hour temporal constraints using basic version of RTC.

Will initially run in Open-Loop i.e. outputs (Commitment instructions) are Advisory/Indicative. i.e. non-bi dibinding.

Phase-3 Spring2014

Real-Time Ancillary Service and Energy Co-optimization using RTC.

Phase-4 Fall 2015 Transmission constraints for future intervals


Phase-4 Fall 2015 Transmission constraints for future intervals

wholesale market operations · pdf filewholesale market operations update john dumas director...

Documents