11/4/2013. NERC RS Meeting October 28, 2013 Sullivan s Island, SC. Troy Blalock SERC NERC RS Representative

Transcription

1 NERC RS Meeting October 28, 2013 Sullivan s Island, SC Troy Blalock SERC NERC RS Representative 2 1

2 Preface: Members of the NERC Resource Subcommittee, whom are representatives of the Eastern Interconnection, are working with Balancing Authorities within the same interconnection on a voluntary basis to support a pilot program in an effort to improve frequency response. Frequency Response is defined as automatic and sustained change in the power consumption or output of a device such as generator that occurs within 5-20 seconds of and is in a direction to oppose a change in the Interconnection Frequency. 3 Frequency response as so defined has been historically declining except for more recent years in the Eastern Interconnection. Typically exhibits the lazy L of primary frequency response withdrawal. 4 2

3 The current initiative focus is on the existing generator fleet with respect to : 1) completeness and accuracy of the data provided in the 2010 NERC Generator Survey, 2) to improve their frequency response capabilities and; 3) for Balancing Authorities and/ or Reliability Coordinators to installs tool(s) to monitor individual generator performance within their authority of control and communicate performance results to the individual generators. 5 Eastern Interconnection: Deadband Setting Given? Yes (2531, 57.21%) No (1893, 42.79%) 6 3

4 % Eastern Interconnection: Deadband Settings er of Units within Range Numbe % % % % 25.99% % % % % % 0 Deadband Setting [mhz] 36 mhz NERC Policy One 7 For all individual generators Eastern Interconnection Phase 2 (including nuclear) 400 MW or larger, July 2014 Survey BAs will request generators to Generation November 2014 provide a completed generator Changes (.036 and survey of governor settings and Proportional) related data by January 1, For all individual generators(excluding nuclear and combined cycles steam turbines) 400 MW or larger, BA s will request generators to modify all dead bands greater than 36 mhz to at most 36 mhz and install proportional response by June 1, If generator advises it is unable to meet timing, then generator should provide reasonable time to complete. Phase 1 Jan 2014 Survey June 2014 Changes (.036 and Proportional) Total Number of Generators (4508) Greater than 400 (309, 6.83%) Between MW (1540, 34.16%) Nuclear (88, 1.95%) 8 4

5 For all individual generators 100 MW to 400 MW, BA s willrequest generators to provide a completed generator survey of current governor settings and related data by July 1, Phase 2 July 2014 Survey November 2014 Changes (.036 and Proportional) Eastern Interconnection Generation For all individual generators 100 MW to 400 MW, BA s will request generators to modify all dead bands greater than 36 mhz to at most 36 mhz and install proportional response by November 1, If generator advises it is unable to meet timing, then generator should provide reasonable time to complete. Phase 1 Jan 2014 Survey June 2014 Changes (.036 and Proportional) Total Number of Generators (4508) Greater than 400 (309, 6.83%) Between MW (1540, 34.16%) Nuclear (88, 1.95%) 9 Balancing Authority send Generator Survey Form to their Generators Balancing Authorities review data from Generators for accuracy and agreement Send completed survey form with Contact info to NERC Coordinating Effort with Modeling Working Groups 10 5

6 Troy Blalock, P.E. South Carolina Electric & Gas Company Transmission Operations Center 601 Old Taylor Rd. Cayce, SC Mail Code J42, Cayce, SC Office #: Cell #: Fax #:

7 Statistical Analysis of Variable Generation Matthew Varghese Resource Subcommittee Meeting October

8 Objectives Relating Variable Generation to CPS1 & CPS10 Metric Frequency Events from ERCOT & CPS1 metric Acknowledgement: Sydney Niemeyer minute Wind data for 1 month & CPS10 metric Acknowledgement: Terry Bilke 2 RELIABILITY ACCOUNTABILITY

9 1. Frequency Event Data & CPS1 CPS1 metric calculated for Event Data 262 Frequency Events in ERCOT from 2008 to 2013 Calculations of CPS1 by Sydney Niemeyer. Sample ERCOT Data Format for the Frequency Event File # UTC (t-0) B-A CPS1 MW Loss FR MW/0.1 Hz System LOAD MW Total Wind MW % of Wind/Total Generation 1 11/01/2008 6:32: , /06/ :20: , /17/ :01: , /21/2008 2:32: , /02/2008 5:20: , , /14/2008 7:23: , /16/ :45: , , /28/ :47: , , /16/ :13: , /25/ :19: , , RELIABILITY ACCOUNTABILITY

10 1. ERCOT Single BA Interconnection DEFINITIONS of CPS1 for a single BA Interconnection: CPS1 = 100* [2 - ((Frequency Error) 2 /( 1) 2 )] 1 = Hz for ERCOT CPS1 < 200 % if Frequency Error is non zero CPS1 is 200 % if Frequency Error is zero 4 RELIABILITY ACCOUNTABILITY

11 1. Explanation of Results Definition of p value With (1-p)% confidence, null hypothesis H0 is rejected For linear regression, the null hypothesis H0 is: The slope of the linear regression is zero CPS1 = 24.5*Wind/Total Gen The p value is With ( ) = 94.74% confidence, H0 is rejected CPS1 = *Total Wind MW The p value is With ( ) = 89.87% confidence, H0 is rejected 5 RELIABILITY ACCOUNTABILITY

12 1. Wind & the Up Direction Margin Wind typically operates at maximum output. With no margin in the up direction, Interconnections get benefit only for high frequency excursions from these generators. Can the CPS1 VG metric detect this lack of margin? Yes. The CPS1 VG metric is influenced by the efficiency of wind curtailment. If wind is ramping up as frequency declines, this results in good CPS1. If wind is curtailed as frequency increases, this also results in good CPS1. 6 RELIABILITY ACCOUNTABILITY

13 2. Multiple BA Interconnection Proposal - Calculating CPS10 of a fictitious wind BA with 10 Minute Data One approach is to determine the support of the wind in terms of CPS10. The aggregate wind is treated as a Balancing Authority (BA). The size of the BA is the MW value at t+n-1. If generation was 2000MW at t+n-1, think of it as a 2000MW BA. Its bias obligation then would be -20MW/0.1Hz. Wind BA s ACE as the difference between MW value at t+n and MW at t+n-1. If wind ramped up 20MW, it would look like an ACE of 20 MW. It should be straightforward to calculate CPS10 for the wind BA. If wind is ramping up as frequency declines, this results in good CPS10. If it provides droop with the decline, even better CPS10. 7 RELIABILITY ACCOUNTABILITY

14 2. Wind BA CPS10 & BAAL violations If 50 MW less, then assume the wind had an ACE of -50MW. Calculate a CPS10 for the average frequency for 10 minute blocks. BAAL triggers roughly at about -500% CPS10. BAAL is triggered in a 10 minute period, the wind is outside limits. BAAL violation is approximated by non-recovery of CPS10. If CPS10 does not recover to normal CPS within two subsequent periods. The wind dragged and didn't start move in the right direction for 30 minutes. 8 RELIABILITY ACCOUNTABILITY

15 2. Scatterplot of CPS10 BA vs. Wind by Terry Bilke 9 RELIABILITY ACCOUNTABILITY

16 2. CPS10 of BA vs. WIND by Terry Bilke Regression Analysis: CPS10 of BA versus CPS10 of WIND The regression equation is CPS10 of BA = CPS10 of WIND Predictor Coef SE Coef T P Constant CPS10 of WIND S = R-Sq = 2.2% R-Sq(adj) = 2.1% 10 RELIABILITY ACCOUNTABILITY

17 2. Summary of CPS10 for BA by Terry Bilke 11 RELIABILITY ACCOUNTABILITY

18 2. Summary of CPS10 for WIND BA by Terry Bilke 12 RELIABILITY ACCOUNTABILITY

19 2. Conclusions about WIND as a BA There is a small correlation between the BA s CPS10 and the wind. 41 periods over this month where CPS10 of the wind exceeded BAAL limits. So if the wind were a BA there were 41 violations for more than 30 minutes. It didn t trigger a problem for the BA. Future studies should consider a smaller BA that had a good share of wind. Absent some additional controls or regulation service, a wind farm would have a difficult time being a Balancing Authority. 13 RELIABILITY ACCOUNTABILITY

20 Background Material

21 1. CPS1 vs. Total Wind MW 15 RELIABILITY ACCOUNTABILITY

22 1. CPS1 vs. Total Wind MW Linear Regression Equation CPS1 = *Total Wind MW RELIABILITY ACCOUNTABILITY

23 1. CPS1 vs. % Wind/Total Gen 17 RELIABILITY ACCOUNTABILITY

24 1. CPS1 vs. % Wind/Total Gen Linear Regression Equation CPS1 = 24.5*Wind/Total Gen RELIABILITY ACCOUNTABILITY

25 Consortium for Electric Reliability Technology Solutions NERC Applications Status NERC Applications Status for Resources Subcommittee Gil Tam, Frank Carrera & Song Xue Sullivan s Island, SC October 29-30, 2013 Agenda Update on EPG/CERTS Support to NERC RS Effective November 1, 2013, Mr. Frank Carrera and Ms. Song Xue will assume responsibilities bl to provide support to NERC RS as follows: EPG day to day support for all NERC/CERTS applications will be coordinated by Pankaj Mishra under the supervision of Frank Carrera Song Xue will provide support on all on going grid performance analysis and reporting Monthly Frequency Response Event Reports Quarterly Interconnection Frequency Control Performance Reports Monthly BAAL Reports NERC Application Status t Summary Inadvertent Web based Application Status HQ Frequency Response Event Detection Threshold Analysis and Recommendation Page 1 1

26 NERC Applications Status Summary Application Resource Adequacy (ACE Frequency) Inadvertent Area Interchange Error (AIE) NERC Applications Status and Authorized Users (Many companies have several authorized users) Release 7.0 Current Production version 170 authorized users. Release 8.0 with updated Maps will be released 11/1/2013 Release 3.5 Current Production version 286 authorized users - New data entry website was published in July FRCC data through Oct have been uploaded in application data base. - Release 4.0 with updated Maps will be released 11/1/2013 Release 1.0 Current production version 122 authorized users Intelligent Alarms Frequency Monitoring and Analysis (FMA) Automated Reliability Reports (ARR) Release 1.0 Current production version 131 authorized users Release 2.5 Current production version 121 authorized users Release 1.0 Current production version 60 authorized users Monthly reports through September 2013 and Seasonal reports through Summer 2013 have been posted in ARR website. Page 2 CERTS/NERC Application Data Base Update All applications have been updated per the NERC CPS2 Bounds Report published on September 16, 2013 The maps in the Inadvertent and Resource Adequacy applications have been updated and will be released on 11/1/2013 EPG is committed to maintaining the accuracy of the Balancing Authority information within the applications twice a year in accordance with the published NERC CPS2 Bounds Report Page 3 2

27 New Web based Inadvertent Application Implementation EPG and NERC IT staff have completed the Web based Inadvertent Application Upgraded to new operating system (Windows 7 with 64 bits) User Guide is complete and posted on: Tabitha Hernandez completed Field Testing and identified several enhancements EPG released application to all authorized users on July 19, 2013 Users can continue to use old application Users of existing Inadvertent application can use same NERC assigned username and password to access new Web based application at following link: EPG has completed many of the recommended enhancements for the Inadvertent Website (details on next slide) User Training EPG can provide a one hour training session via WebEx, if necessary Page 4 Web based Inadvertent Application Enhancements Completed Inadvertent Worksheet: AddedTotals to Worksheet for Data EntryUsers Refrained the Blue Data entry screen from displaying by default Adding/Removing Interchange partners is reflected in the current month Inadvertent Reports: Removed Duplicate Month/Year line items Added a Pop Up with details on Reliability Region Status (Lock Status, Number of BA s Submitted) Added link within Pop Up to navigate directly to selected Region for detailed reports Unlocked drop down items to quickly change report criteria Page 5 3

28 Inadvertent Application Questions for NERC RS Should the Pseudo Balancing Authority be Removed? Was created by the Inadvertent Working Group circa 2008 as a crutch to balance the Eastern Interconnection on a monthly basis The addition of FRCC BA data negates the need for the Pseudo Balancing Authority Should all users be allowed to view Monthly Interconnection Reports? (A request from some BAs) Currently only Administrators have access to any reports in the system. Users have asked for reporting tools that span are above their privileges. Since the new Web based Inadvertent application is operational, should the old Inadvertent application be decommissioned? More users are having problem loading/using the old application. The Web based Inadvertent application has more functionalities than the old application except for the map. Eliminate duplicate application. New features can be incorporated in the new application. July 2013 Page 6 HQ Frequency Event Detection Threshold Analysis The current frequency delta threshold for HQ is 300 mhz The number of events detected have increased substantially over the last several months. 25 events were detected in July for a year to date total of 88 events CERTS/EPG analyzed the results of the last 9 months events detected to determine if the current frequency delta threshold of 300 mhz should be revised (see following slide) Page 7 4

29 Month Jan Feb Mar April May June July August September Total: HQ Frequency Event Detection Threshold Analysis Frequency Delta 300 mhz 350 mhz 400 mhz Gen - 3; Load - 1 Gen - 2; Load - 0 Gen - 1; Load Gen - 2; Load - 2 Gen - 2; Load - 0 Gen - 1; Load Gen - 1; Load - 1 Gen - 0; Load - 1 Gen - 0; Load Gen - 2; Load - 3 Gen - 2; Load - 1 Gen - 2; Load Gen - 4; Load - 6 Gen - 4; Load - 4 Gen - 3; Load Gen - 9; Load - 6 Gen - 9; Load - 3 Gen - 8; Load Gen - 15; Load - 10 Gen - 10; Load - 6 Gen - 9; Load Gen - 8; Load - 9 Gen - 3; Load - 5 Gen - 2; Load Gen - 2; Load - 4 Gen - 2; Load - 3 Gen - 2; Load Gen - 46; Load - 42 Gen - 34; Load - 23 Gen - 28; Load - 11 Page 8 HQ Frequency Event Detection Threshold Recommendation Based on results of analysis, CERTS/EPG recommends that the frequency event detection threshold for HQ be revised as follows: 400 mhz for loss of Generation events 350 mhz for loss of Load events Implementation of these revised thresholds will keep the number of events detected manageable and still provide sufficient number (2 to 3) of candidate events per month for BAL 003 The RS decided to revise the HQ detection threshold to 350 mhz for loss of Generation and for loss of Load events. Page 9 5

30 Q & A Page 10 EPG Personnel Contact Information NERC Application Support: Pankaj Mishra Mishra@electricpowergroup.com i (626) Ext. 136 Frank Carrera Carrera@electricpowergroup.com (626) Ext. 131 Grid Performance Analysis & Reporting: Song Xue Xue@electricpowergroup.com (626) Ext. 135 Page 11 6

31 Consortium for Electric Reliability Technology Solutions NERC Grid Performance Metrics Analysis Frequency Control & Time Error Correction Performance Report 3rd Quarter 2013 Song Xue & Gil Tam Sullivan s Island, SC October 29-30, 2013 Page 0 Interconnection Frequency Control Performance Reports Reports for all four interconnections through September 2013 were prepared p and submitted to the respective interconnection representatives and Larry Kezele for review Per RS request, EPG has summarized each performance metric for all four interconnections on a single slide to facilitate review The Time Error Correction report summary for WECC through September 2013 is included in this report If ERCOT, EI and HQ can provide time error correction dt data in spreadsheet format (i.e., date, start & stop time), EPG can produce similar time error correction reports EPG will continue to produce these reports for future RS quarterly meetings Page 1 1

32 Monthly CPS1, Page 2 Monthly CPS2, Page 3 2

33 Frequency Deviation Profile, * * Through September Only Page 4 Frequency Profile, * * Through September Only Page 5 3

34 Frequency Profile January through December * Through September Only Page 6 Frequency Error 3rd Quarter, * * Through September Only Page 7 4

35 Frequency Error 2 nd Quarter, * * Through September Only Page 8 Frequency Error 1 st Quarter, * * Through September Only Page 9 5

36 Frequency Error 4 th Quarter, Page 10 Daily Frequency Performance 2010 Page 11 6

37 Daily Frequency Performance 2011 Page 12 Daily Frequency Performance 2012 Page 13 7

38 Daily Frequency Performance 2013 Page 14 1 Minute Frequency Profile, * * Through September Only Page 15 8

39 10 Minute Frequency Profile, * * Through September Only Page 16 Daily RMS1, * * Through September Only Page 17 9

40 Daily RMS1 Boxplot, * * Through September Only Page 18 Daily Average Frequency Error Boxplot, * Through September Only Page 19 10

41 Daily RMS1 Control Plot, 2013* * Through September Only Page 20 Daily Average Frequency Error Control Plot * Through September Only Page 21 11

42 Y WECC TIME ERROR STATISTICS July 2013 Total Number of Time Error Corrections 21 Fast Time Error Corrections 10 Slow Time Error Corrections 11 Net Total Duration of TE Correction (Hrs) Hours Fast Time Error Corrections 59.5 Hours Slow Time Error Corrections 55.6 Hours Average Duration of Time Error Corrections(Hrs) 5.5 Hours Fast Time Error Corrections 6.0 Slow Time Error Corrections 5.1 Actual Time Error Correction Achieved Second s Page 22 August WECC TIME ERROR STATISTICS August 2013 Total Number of Time Error Corrections 13 Fast Time Error Corrections 5 Slow Time Error Corrections 8 Net Total Duration of TE Correction (Hrs) 78.8 Hours Fast Time Error Corrections 28.0 Hours Slow Time Error Corrections 50.8 Hours Average Duration of Time Error Corrections(Hrs) 6.1 Hours Fast Time Error Corrections 5.6 Slow Time Error Corrections 6.3 Actual Time Error Correction Achieved Second s Page 23 12

43 September WECC TIME ERROR STATISTICS September 2013 Total Number of Time Error Corrections 12 Fast Time Error Corrections 5 Slow Time Error Corrections 7 Net Total Duration of TE Correction (Hrs) 66.5 Hours Fast Time Error Corrections 30.6 Hours Slow Time Error Corrections 35.9 Hours Average Duration of Time Error Corrections(Hrs) 5.5 Hours Fast Time Error Corrections 6.1 Slow Time Error Corrections 5.1 Actual Time Error Correction Achieved Second s Page 24 WECC Time Error Statistics, 2013 Summary Annual Totals (YTD) Total Number of Time Error Corrections 137 Fast Time Error Corrections 73 Slow Time Error Corrections 64 Net Total Duration of TE Correction (Hrs) Hours Fast Time Error Corrections Hours Slow Time Error Corrections Hours Average Duration of Time Error Corrections(Hrs) 5.86 Hours Fast Time Error Corrections 6.00 Slow Time Error Corrections 5.69 Actual Time Error Correction Achieved Second s 13.9 minutes Page 25 13

44 Q & A Page 26 14

45 Daily Avg Daily Epsilon 1 (RMS) Eastern Interconnection RMS1 by Year (through October 21, 2013) Year Eastern Interconnection Daily Frequency Error (through October 21, 2013) Year

46 Sample Range Sample Mean Sample Range Sample Mean Xbar-R Chart of Daily Ave Freq Error by Week (10/1/12 to 10/21/13) UCL= _ X= LCL= Sample UCL= _ R= LCL= Sample Tests performed with unequal sample sizes Xbar-R Chart of Daily RMS1 by Week (10/1/12 to 10/21/13) UCL= _ X= LCL= Sample UCL= _ R= LCL= Sample Tests performed with unequal sample sizes

47 NERC Resources Subcommittee Eastern Interconnection Time Error Report 3Q Time Error Summary Y ear Mo onth # OF CO (FA ORRECTIONS AST) # OF CO (SL ORRECTIONS LOW) TOTAL HO OURS (FAST) TOTAL HO OURS (SLOW) AVERAGE (FA E DURATION AST) AVERAGE (SL E DURATION LOW) FEB MAR MAY JUN AUG SEP NOV DEC 1

48 400 Monthly Time Error Summary - Total Hours Total Hours on Correction Fast Slow 45 Monthly Time Error Summary - Number of Corrections Number of corrections Fast Slow 2

49 400 Yearly Time Error - Fast Corrections vs Total Hours Number of Corrections Number of Fast TEC's Total Hours Fast 3

50 ERCOT Frequency Control and Time Correction Report Resources Subcommittee Meeting Sullivan Island, SC October 30, 2013 Frequency Profile, CPS1 & 2 and Daily Frequency RMS1. ERCOT FREQUENCY CONTROL REPORT 1

51 NERC CPS ERCOT CPS1 By Day - September-2013 ERCOT Monthly CPS1 = Day CPS 1 Avg CPS 1 Peak Energy ERCOT Peak Energy NERC CPS 2 100% 98% 96% 94% 92% 90% 88% 86% 84% 82% 80% ERCOT CPS2 By Day - September-2013 ERCOT Monthly CPS2 = Day CPS 2 Avg CPS 2 Peak Energy ERCOT Peak Energy 2

52 170 ERCOT CPS1 CPS1 12 Month Rolling Average = CPS1 Average e Monthly Average 12 Month Rolling Average 100 ERCOT CPS2 Score* *ERCOT as a single control area is exempt from CPS2. These scores are For Information Only CPS Month CPS2 3

53 4 Daily RMS1 of ERCOT Frequency /1/2000 7/1/2000 1/1/2001 7/1/2001 1/1/2002 7/1/2002 1/1/2003 7/1/2003 1/1/2004 7/1/2004 1/1/2005 7/1/2005 1/1/2006 7/1/2006 1/1/2007 7/1/2007 1/1/2008 7/1/2008 1/1/2009 7/1/2009 1/1/2010 7/1/2010 1/1/2011 7/1/2011 1/1/2012 7/1/2012 1/1/2013 7/1/2013 Daily RMS1 of ERCOT Frequency /1/2004 4/1/2004 7/1/ /1/2004 1/1/2005 4/1/2005 7/1/ /1/2005 1/1/2006 4/1/2006 7/1/ /1/2006 1/1/2007 4/1/2007 7/1/ /1/2007 1/1/2008 4/1/2008 7/1/ /1/2008 1/1/2009 4/1/2009 7/1/ /1/2009 1/1/2010 4/1/2010 7/1/ /1/2010 1/1/2011 4/1/2011 7/1/ /1/2011 1/1/2012 4/1/2012 7/1/ /1/2012 1/1/2013 4/1/2013 7/1/2013

54 Daily RMS1 of ERCOT Frequency /1/2007 4/1/2007 7/1/ /1/2007 1/1/2008 4/1/2008 7/1/ /1/2008 1/1/2009 4/1/2009 7/1/ /1/2009 1/1/2010 4/1/2010 7/1/ /1/2010 1/1/2011 4/1/2011 7/1/2011 Comparing July 2010 vs July 2013 profile of frequency in 5 mhz bins 10/1/2011 1/1/2012 4/1/2012 7/1/ /1/2012 1/1/2013 4/1/2013 7/1/ <= HZ < Jul-10 Jul

55 Comparing August 2010 vs August 2013 profile of frequency in 5 mhz bins <= HZ < Comparing September 2010 vs September 2013 profile of frequency in 5 mhz bins <= HZ < Aug-10 Aug Sep-10 Sep-13 6

56 50000 ERCOT Frequency Profile Comparison January through December of each Year One Minute Occurances System Change Request 773 Adjust Generation To Be Dispatched by a percentage of the Regulation Service Deployed. Modification installed May 23, Metrics developed to determine improvement in utilization of Regulation Service. 7

57 Percent of 15 minute Inetervals when both Reg UP and Reg Down were Deployed % SCR 773 was implemented on May 23, % 87.19% 85.00% 80.51% 83.16% 80.87% 80.00% 77.77% 76.95% 75.90% 73.81% 75.00% 72.53% 71.57% 71.61% 71.33% 71.15% 70.43% 68.18% 71.78% 70.80% 70.19% 70.00% 68.62% 67.44% 66.04% 65.58% 65.09% 65.86% 66.87% 65.17% 64.89% 65.00% 63.51% 61.72% 62.43% 59.91% 59.86% 60.32% 60.00% 55.00% 53.47% 50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time Error, Number of Time Corrections and Total Hours in Time Correction. ERCOT TIME ERROR AND TIME CORRECTIONS PERFORMANCE 8

58 ERCOT Time Error Corrections July None. None. ERCOT Time Error Corrections August 9

59 None. ERCOT Time Error Corrections September None. ERCOT Time Error Corrections October through 28th 10

60 350 Monthly Cumulative TEC Monthly TEC Hours (Nov 2011 to Oct 2012) Monthly TEC Hours (Nov 2012 to August 2013) HOURS November 2.5 December January February March April May June July August September October Time Correction History Time Corrections Total Hours on Avg Hours Avg Corrections % Time on Year Month Days Fast Slow Count Control Per Correction Per Day Correction 2012 Jan % Feb % Mar % Apr % May % Jun % Jul % Aug % Sep % Oct % Nov % Dec % 2013 Jan % Feb % Mar % Apr % May % Jun % Jul % Aug % Sep % Oct % 11

61 Monthly TEC comparison Monthly TEC Hours (Nov 2011 to Oct 2012) Monthly TEC Hours (Nov 2012 to October 28, 2013) Hours November December January February March April May June July August September October 10 November-2011 ERCOT Time Error & Time Change October Loss - Daily Time - Gain 2 0 October-2012 Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Time Error Time Change Time Error 12

62 ERCOT Time Error & Time Change November October Loss - Daily Time - Gain 2 0 October-2013 Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Time Error Time Change Time Error Monthly Hours on Time Correction 90 Total Hours on Correction Between October 2001 and June 2013, only three months had zero TEC. Since June 2013 there havebeen 4 consecutive months with no TEC Hours on Control 13

63 Monthly Time Correction Summary Number of Corrections Slow Fast What Changed? Regulation Deployment feedback into Generation To Be Dispatched was added. ERCOT identified generators not following Startup procedures and worked with generators to improve performance. Load Frequency Control (LFC) tuning improved Regulation Deployments. Primary Frequency Response performance improved. More generators delivering sustained PFR. 14

64 Performance as measured by BAL for current year. ERCOT INTERCONNECTION FREQUENCY RESPONSE PERFORMANCE ERCOT Primary Frequency Response BAL Bias and Frequency Response Measure Page 1 Date/Time (t-0) BA BA BA Relay Lmt Value "A" Information Value "B" Information SEFRD (FRM) Time Bias R1 for Bias for R1 BA Time Zone DelFreq Time DelFreq DelFreq MW/Load Lost Adjustment MW/Load Lost Adjustment (MW/0.1Hz) (MW/0.1Hz) 12/10/ :32:54 CST :32: /12/2012 7:31:10 CST :31: /25/2012 4:30:20 CST :30: /2/ :41:26 CST :41: /13/ :35:06 CST :35: /15/ :04:14 CST :04: /19/ :10:16 CST :10: /22/ :06:24 CST :06: /4/ :47:48 CST :47: /7/ :05:56 CST :05: /15/ :39:50 CDT :39: /18/ :21:22 CDT :21: /2/ :25:34 CDT :25: /9/ :21:40 CDT :21: /23/ :19:30 CDT :19: /29/ :59:14 CDT :59: /7/2013 3:07:26 CDT :07: /11/ :13:48 CDT :13: /17/ :38:22 CDT :38: /22/ :11:30 CDT :11: /29/ :47:34 CDT :47: /30/ :46:22 CDT :46: /4/ :04:26 CDT :04: /11/ :56:36 CDT :56: /12/ :45:22 CDT :45: /25/ :47:46 CDT :47: /26/ :09:56 CDT :09: /21/ :00:50 CDT :00:

65 ERCOT Primary Frequency Response BAL Bias and Frequency Response Measure Page 2 Date/Time (t-0) BA BA BA Relay Lmt Value "A" Information Value "B" Information SEFRD (FRM) Time Bias R1 for Bias for R1 BA Time Zone DelFreq Time DelFreq DelFreq MW/Load Lost Adjustment MW/Load Lost Adjustment (MW/0.1Hz) (MW/0.1Hz) 7/25/2013 1:15:52 CDT :15: /2/ :35:08 CDT :35: /11/2013 5:43:36 CDT :43: /12/2013 2:58:06 CDT :58: /18/ :45:34 CDT :45: /20/ :19:12 CDT :19: /24/ :46:10 CDT :46: /29/ :55:24 CDT :55: /3/ :25:10 CDT :25: /25/2013 9:17:30 CDT :17: /8/ :56:08 CDT :56: /12/ :23:26 CDT :23: FRM - Average Estimated Frequency Response MW/0.1 Hz using SEFRD for R FRM - Regression Estimated Frequency Response MW/0.1Hz using SEFRD for R FRM - Median Estimated Frequency Response MW/0.1Hz for BA Compliance to R1, minimum Frequency Response Historical and current year to date totals. ERCOT INTERCONNECTION TOTAL ENERGY AND TOTAL ENERGY FROM WIND RESOURCES 16

66 ERCOT Total Energy 45,000,000 40,000,000 35,000,000 30,000,000 MWH 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 January February March April May June July August September October November December ERCOT Total Energy from Wind Generation 4,000,000 3,500,000 3,000,000 2,500,000 MWH 2,000,000 1,500,000 1,000, ,000 0 January February March April May June July August September October November December

67 ERCOT % Energy from Wind Generation 16.00% 14.00% 12.00% 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% January February March April May June July August September October November December Questions? 18