Frequency Response and Standard background Overview of BAL-003-1

Industry Webinar BAL-003-1 Draft Frequency Response Standard and Supporting Process July 18, 2011

Agenda Frequency Response and Standard background Overview of BAL-003-1 What s changing Field Trial Frequency Response Survey Form 2 Frequency Response Survey Form 1 Other field trial activities Questions Next Steps 2

Background FERC Order 693 directed specific changes for BAL-003 SAR drafted to address the issues identified by FERC NERC OC authorized Frequency Response field trial to help standard development at the June 7 8, 2011 OC meeting 3

Frequency Response Basics Frequency Response, termed beta (β), is A fundamental reliability service A combination of governor and load response Following a generator (or load loss) β stabilizes frequency at a new point Frequency does not return to 60 Hz until the contingent Balancing Authority replaces the loss 4

Typical Frequency Events Typical Events (5 seconds before unit trip to 60 seconds thereafter) 5

Bias vs. Frequency Response Frequency Bias (B) is not the same as Frequency Response (β) Frequency Response is actual MW contribution to stabilize frequency Bias is an approximation of β used in the ACE equation (prevents AGC withdrawal of β) Both are negative numbers (as frequency drops, MW output increases and vise versa) Both are measured in MW/0.1Hz Bias (absolute value) must be > β (absolute value) In the East, B (absolute value) is about 2-3 x as large as β (absolute value) Bias (absolute value) must be at least 1% of Balancing Authority peak load under current standard If there is to be a difference between B and β, it is preferable to be over-biased 6

FRS Original Goals Original SAR Objectively benchmark and track BA and Interconnection performance Confirm trends Learn why and where response is low Additional generator level data from low responders Accurate data for models Enable technically sound decisions on setting any future performance obligations FERC Order No. 693 directed additional work 7

BAL-003-1 Overview Drafting Team proposing five Requirements (only one is a significant ifi change) Frequency Response performance obligation Frequency Bias Setting Implementation AGC to be operated in Tie-Line Bias Appropriate Frequency Bias Setting for those providing Overlap Regulation Service, Minimum Frequency Bias Setting NERC has committed to file BAL-003-1 inthe NERC has committed to file BAL 003 1 in the first quarter of 2012 8

Reasons for Changes FERC directives in Order No. 693 Determine the appropriate periodicity of frequency response surveys Define the necessary amount of Frequency Response needed for Reliable Operation for each balancing authority with methods of obtaining and measuring that the frequency response is achieved Details are described in the Background Document that will be posted with the Standard 9

Requirement R1 BA to provide an average (median) amount of Frequency Response for defined set of events Frequency Response Obligation (FRO) is defined for upcoming year (based on BA size) BA reports performance at the end of the year for frequency excursions during the year Most, if not all BAs should have no trouble meeting FRO Generally sufficient Frequency Response in each Interconnection Standard provides mechanisms to obtain response 10

R2-R5 R5 Similar to Today 2. Implement Frequency Bias Setting on date specified by NERC 3. AGC operation in Tie-Line Bias, unless detrimental to reliability 4. Defines how Overlap Regulation providers implement Bias Setting 5. Identifies minimum Bias Setting Drafting team proposes 0.8% of peak/0.1hz Attachment B outlines process for changes to Bias Setting floor 11

Attachment A to the Standard Defines Process NERC will follow to Select events for analysis Allocate the Interconnection s ti Frequency Response Obligation to BAs NERC will publish lists of events during the year so BAs will have heads up on events that will be used BAs encouraged to develop local tools to scan for events and capture data for ongoing analysis 12

Allocation Methodology in Attachment A Determine FRO based on the average of forecasted peak load and installed capacity Formula FROBA = FROInt x 13

Estimating your FRO 1. Use the proposed FRO for your Interconnection Eastern Western Texas HQ Starting Frequency 60 60 60 60 Hz *Highest UFLS 59.6 59.5 59.3 58.5 Hz Contingency Protection Criteria 4500 2740 2750 1700 MW **Base Obligation 1125 548 229 113 MW/0.1Hz With 25% Safety Margin 1406 685 286 141 MW/0.1Hz 2. Multiply pythis value by: Your BA s Bias Setting Your Interconnection s Total Bias You can find Bias Setting values at: http://www nerc com/docs/oc/rs/2011%20cps2%20bounds%20report http://www.nerc.com/docs/oc/rs/2011%20cps2%20bounds%20report %20v6b(Update20110601).pdf 14

Key Field Trial Activities Frequency events have been identified for the period December 2010 through present utilizing Attachment A of the proposed standard NERC is requesting that the BAs fill out FRS Forms 1 and 2 for those events related to the interconnection in which the BA operates This data request was approved by the NERC Operating Committee at their June 7 8, 2011 meeting 15

Other Field Trial Activities 2010 NERC Recommendations (Alerts) on governors and BA Frequency Response 2011 Bias calculation 2012 Bias calculation (following Attachment A process) Evaluate impact due to change from 1% to 0.8% floor for Bias Setting Adjust based on observations 16

Candidate Frequency Event Detection Combination of tools used: Univ. of Tenn. Enhanced FNet System High-speed sensors (10 samples/second) Detects frequency excursions and estimates gen or load lost Triangulates to estimate location of disturbance Records data & calculates key frequency response factors CERTS Resource Adequacy (RA)Tool Smart Alarm System Based on 1-minute average frequency ICCP readings Indicates frequency deviations and alarms beyond trigger levels Provides BA-level variances in ACE CERTS Frequency Monitoring and Analysis (FMA) Tool After-the-fact analysis of detected and recorded data 17

Frequency Triggers for Detection Frequency Deviation Eastern ±60 mhz Western ±80 mhz Texas ±100 mhz Resource / Load Loss (FNet Estimate) Eastern ±1,000 MW Western ±700 MW Texas ±450 MW 18

FNet Sample Detection Output Trace of All Eastern Interconnection FDRs Courtesy of Power Information Technology Laboratory, University of Tennessee 19

FNet Sample Output 5 Point Median Eastern Interconnection Point A 60.0080 Hz Point C Point B 59.9492 Hz 59.9541 Hz Estimated Resource Loss 1,360 MW Actual Resource Loss 1,178 MW Delta Frequency A to C 58.8 mhz A to B 53.9 mhz Interconnection Frequency Response A-C -2,003 MW/0.1 Hz A-B -2,186 MW/0.1 Hz Courtesy of Power Information Technology Laboratory, University of Tennessee 20

Frequency Event Selection Criteria 1. Freq. Resp. Measurement Periods (12 months) December 1 through November 30 2. At least 25 Freq. excursions/interconnection Older events used if necessary 3. Event stabilization must be in less than 18 seconds, AND: Point C Point C Interconnection Delta F (Hz) Under Freq. Over Freq. Eastern 004 0.04 <5997 59.97 >6003 60.03 Western 0.04 < 59.97 > 60.03 Texas 0.15 < 59.97 > 60.03 Quebec 0.04 < 59.97 > 60.03 21

Frequency Event Selection Criteria 4. Pre-disturbance frequency stable and ~60 Hz Value A calculated for -16 sec. to 0 sec. before transient 5. Events with a second event before first scan of Value B not used 6. Exclude frequency events near top of hour when large changes in load or interchange are occurring 7. Use cleanest 2-3 frequency excursions each month older or other events from same season may be substituted Other detected candidate events may be used by ERO to evaluate interconnection frequency response performance 22

Introduction of FRS Form 2 FRS Form 2 receives BA AGC scan rate data to measure the BA s primary frequency response performance to a single, identified frequency perturbation on the system. Measures the change in Net Actual Interchange and the change in frequency due to the perturbation. Standardizes the measurement process for all BAs in all Interconnections. The field trial is testing 5 different average periods in this measure. 23

Objective We will review the function of FRS Form 2. The general layout where input from the BA is required. What data is required for each event. Where this data is placed in Form 2. How to execute the calculation of frequency response performance. File naming convention. 24

FRS Form 2 Function 60.08 60.06 Monday, October 12, 2009 1.171 Balancing Authority My BA Initial P.U. Performance 2.046 Initial P.U. Performance Adjusted 3850.0 20 to 52 second Average Period 60.04 60.02 3788.85 3800.0 Frequ uency Hz 60 59.98 59.96 59.94 59.92 59.9 59.88 59.86 1. Aligns the data for averaging for each average period. 3715.464 59.8888 3750.0 MW 3700.0 59.84 3650.0 59.82 59.8 59.78 59.76 59.74 2. Calculates pre and post event average change. 3645.3030 3. Calculates l expected response to meet the FRO and actual response. 3600.0 59.72 3550.0 2:26:24 2:26:34 2:26:44 2:26:54 2:27:04 2:27:14 2:27:24 2:27:34 2:27:44 2:27:54 2:28:04 2:28:14 2:28:24 Hz Average Frequency MW Average MW EPFR for FRO Adjusted 25

FRS Form 2 Workbook Tabs Entry Data Entry Data Tab: General BA information and evaluation execution. 26

FRS Form 2 Workbook Tabs - Data Data Tab: BA Event scan rate data placed in specific columns. 27

FRS Form 2 Workbook Tabs - Evaluation Evaluation Tab: All calculations of performance with data tables for averaging data and graphing are in columns A through DR. 28

FRS Form 2 Workbook Tabs Form 1 Data Form 1 Summary Data Tab: This single event data and performance results needed for FRS Form 1 is contained in row 7, columns A through CV. 29

FRS Form 2 Workbook Tabs - Instructions Instructions Tab: Detailed instructions for each step of the process. 30

FRS Form 2 Spreadsheet Setup Entry Data 1. Enter your BA s name (pneumonic) in cell B1 and Frequency Response Obligation (FRO) in cell B2 in the green highlighted cells of the Entry Data worksheet. Save the file as a master copy for your BA. Balancing Authority Name: My BA Balancing Authority Frequency Response Obligation (FRO from FRS Form 1) 80 Note: See "Instruction" tab for more detailed instructions. Step 1. Copy and Paste Event Data into the appropriate cells of the "Data" worksheet. Maintain date and time format of mm/dd/yy hh:mm:ss. Determine Time of T(0) and edit formula in cell "C8" to reference the correct row of Step 2. the "Data" worksheet. 2:27:24 T(0) is the first change in frequency of about 0.010 Hz (10 mhz) which should be the first scan of frequency data of the event. Cell beg wor refe freq Step 3. Time of Frequency Recovery to 60 Hz or Pre Perturbation Hz 2:33:08 Step 4. Enter MW output of generator or load that caused event (+ for gen loss, for load loss) (Value from NERC Event List. If multiple units, enter total MW loss.) 633 MW Hit the big blue button to copy your data for pasting into FRS Form 1 "BA Event Data" Step 5. worksheet. Copy Form 2 Data for Pasting into Form 1 60.1 60.05 60 59.95 59.9 59.85 59.8 59.75 59.7 Step 6. Paste data into FRS Form 1 in the appropriate row on the "BA Event Data" worksheet. Step 7. Save this workbook using the following file name format:nyiso_yymmdd_hhmm_frs_form2.xlsx 10/12/09 Date mmddyy 2:27 Time hh:ss of T(0) 31

FRS Form 2 Spreadsheet Layout - Data Critical Data: Time, frequency and NAI. Important Data: Transferred Frequency Response, Contingent BA Loss, Bias Setting and BA Load JOU Non Transferred Contingent Net Dynamic Conforming Pumped Ramping Frequency BA BA BA Actual Schedules Load Hydro Units Response Lost Generation Bias Load Interchange Imp( ) Exp (+) Load ( ) Load ( ) Gen (+) Gen (+) Rec ( ) Del (+) Load ( ) Gen (+) Setting Time (T) Hz MW MW MW MW MW MW/0.1 Hz MW MW/0.1 Hz MW 10/12/09 02:12:00 59.981 3669.878418 350 351.361511 0 0 10 15 103 7500 10/12/09 02:12:04 59.981981 3671.699707699707 350 351.361511361511 0 0.5 10 15 103 7500.33 10/12/09 02:12:08 59.98 3671.698486 350 351.361511 0 1 10 15 103 7500.66 Optional Data: Adjustments JOU, Non-conforming load, Pumped Hydro, Ramping Units. DO NOT delete columns for any optional data not utilized or any non-applicable data for Transferred Frequency Response, Contingent BA Loss or Load,. Leave data as zero or blank. 32

Data Sign Convention Net Actual Interchange +/- Joint Owned Unit Import (-) Export (+) Non-Conforming Load (-) Option on Evaluation worksheet to flip sign to positive in cell AS1. Enter + if your load data is positive. Enter - if your load data is negative. Pumped-Hydro Load (-) Gen (+) Ramping Units Gen (+) 33

Sign Convention part 2 Transferred Frequency Response Receiving i BA Rec (-) Delivering BA Del (+) Interconnection Transferred Frequency Response should net to zero. Contingent BA Lost Load (-) Lost Generation (+) BA Bias Setting (-) BA Load (+) 34

Form 2 Pasting Data into Data Worksheet Historian data collection formulas should be removed from Form 2 before submitting. Use PasteSpecial/Values to strip formulas and only paste data values into the worksheet. Do not delete any data columns. Provide a minimum of 5 minutes of data before the beginning of the event and a minimum of 15 minutes of data after the event. The spreadsheet will easily accommodate 60 minutes of data. 35

FRS Form 2 - Execution After pasting event data into Form 2, determine the time of the beginning of the event. Observe the frequency graph on the Entry Data worksheet for the approximate time of the event. On the Data worksheet, scroll down the frequency data column to the approximate time to determine the first change in frequency at the beginning of the event. Note the row number in the Data worksheet that this occurs. 36

FRS Form 2 Execution Beginning of Event Row 237 contains the first change in frequency identifying the beginning of the event. 37

FRS Form 2 Execution Set Event Beginning g Return to the Entry Data worksheet and edit the formula in cell C8 to reference the row number (237) in the previous step. Select cell C8 and edit formula here to reference row identified in previous step. Press enter. 38

FRS Form 2 Execution Event End Time Return to the Data worksheet and determine when frequency returns to 60.00 Hz, schedule or the pre-event frequency. Note the row number on the Data worksheet that this occurs. 39

FRS Form 2 Execution Find Event End Time Row 323 contains the recovery frequency identifying the end of the event. 40

FRS Form 2 Execution Set Event End Time Return to the Entry Data worksheet and edit the formula in cell C11 to reference the row number (323) in the previous step. Select cell C11 and edit formula here to reference row identified in previous step. Press enter. 41

FRS Form 2 Execution complete If available from Form 1, enter the size (in MW) of the generator or load that caused this event into cell C15 of the Entry Data worksheet. This is optional for the Interconnection ti performance evaluation. Execution is now complete. Performance of your BA for all the average periods in the field trial is ready to be copied into FRS Form 1. Perform a mouse click on the Copy Form 2 Data for PastingintoForm1 into macro on the Entry Data worksheet. The correct data is now copied. 42

Alternate Copy of Performance Data If you prefer not to enable macros. Navigate to the Form 1 Summary Data worksheet. Select cells A7 through h CV7 Select copy. PasteSpecial/Values into FRS Form 1. 43

FRS Form 2 File Naming Convention Use the following file name convention. Xxxxx_yymmdd_hhmm_FRS_Form2.xlsm Where xxxxx is your BA pneumonic yy is year mm is month dd is day hh is hour mm is minute 44

Introduction of FRS Form 1 One FRS Form 1 for each Interconnection will be generated and posted by NERC. Contains list of events: date and time that each BA in that Interconnection must measure performance by completing a FRS Form 2. As each FRS Form 2 is completed, the resulting BA performance data is added to FRS Form 1 by using the PasteSpecial/values action. Each BA creates a FRS Form 1 with the Each BA creates a FRS Form 1 with the performance of their BA to each event. 45

Objective FRS Form 1 layout BA data entry. FRS Form 2 data entry. FRS Form 1 file naming convention. 46

FRS Form 1 Layout Instructions Tab: Basic instructions for completing FRS Form 1. 47

FRS Form 1 Layout Data Entry Tab Selecting N will add the performance of the event to the yearly average. Enter appropriate information in the green cells. The reason s section will be entered only if adjustments are utilized in the analysis. Correct year information will automatically update when the 24 th frequency event is identified by NERC and added to the Event Date/Time column. 48

FRS Form 1 BA Form 2 Event Data PasteSpecial/Values your BA performance results from FRS Form 2 on the appropriate event row starting in row 7, column D through CY for all identified events. BA Form 2 Event Data Tab: The BA will enter the performance data from FRS Form 2 in this section. 49

FRS Form 1 Naming Convention Use the following file name convention. Xxxxx_yyyy_FRS_Form1.xlsm Where xxxxx is your BA pneumonic yyyy is year 50

Additional Worksheets in FRS Form 2 15 Worksheets are Hidden within FRS Form 2. Graphs of BA Initial performance as measured by each average period being tested in the field trial. Graphs of BA Sustained frequency response performance as a function of the FRO. Per Unit evaluation (P.U.) is used for this measure where 1.0 P.U. represents that the BA delivered exactly its FRO during the measurement period. Graphs of BA Adjustments and the Interconnection Frequency Response performance are available if the BA utilized this functionality. 51

To View Hidden Worksheets Right click any tab at the bottom of the workbook. Select Unhide. Select the worksheet to be displayed. Select ok. 52

Graph Example BA Performance 20 to 52 second evaluation. 60.08 P.U. performance is displayed. 60.06 (Adjusted and unadjusted). 60.04 60.02 Monday, October 12, 2009 1.171 Balancing Authority My BA Initial P.U. Performance 2.046 Initial P.U. Performance Adjusted 3850.0 20 to 52 second Average Period 3788.85 3800.0 Frequ uency Hz 60 59.98 59.96 59.94 59.92 59.9 59.88 59.86 59.84 59.82 59.8 59.78 59.76 59.74 The first change in frequency will always align in the center of this graph. 3715.464 Actual Net Interchange MW 3750.0 MW 3700.0 59.8888 The minimum frequency 3650.0 response to 3645.3030 meet FRO for The average periods of the this single event pre event and post event is displayed. 3600.0 data is highlighted. 59.72 3550.0 2:26:24 2:26:34 2:26:44 2:26:54 2:27:04 2:27:14 2:27:24 2:27:34 2:27:44 2:27:54 2:28:04 2:28:14 2:28:24 Hz Average Frequency MW Average MW EPFR for FRO Adjusted 53

Graph Example Sustained Graph Monday, October 12, 2009 My BA 0.895 Sustained P.U. Performance 60.08 3850.0 60.06 60.04 60.02 3800.0 60 59.98 3750.0 59.96 Freq quency Hz 59.94 59.92 59.9 59.88 59.86 59.84 59.82 The Event recovery period Target is calculated and displayed based on the BA s FRO, starting and ending NAI. 3700.0 MW 3650.0 3600.0 59.8 59.78 59.76 59.74 The Ramp MW trend is calculated based on beginning and ending NAI adjusted for starting and ending frequency. 3550.0 59.72 3500.0 2:26:24 2:27:24 2:28:24 2:29:24 2:30:24 2:31:24 2:32:24 2:33:24 2:34:24 2:35:24 2:36:24 2:37:24 2:38:24 2:39:24 2:40:24 2:41:24 2:42:24 Hz Interchange MW Recovery Period Target MW Recovery Period Ramp MW 54

Question & Answer 55

Working Documents Draft BAL-003-1 Draft Attachment A to BAL-003-1 Draft Attachment B to BAL-003-1 Draft FRS Form 1 Draft FRS Form 2 BAL-003-1 Background Document 56

Next Steps Download FRS Forms 1 and 2 for your Interconnection and Scan Rate Files can be found at: http://www.nerc.com/filez/standards/frequency_resp onse-rf.html Extract data (5 minutes before to 15 minutes after) for each event Populate FRS Forms 1 and 2 (one FRSForm 2 for each event) The address to send FRS Forms 1 and 2 will be distributed with the announcement of event dates. 57

Contact Information If you have questions related to the data request, contact: Questions concerning FRS Form 1 and FRS Form 2 Sydney Niemeyer (T) 713.537.3715 (E) sydney.niemeyer@nrgenergy.com Questions concerning the Standard and Associated Documents Darrel Richardson (T) 609.613.1848 (E) darrel.richardson@nerc.net 58