Objectives - Electric Reliability Council of Texas

Objectives - Electric Reliability Council of Texas

WIND OPERATIONS GENERATION CHALLENGES AND EXPECTATIONS OPERATIONS PLANNING STAFF ERCOT Objectives List challenges in forecasting wind in Texas. List actions a Wind facility experiencing icing issues should take. Given an example of telemetry being received from a renewable resource, determine if it is an issue and whether the QSE representing this resource should be contacted to rectify the issue. PUBLIC 2 Wind Generation and Wind Forecasting in ERCOT PUBLIC 3 Wind Growth Potential PUBLIC 4 Wind Regions within ERCOT

Panhandle: 4,037 MW North: 1,049 MW West: 10,398 MW South: 1639 MW PUBLIC Coast: 2,161 MW 5 Wind Records PUBLIC TIME LOAD (MW) WIND GENERATION (MW) PENETRATION AT RECORD TIME 03/31/2017 20:56 40,863

16,141 39.50% 03/23/2017 03:50 28,780 14,391 50.00% 6 Wind Generation Forecast AWS Truepower computes a Short Term Wind Power Forecast (STWPF) for all Wind-powered Generation Resources (WGR) in ERCOT for a rolling 168 hour timeframe. This process uses information from multiple systems, including: RARF (Resource Asset Registration Form) (ex. site geo-location, met tower geo-location) EMS (Energy Management System) (ex. Resource Status, telemetered site specific meteorological data) Outage Scheduler (ex. WGR derate start/stop dates, derate values) When submitting COPs, Wind QSEs must use this STWPF for the applicable 168 hours timeframe. (COP HSL must be <= STWPF.) Note that NPRR 785, upon approval will only require Wind QSEs to update a COP only when wind farm operating conditions dictate COP HSL to be lower STWPF. *Note that for a WGR going through commissioning process, ERCOT will start providing STWPF to the QSE once WGR receives approval to Generate into the ERCOT Grid (Part II of the Commissioning check list) PUBLIC 7

NPRR 785 Synchronizes Short Term Wind Forecast (STWPF) with Current Operating Plan (COP) of Wind Generation Resources (WGRs) and Photovoltaic Generation Resources (PVGRs) ERCOT will automatically prepopulate every WGR and PVGRs COP with most recent forecast for the next 168 hours. QSEs representing WGRs or PVGRs will be required to either submit a COP with the automatically prepopulated forecast or submit a lower number (when wind farm operating conditions dictate COP HSL to be lower STWPF). PUBLIC 1 0 Challenges in Forecasting Wind Relatively concentrated geographic region for wind installation (lack of spatial diversity) Texas weather can change rapidly Severe changes in wind generation output caused by different types of extreme weather (large wind ramps) Frontal system, trough, or dry line/Thunderstorms/Low-level jets/Weakening pressure gradients/Strengthening pressure gradients There is an approximation of hypothetical wind speed-power curve to the real one Hypothetical Power Curve Actual Wind Farm Power *The steep part of the power curve from 4-12 m/s is where power increases strongly with speed.

PUBLIC 1 1 Wind Forecast Errors (Day-Ahead) Day Ahead Wind Forecast Performance MAPE = MEAN ABSOLUTE %ERROR 14.0% Jun; 12.3% 12.0% 10.0% MAPE Jan; 8.7% Jan; 8.3% 8.0% 6.0% 6.0% Jan; 5.8% Apr; 10.4% Feb; 10.0% Feb; 9.7% Feb; 8.5% May; 9.9%

Jun; 10.2% Mar; 9.0% Apr; 8.8% May; 9.1% Mar; 8.5% May; 7.6% Mar; 7.1% Apr; 7.2% 6.9% 6.7% 6.0% 5.7% Nov; 8.6% Sep; 7.3% Sep; 7.0% Jul; 7.2% Jul; 6.6% Jun; 5.6% 5.6% 5.2% Jul; 4.7% Aug; 6.0% 5.8% Aug; 5.6% Oct; 6.3% 5.1% Sep; 5.0% Aug; 4.5%

4.0% Oct; 7.5% Oct; 5.0% Nov; 7.8% Nov; 6.7% 5.3% 5.3% Nov Dec 4.0% 2.0% 0.0% Jan Feb Mar Apr 2013 PUBLIC May 2014 Jun

2015 Jul Aug 2016 Sep Oct 2017 1 2 Wind Forecast Errors (Hour-Ahead) Hour Ahead Wind Forecast Performance MAPE = MEAN ABSOLUTE %ERROR 8.0% 7.0% Jun; 6.7% Apr; 6.4% 6.0%Jan; 5.8% Feb; 6.0% Feb; 5.7% Mar; Mar; 5.8% 5.7%

MAPE 5.0%Jan; 5.0% Feb; 4.46% 4.0% 3.84% Jan; 3.55% 3.52% May; 6.3% Jun; 6.1% May; 5.8% Nov; 5.7% Nov; 5.5% Apr; 5.5% Apr; 5.10% May; 5.23% 4.26% 4.51% Jul; 5.1% Jul; 4.7% Jun; 3.95% 3.88% 3.81% Mar; 3.60% Sep;

Sep; 4.6% 4.5% Aug; 4.2% 3.98% Aug; 3.9% 3.64% Jul; 3.39% Aug; 3.36% 3.40% 3.0% Oct; Oct; 4.8% 4.7% Oct; 3.84% Nov; 4.01% 3.50% 3.30% Sep; 2.84% 2.70% 2.0% 1.0% 0.0% Jan Feb Mar Apr 2013 PUBLIC

May 2014 Jun 2015 Jul Aug 2016 Sep Oct Nov Dec 2017 1 3 Objectives List challenges in forecasting wind in Texas. List actions a Wind facility experiencing icing issues should take. Given an example of telemetry being received from a renewable resource, determine if it is an issue and whether the QSE representing this resource should be contacted to rectify the issue. PUBLIC 1 4

Wind Generation Cold Weather Preparation PUBLIC 1 5 Icing Impacts Source: Wind Turbines in Cold Climates: Icing Impacts and Mitigation Systems, http://www.springer.com/us/book/9783319051901 PUBLIC 1 6 Wind Units Failure in Cold Winter Frigid Temperatures Exceeds Turbine Limits Wind turbines are typically designed to operate within ambient air temperatures of -15C/-20C (5F/4F) Wind turbines have an automatic shutdown feature to protect components if that range is exceeded. Blade Icing Icing on wind turbines affects three different aspects simultaneously the design (aerodynamics, load, control system, and material) the safety (ice throw, unbalanced rotor spinning, over-power, and fatigue) performance (annual energy output, wind measurements, and design life duration) Other impacts wind sensorsrendering ineffective wind-measuring equipment increase noise levels and generally decrease a turbines cost-effectiveness. PUBLIC 1 7

Safety PUBLIC 1 8 Cold Weather Extreme Package This ensures that wind turbine operates in temperatures as low as -30C (-22F), and in survival mode without operation, at temperatures as low as -40C (-40F) (one example). Minimum temperature (standard) operational / survival Standard weather package -15C/-20C (5F/-4F) Cold weather package -30 C/-40 C (-22F/-40F) ERCOT is aware of at least one vendor that offers an extreme cold weather package. Resource Entities should check with their manufacturer. PUBLIC 1 9 Icing Prevention Technologies Passive icing prevention methods Rely on the physical properties of the blade surfaces to prevent ice accumulation. An example is application of an anti-adhesive coating on the blade such as teflon. Another approach takes advantage of the heat absorbing capacity of dark colored surfaces and consists in the use of black coated blades.

Active de-icing methods Consist of thermal, chemical and impulse de-icing In thermal de-icing, electrical elements, similar to the one found on the rear window of a car, can be used to warm and melt the ice accumulation off the blades. In a comprehensive wind turbine icing prevention approach, sensors that could detect the build-up of ice on the rotor could be considered. PUBLIC 2 1 Cold Weather Preparation Inclement Weather Plan Preventative Maintenance Schedule Emergency Plan Communication PUBLIC 2 2 Wind Operations During Icing Conditions PUBLIC 2 3

Challenges Created by Icing Icing creates a mismatch between the forecast and actual production. This creates an error in the forecast which then was provided to WGRs. This error may be used in the COP which feeds into RUC. Could lead to ERCOT counting on more wind than WGRs are able to generate. PUBLIC 2 4 Icing Event Dec 26-27, 2015 Freezing temperatures on December 26, 2015 in the panhandle region of Texas marked the beginning of icing conditions that lasted throughout the duration of December 27 . By 3:00 AM on December 27, the wind forecasting error had reached about 1700 MW. Icing was continually reported throughout the day and forecasting error reached nearly 4500 MW by the hour ending HE0800. The outage scheduler reported new WGRs related derates and forced outages totaling around 6000 MW by the end of the day. PUBLIC 2 5 Wind Outages 12/26 12/27 PUBLIC 2 6 West Wind Forecasts Over Time HE8 Observed: 2,285 Last COP: 5,864 6 hour: 6,937

PUBLIC 2 7 West Wind Forecasts Over Time HE19 Observed: 1,352 Last COP: 1,774 6 hour: 3,110 12 hour: 5,592 00 PUBLIC 2 8 West Wind Hour Ahead Summary 10,000.0 9,000.0 8,000.0 7,000.0 6,000.0 5,000.0 4,000.0 3,000.0 2,000.0 1,000.0 0.0 1 2 3

4 5 6 7 8 9 Aggr COP PUBLIC 10 11 12 13 14 15 Forecast Error RT Aggr Wind-Output 16 17

18 19 20 21 22 23 24 STWPF 2 9 Panhandle Wind Hour Ahead Summary 2,500.0 2,000.0 1,500.0 1,000.0 500.0 0.0 1 2 3

4 5 6 7 8 9 Aggr COP PUBLIC 10 11 12 13 14 Forecast Error RT Aggr Wind-Output 15 16 17

18 19 20 21 22 23 24 STWPF 3 0 Forecast vs. COP Comparison West Wind Hour Ahead Summary Panhandle Wind Hour Ahead Summary 10,000.0 2,500.0 9,000.0 2,000.0 8,000.0 7,000.0 1,500.0

6,000.0 5,000.0 1,000.0 4,000.0 3,000.0 500.0 2,000.0 1,000.0 0.0 1 3 5 7 9 11 13 STWPF RT Aggr Wind-Output PUBLIC 15 17

19 21 Forecast Error Aggr COP 23 0.0 1 2 3 4 5 6 7 8 9 101112131415161718192021222324 STWPF RT Aggr Wind-Output Forecast Error Aggr COP 3 1 Forecast Performance Notice the disparity between the last forecast value before the adjustment period and the observed value in HE 8 and 12 versus in HE 19. Delayed outages caused these errors in earlier hours which were then minimized in HE 19. The faster these outages are entered, the faster they are incorporated to the forecast. Panhandle Wind 2500 West Wind 8000 7000 2000

6000 5000 1500 4000 1000 3000 2000 500 1000 0 HE 3 HE 8 HE 12 HE 19 5 Hour Forecast MW Last Forecast Before Adjustment Period MW Observed PUBLIC 0 HE 3 HE 8

HE 12 HE 19 5 Hour Forecast MW Last Forecast Before Adjustment Period MW Observed 3 2 Units STWF with no Outage Updates WRONG Forecast NEVER Converged to zero PUBLIC 3 3 Units STWF with Outage Updates CORRECT Forecast Converged to zero PUBLIC 3 4 Telemetry During Icing Conditions WRONG

Number of offline turbines *Remember this is under icing conditions. Number of online turbines While the NTON line does decrease, it does not fall as the units MW goes to zero, but is very delayed. MW output PUBLIC This is too slow. 3 5 Telemetry During Icing Conditions CORRECT Number of offline turbines PUBLIC *Remember this is under icing conditions. Number of online turbines

The NTON line falls as the units MW goes to zero. MW output This is very good. 3 6 Addressing the Wind Turbine Icing Use the following when a WGR is experiencing an icing related event, Call the ERCOT Control Room and notify them. Ensure the Plant has a Communication Plan to inform QSE & ERCOT of the event. Continuously update the telemetry for number of turbines on and off (NTON and NTOFF) as conditions on the ground deteriorate. If the outage/derate is expected to last greater than 2 hours, submit this into ERCOTs Outage Scheduler. In comparison to turbine telemetry, Outage submissions have a greater impact in correcting the forecast. 3.1.4.5 Notice of Forced Outage or Unavoidable Extension of Planned or Maintenance Outage Due to Unforeseen Events (2) Any Forced Outage that occurs in Real-Time must be entered into the Outage Scheduler if it is to remain an Outage for longer than two hours . PUBLIC 3 7 Objectives List challenges in forecasting wind in Texas. List actions a Wind facility experiencing icing issues should take.

Given an example of telemetry being received from a renewable resource, determine if it is an issue and whether the QSE representing this resource should be contacted to rectify the issue. PUBLIC 3 8 Current and Ongoing Telemetry Concerns PUBLIC 3 9 Meteorological Telemetry Requirements Meteorological telemetry must be telemetered with the specified data units at the refresh rates mentioned below. TELEMETRY NAME Wind Speed (MPH) Wind Direction (DEG) Temperature (TEMP) Barometric Pressure (PRES) Num. Of Turbines Available (NTON) Num. Of Turbines Unavailable (NTOF) Num. Of Turbines Unknown (NTUN) DATA UNIT REFRESH RATE MPH

Degrees (between 0 and 360) C mbar 10s Integer (Count of turbines) 10s 10s 10s 10s References: ERCOT and QSE Operations Practices During The Operating Hour (Section 3.9.1) ERCOT Nodal ICCP Communications Handbook PUBLIC 4 0 Telemetry Examples (from 9/6/2016) PUBLIC 414 1 Turbine Availability Telemetry Requirements Num. Of Turbines Available (NTON)

Num. Of Turbines Unavailable (NTOF) Should indicate count of turbines that are offline & not available Turbines that are under maintenance, or experiencing forced outages (caused by meteorological phenomenon like icing, hurricane, etc.) or mechanical issues) should be communicated as unavailable Num. Of Turbines Unknown (NTUN) Should indicate count of turbines that will generate power if the operating conditions are met regardless of what current output from the wind generation resource may be. If a turbine is disconnected from the grid to reach curtailed Base Point, it should be communicated as Available. should only be used in the few and temporary occasions when there is a failure with the wind plants communication system Sum of the telemetered number of turbines Available, Unavailable and Unknown should equal the total number of turbines identified in units RARF. Reference: ERCOT and QSE Operations Practices During The Operating Hour (Section 3.9.2) PUBLIC

4 2 Number of Turbines Online (NTON) WRONG CORRECT 35 40 140 30 35 120 30 25 100 25 20 80 20 15 60 15

10 40 10 20 0 0 0 MW HSL NTON NTOF 06:00 06:17 06:35 06:53 07:10 07:28 07:46 08:03 08:21 08:39 08:56 09:14 09:32

09:49 10:07 10:25 10:42 11:00 11:18 11:35 11:53 12:11 12:28 12:46 13:04 13:21 13:39 13:57 14:14 14:32 14:50 15:07 15:25 15:43 5 06:00 06:16 06:33 06:50 07:06 07:23 07:40 07:57 08:13 08:30 08:47 09:04

09:20 09:37 09:54 10:11 10:27 10:44 11:01 11:17 11:34 11:51 12:08 12:24 12:41 12:58 13:15 13:31 13:48 14:05 14:22 14:38 14:55 15:12 15:28 15:45 5 MW HSL NTON NTOF Wind farms interpret the NTON flag differently.

The NTON is meant to describe the number of turbines available for production, which has nothing to do with the actual output. The left image is taking the units MW value into account - Wrong The right image is expected method for computing and telemetering this data. PUBLIC 4 3 06:00 06:16 06:33 06:50 07:06 07:23 07:40 07:57 08:13 08:30 08:47 09:04 09:20 09:37 09:54 10:11 10:27 10:44 11:01 11:17 11:34 11:51 12:08 12:24 12:41 12:58

13:15 13:31 13:48 14:05 14:22 14:38 14:55 15:12 15:28 15:45 NTON Example 1 PUBLIC WRONG 35 30 25 20 15 10 5 0 MW HSL NTON

NTOF 4 4 06:00 06:17 06:35 06:53 07:10 07:28 07:46 08:03 08:21 08:39 08:56 09:14 09:32 09:49 10:07 10:25 10:42 11:00 11:18 11:35 11:53 12:11 12:28 12:46 13:04 13:21 13:39 13:57 14:14 14:32 14:50

15:07 15:25 15:43 NTON Example 2 35 PUBLIC The NTON line stays constant despite the units MW value going to zero. 40 140 CORRECT 120 30 100 25 80 20 60 15 10 40 5 20 0

0 MW HSL NTON NTOF 4 5 00:00 00:37 01:15 01:53 02:31 03:09 03:47 04:25 05:03 05:41 06:19 06:57 07:35 08:13 08:51 09:29 10:06 10:44 11:22 12:00 12:38 13:16 13:54 14:32 15:10 15:48

16:26 17:04 17:42 18:20 18:58 19:35 20:13 20:51 21:29 22:07 22:45 23:23 MW exceeding HSL telemetry MW telemetry must never exceed the HSL. When a WGR is UNCURTAILED its MW and HSL values should be exactly the same. 140 120 100 80 WRONG PUBLIC 60 40 20

0 MW HSL 4 6 MW exceeding HSL telemetry MW telemetry must never exceed the HSL. When a WGR is UNCURTAILED its MW and HSL values should be exactly the same. 120 100 80 CORRECT 60 40 20 00:00 00:41 01:22 02:03 02:44 03:26 04:07 04:48 05:29 06:10 06:52

07:33 08:14 08:55 09:36 10:18 10:59 11:40 12:21 13:02 13:44 14:25 15:06 15:47 16:28 17:10 17:51 18:32 19:13 19:54 20:36 21:17 21:58 22:39 23:20 0 MW PUBLIC HSL 4 7 WGR Under Curtailment ERCOT nodal protocols and Business Practices dictate that when under curtailment a

WGRs HSL must represent actual power generation potential. HSL should never lower dramatically under curtailment scenarios. 80 5 4.5 70 4 60 3.5 50 3 40 WRONG 2.5 2 30 1.5 20 1 10 0.5

0 19:00 19:03 19:07 19:11 19:15 19:19 19:22 19:26 19:30 19:34 19:38 19:41 19:45 19:49 19:53 19:57 20:00 20:04 20:08 20:12 20:16 20:19 20:23 20:27 20:31 20:35 20:38 20:42 20:46 20:50 20:54 20:57 0

MW PUBLIC HSL Curtailment Flag 4 8 WGR Under Curtailment ERCOT nodal protocols and Business Practices dictate that when under curtailment a WGRs HSL must represent actual power generation potential. HSL should never lower dramatically under curtailment scenarios. 160 5 4.5 140 4 120 3.5 100 3

80 2.5 2 60 CORRECT 1.5 40 1 20 0.5 0 12:00 12:28 12:56 13:24 13:53 14:21 14:49 15:17 15:46 16:14 16:42 17:10 17:39 18:07 18:35 19:04 19:32 20:00 20:28 20:57

21:25 21:53 22:21 22:50 23:18 23:46 00:14 00:43 01:11 01:39 02:08 02:36 03:04 03:32 0 MW PUBLIC HSL Curtailment Flag 4 9 18:00 18:03 18:07 18:11 18:15 18:19 18:22 18:26

18:30 18:34 18:38 18:41 18:45 18:49 18:53 18:57 19:00 19:04 19:08 19:12 19:16 19:19 19:23 19:27 19:31 19:35 19:38 19:42 19:46 19:50 19:54 19:57 WGR Under Curtailment WGRs must follow ERCOT Base Point when under curtailment 150 145 140 135

130 CORRECT PUBLIC 125 120 115 110 105 100 MW HSL Base Point 5 0 Objectives List challenges in forecasting wind in Texas. List actions a Wind facility experiencing icing issues should take. Given an example of telemetry being received from a renewable resource, determine if it is an issue and whether the QSE representing this resource should be contacted to rectify the issue. PUBLIC

5 1 Questions? PUBLIC 5 2

Recently Viewed Presentations

  • Local Authority Business License Reform - World Bank

    Local Authority Business License Reform - World Bank

    MOBILIZING LOCAL REVENUE FROM THE BUSINESS SECTOR Roy Kelly Duke University <[email protected]> Innovations in Local Revenue Mobilization World Bank Workshop Sponsored by the
  • The Lightning Thief Rick Riordan - Weebly

    The Lightning Thief Rick Riordan - Weebly

    Discussion Question. The Lotus Casino in Las Vegas is a modern-day version of the Land of the Lotus Eaters, which Odysseus visited on his way back from Troy.
  • Writing a Method Section - University of Massachusetts Lowell

    Writing a Method Section - University of Massachusetts Lowell

    Articulate what the method section of a psychology paper is. State what goes in that section. ... Participants will be a convenience sample of 30 couples who have been married or cohabiting at least 10 years, and are at least...
  • Chapter 10 Worked Example 1 Explain the trend

    Chapter 10 Worked Example 1 Explain the trend

    - Electronegativity differences: HCl > HBr > HI - Number of electrons and London forces: HCl < HBr < HI → not by dipole-dipole forces, but by London forces ... C2H5OC2H5 (diethyl ether) or C4H9OH (butanol); (c) CHI3 or CHF3;...
  • Viper FilmStream Camera - Free

    Viper FilmStream Camera - Free

    Block diagram (simplified, FilmStream mode) From field to post Hard Disk Recording at the set by "director's friend" director's friend FilmStream or RGB-video system diagram YUV-Video system diagram Post-Production Overview, including recorders Viper FilmStream Camera Summary No Slide Title
  • Generation Adequacy Task Force Update to WMS

    Generation Adequacy Task Force Update to WMS

    Generation Adequacy Task Force Update to WMS November 7, 2012 Items Discussed: Status of 2012 Loss of Load Expectation (LOLE) Study Update on NPRR489, Planning Reserve Margin, and PGRR022, Removal of Planning Reserve Margin Calculation Methodology Discussion on next scheduled...
  • Sharing eCrime Data Across National Borders - FIRST

    Sharing eCrime Data Across National Borders - FIRST

    A Look Backwards. Many years ago, we started a block list of phishing 'site' URLs. We ignored spam, malware, etc., sites. But phishing is now smushed with everything else - one lure gets you a phish, malware, and other prizes
  • Year 12, 2016 Curriculum Information Evening Penny Houghton

    Year 12, 2016 Curriculum Information Evening Penny Houghton

    Moderation and Scaling - Applied to ATAR Courses only. Moderation and scaling are both part of the marks adjustment process that is used by the Tertiary Institutions Service Centre (TISC) as students are ranked for University entry