2019 assessment of reliability performance services/2019 texas … · 2019 assessment of...

2019 Assessment of Reliability Performance for the Texas Interconnection

By Texas Reliability Entity, Inc.

May 2020

2019 ASSESSMENT OF RELIABILITY PERFORMANCE PAGE 2 OF 123 MAY 2020

Table of Contents

Executive Summary ................................................................................................................... 6

Summary of Key Findings and Observations ............................................................................. 8

Recommended Focus Areas for 2020 ........................................................................................ 9

Introduction ...............................................................................................................................11

ERCOT Region Bulk Electric System – By the Numbers ...........................................................13

Key Performance Indicators ......................................................................................................16

I. Event Analysis Review .......................................................................................................18

II. Resource Adequacy and Performance ..............................................................................22

III. System Resilience .............................................................................................................41

IV. Changing Resource Mix .....................................................................................................59

V. Human Performance ..........................................................................................................79

VI. Bulk Power System Planning .............................................................................................84

VII. Loss of Situational Awareness ...........................................................................................92

VIII. Increasing Complexity in Protection and Control Systems .................................................99

IX. Physical and Cyber Security ............................................................................................ 104

Appendix A – Transmission Availability Analysis ..................................................................... 106

Appendix B – Generation Performance Analysis ..................................................................... 114

Appendix C – Frequency Control Performance Analysis ......................................................... 119

Table of Figures and Tables

Figure 1 – ERCOT Region Map and 2019 Texas RE Registered Entities ..................................11

Table 1 – Key Performance Indicator Trends ............................................................................17

Figure 2 – Bulk Power System Awareness - By the Numbers ...................................................18

Figure 3 – 2019 Events by Category .........................................................................................19

Figure 4 – Summary of 2015-2019 Event Analysis Trends ........................................................21

Figure 5 – Resource Adequacy and Performance - By the Numbers ........................................24

Figure 6 – Summer 2019 SARA versus Actual at Peak .............................................................24

Figure 7 – August 12, 2019 Capacity, Demand, and Reserves .................................................25

Table 2 – System Conditions, August 12, 13, and 15, 2019 ......................................................25

Figure 8 – Summer 2019 Generation Scheduled and Forced Outages .....................................26

Figure 9 – Typical Frequency Disturbance ................................................................................27

Table 3 – Frequency Event Requirements and Metrics .............................................................27

Figure 10 – Frequency Disturbance Nadir versus Gen Loss MW/Inertia, 2015-2019.................28

Figure 11 – Rate of Change of Frequency versus Normalized Generation Loss, 2015-2019.....29

Figure 12 – Time (sec) between T0 (Start of Frequency Disturbance) to Nadir .........................30

Figure 13 – Annual Primary Frequency Response Trend for ERCOT Region ...........................31

Figure 14 – Event Recovery Time 2012-2019 ...........................................................................32

Table 4 – ERCOT Generation Performance Metrics 2015 through 2019 ...................................33

Figure 15 – MW-Weighted EFOR Metric by Fuel Type and Year ..............................................34

Figure 16 – Time Trend for MW-Weighted EFOR .....................................................................34

Table 5 – Generator Immediate De-rate and Forced Outage Data (Jan. – Dec. 2019) ..............35


Table 6 – 2019 Major Category Cause of Immediate Forced Outage Events from GADS .........35

Figure 17 – 2019 Average Forced Outages per Unit .................................................................36

Figure 18 – 2019 Average Unavailability from Forced Outages per Unit ...................................36

Figure 19 – Reportable Balancing Contingency Events by Year ...............................................37

Figure 20 – Cumulative Unavailable MW Due to Natural Gas Curtailments By Season ............38

Figure 21 – Cumulative Unavailable MW Due to Natural Gas Curtailments By Year .................39

Figure 22 – History of Demand Response Deployed by ERCOT ...............................................40

Figure 23 – Cumulative MW of Economic Demand Response Deployments .............................40

Figure 24 – System Resilience - By the Numbers .....................................................................43

Table 7 – Summary of Event Analyses .....................................................................................43

Figure 25 – Events Reported by Quarter ...................................................................................44

Figure 26 – 2011-2019 Event Cause Summary .........................................................................44

Table 8 – Extreme Transmission Event Day Analyses ..............................................................45

Table 9 – Extreme Generation Event Day Analyses ..................................................................45

Figure 27 – OE-417 Reports of Lost Load .................................................................................46

Figure 28 – Event Attributes by Category ..................................................................................47

Table 10 – 2019 Momentary and Sustained Outages................................................................47

Figure 29 – 2010-2019 345 kV Automatic Outage Metrics ........................................................48

Table 11 – TADS Circuit and Automatic Outage Historical Data for ERCOT Region .................48

Figure 30 – 2019 Automatic Outages by Month ........................................................................49

Figure 31 – 2019 Automatic Outage Duration by Month ............................................................49

Figure 32 – 2019 345 kV Sustained Outage Cause versus Duration .........................................50

Figure 33 – 2019 138 kV Sustained Outage Cause versus Duration .........................................51

Figure 34 – 2015-2019 345 kV Sustained Outages by Event Type ...........................................53

Figure 35 – Interface Operation Minutes Greater Than 90 percent of GTL ................................55

Table 12 – 2019 Top Constraints by Duration ...........................................................................55

Figure 36 – Constraints by Month for 2019 ...............................................................................56

Table 13 – 2019 Chronic Constraint Causes .............................................................................57

Figure 37 – 2019 Chronic Constraint Causes by Total Congestion Rent ...................................57

Figure 38 – 2019 Hourly Reliability Unit Commitments by Month and Cause ............................58

Figure 39 – Changing Resource Mix - By the Numbers .............................................................61

Figure 40 – 2019 Generation Nameplate Capacity ....................................................................62

Figure 41 – 2019 Energy by Fuel Type .....................................................................................63

Figure 42 – Energy by Fuel Type Trend ....................................................................................64

Figure 43 – Planned Generation with Signed Interconnect Agreements and Meeting Planning

Guide Requirements .................................................................................................................64

Figure 44 – Historical System Inertia Type ................................................................................65

Figure 45 – Inertia versus Net Load, 2017-2019 .......................................................................66

Figure 46 – 2019 Inertia by Month and Operating Hour .............................................................67

Figure 47 – Inertia versus Percentage of Load Served by IRRs ................................................68

Table 14 – Minimum Inertia for 2015-2019 ................................................................................68

Figure 48 – Inertia Duration Curve ............................................................................................69


Table 15 – Maximum and Minimum Load, Wind, Solar, and Net-Load Ramps for 2019 ............69

Figure 49 – Maximum One-Hour Ramps for 2013-2019 ............................................................71

Table 16 – Fossil Generation Capacity/Service Factor Metrics .................................................72

Figure 50 – 2015-2019 Fossil Generation Net Capacity Factors ...............................................72

Figure 51 – 2008-2019 Wind Generation as a Percentage of ERCOT Total Energy .................73

Figure 52 – 2008-2019 Wind Generation as Percentage of ERCOT Total Energy by Month .....73

Figure 53 – 2019 Wind Capacity Factor for Summer Peak Hours .............................................74

Table 17 – ERCOT Wind Generation Performance Metrics, 2019 .............................................75

Figure 54 – 2019 Wind GADS Equivalent Availability Factors ...................................................75

Figure 55 – 2019 GADS-Wind Equivalent Outage Rates ..........................................................76

Figure 56 – 2019 GADS-Wind Net Capacity Factors .................................................................76

Figure 57 – 2015-2019 Solar Generation MWH ........................................................................77

Figure 58 – 2015-2019 Solar Generation as a Percentage of ERCOT Total Energy .................78

Figure 59 – 2019 Solar Capacity Factor for Summer Hours ......................................................78

Table 18 – Outages Rates Caused by Human Error .................................................................80

Figure 60 – Outage Rates Caused by Human Error ..................................................................80

Figure 61 – Generator Forced Outage Human Error Issues ......................................................81

Figure 62 – Protection System Misoperations Trend Caused by Human Performance..............82

Figure 63 – Event Analysis Human Performance Cause Coding ...............................................83

Figure 64 – Annual Energy and Peak Demand .........................................................................85

Figure 65 – Energy by Load Zone .............................................................................................86

Figure 66 – Peak Demand by Load Zone ..................................................................................86

Figure 67 – Energy by Weather Zone .......................................................................................87

Figure 68 – Peak Demand by Weather Zone ............................................................................87

Figure 69 – Summer Peak Reserve Margins .............................................................................88

Figure 70 – EEA Events by Year ...............................................................................................89

Figure 71 – Non-Modeled Generation Capacity by Fuel Type ...................................................91

Figure 72 – Loss of EMS and SCADA Events by Year ..............................................................93

Figure 73 – Loss of EMS and SCADA Events by Duration ........................................................94

Figure 74 – Loss of EMS and SCADA Events by Type and Attribute ........................................94

Figure 75 – ERCOT State Estimator Convergence Rate ...........................................................95

Figure 76 – ERCOT Telemetry System Availability ...................................................................96

Figure 77 – State Estimator versus Telemetry Accuracy ...........................................................97

Figure 78 – Bus Summation Telemetry Accuracy ......................................................................97

Figure 79 – Bus Voltage Telemetry Accuracy ...........................................................................98

Table 19 – Protection System Misoperation Data .................................................................... 100

Figure 80 – Protection System Misoperation Trends ............................................................... 100

Figure 81 – Protection System Misoperation Count by Cause 2011-2019 ............................... 101

Figure 82 – Protection System Misoperation Rates by Entity 2015-2019 ................................ 101

Figure 83 – Protection System Misoperation Rates by Region 2015-2019 .............................. 102

Figure 84 – Outage Rates Caused by Failed Protection Equipment ........................................ 103

Figure 85 – ERCOT Trend in Substation Intrusions/Copper Theft/Cyber Security Issues ........ 105


Figure 86 – Locations for Reported Intrusions/Copper Theft/Cyber Security Issues ................ 106

Table A.1 – 2010-2019 End of Year Circuit Data ..................................................................... 106

Table A.2 – 2010-2019 345 kV Circuit and Transformer Outage Data ..................................... 106

Figure A.1 – 345 kV Circuit Automatic Outages by Month ....................................................... 107

Figure A.2 – Multi-Year Comparison of TADS Outages and Duration by Month (> 200 kV) ..... 107

Figure A.3 – 345 kV Circuit Momentary Outage Count by Cause ............................................ 108

Figure A.4 – 345 kV Circuit Sustained Outage Count by Cause .............................................. 108

Figure A.5 – 345 kV Circuit Sustained Outage Duration (Hours) by Cause ............................. 109

Table A.3 – Sustained Outage Data by Average Outage Duration .......................................... 110

Figure A.6 – 138 kV Circuit Sustained Outage Counts by Month ............................................ 111

Figure A.7 – 138 kV Circuit Sustained Outage Duration (Hours) by Month .............................. 112

Figure A.8 – 138 kV Circuit Sustained Outage Count by Cause .............................................. 112

Figure A.9 – 138 kV Circuit Sustained Outage Duration by Cause .......................................... 112

Table B.1 – 2015-2019 GADS and GADS-Wind Units Reporting ............................................ 114

Figure B.1 – GADS Fossil Generation in ERCOT by Age and Fuel Type ................................ 115

Figure B.2 – 2019 GADS Metrics by Age ................................................................................ 115

Figure B.3 – 2015-2019 GADS EFOR by Age ......................................................................... 116

Table B.2 – ERCOT Generation Performance Metrics by Fuel Type for 2019 ......................... 116

Figure B.4 – 2015-2019 Count of Immediate Forced Outage Events by Month ....................... 117

Figure B.5 – 2015-2019 Count of Immediate Forced De-rate Events by Month ....................... 118

Figure C.1 – CPS1 Average January 2010 to December 2019 ............................................... 119

Figure C.2 – ERCOT CPS1 Annual Trend since January 2008 ............................................... 120

Figure C.3 – Frequency Profile Comparison ........................................................................... 121

Figure C.4 – CPS1 Score by Operating Hour for 2015 through 2019 ...................................... 122

Figure C.5 – Daily RMS1 for 2015 through 2019 ..................................................................... 122

Table C.1 – Frequency Trigger Limit Performance .................................................................. 123


Executive Summary

The goals of the 2019 Assessment of Reliability Performance report for the Texas Interconnection are to illuminate the historical and overall bulk power system (BPS) reliability picture, help identify risk areas, and develop measurable priorities for reliability improvement.

This report represents an ongoing effort by Texas Reliability Entity, Inc. (Texas RE) to provide a view of risks to electric reliability based on historic performance in the Electric Reliability Council of Texas (ERCOT) region. By integrating many ongoing efforts and addressing key measurable components of BPS reliability, this report seeks to provide insight, guidance, and direction to those areas in which reliability goals can be more effectively achieved. Additionally, this report seeks to streamline and align the data and information reported from multiple sources, thereby providing efficient data dissemination and information transparency. The key findings and observations can serve as inputs to process improvements, event analyses, reliability assessments, and critical infrastructure protection.

For 2019, overall BPS reliability fell within the defined acceptable performance metrics. The report details key performance observations tied to NERC’s identified risk areas for the BPS:

Resource Adequacy was maintained despite reduced peak reserve margins, a record all-time peak, and challenges resulting in two Level 1 Energy Emergency Alerts (EEAs). An increasing trend in conventional generator outages merits closer monitoring, as does use of demand response.

A Changing Resource Mix still provided adequate inertia and frequency response, and available resources handled ramping requirements and voltage support. A long-term increasing trend in the maximum one-hour up ramp magnitudes for net load and solar generation should be watched closely given continued conventional plant retirements.

Bulk Power System Planning will need to account for observed and ongoing growth in load along with the increased penetration of renewables and distributed energy resources. Adequacy and stability analysis (with necessary modeling enhancements) are an area of emphasis, particularly for inverter-based resources.

System Resilience for the BPS is reflected in limited impacts from numerous storm events that caused loss of transmission and generation facilities, and in improved extreme outage day transmission system performance.

Human Performance as measured in generation and protection system misoperations is improving overall, but causal analysis shows repeated issues with error-checking and procedures.

Loss of Situational Awareness was less of a concern in 2019 given high availability and improving accuracy of telemetry along with high convergence rates for state estimation.

Increasing Complexity in Protection and Control Systems is evident in misoperations and low voltage ride-through failures during system events, but the overall the Protection System Misoperation rate trended downward.

Performance highlights:

Resource Adequacy


o Sufficient operating reserves were maintained during summer and winter peak hours, with a new all-time record set for summer peak. ERCOT issued two EEAs at level 1 in August 2019.

o Demand response played a key role reducing summer peaks, especially during the two EEA level 1 events in August 2019.

o Frequency control was maintained at high levels in 2019. Control Performance Standard 1 (CPS1) was 174.8 for calendar year 2019 versus 175.7 for calendar year 2018. Balancing Authority ACE Limit (BAAL) exceedances were 16 clock-minutes for calendar year 2019.

o GADS Equivalent Forced Outage Rate (EFOR) (MW Weighted) was 7.6 percent for 2019 versus 6.1 percent for 2018. The EFOR rate during the summer of 2019 was 5.6 percent.

o No Reportable Balancing Contingency event recovery failures and no Reportable Balancing Contingency events greater than the Most Severe Single Contingency (MSSC) occurred in 2019. Average recovery time from generation loss events was 5.1 minutes for calendar year 2019 versus 6.2 minutes for calendar year 2018.

o The cumulative generation capacity impacted by natural gas fuel curtailments decreased in 2019 when compared to 2018, likely due to electric-gas regulatory coordination in the summer and a milder winter.

System Resilience o The BPS withstood several events in 2019 that involved the loss of multiple

transmission and generation facilities with minimal impact to the system. o No reported ERCOT Interconnection Reliability Operating Limit (IROL) exceedances

occurred in 2019. o 138 kV circuit outages rates per element remained stable in 2019 when compared to

previous years and the five-year average rates. o The number of transmission and generation outages on extreme days improved in

2019 when compared to extreme days in 2017-2018.

Changing Resource Mix o Average synchronous inertia across most operating hours continued to increase in

2019 when compared to previous years in spite of the retirement of additional large coal units.

o Primary Frequency Response achieved a median value of 918 MW/0.1 Hz, well above the NERC BAL-003 Standard’s obligation of 381 MW/0.1 Hz.

o Peak of net load (the subtraction of renewable generation from system load) became an additional challenge when wind did not reach forecast levels just prior to summer system peaks.

Human Performance o Outage rates for Protection System Misoperations caused by human error are showing

an improving, downward trend. o Generator forced outage rates caused by human error dropped significantly in 2019

compared to previous years, while those for transmission ticked upward - doubling for 138 kV circuits to two percent of the total.

Bulk Power System Planning o Summer Peak: Actual 74,533 MW versus projected 74,853 MW o Winter Peak: Actual 60,646 MW versus projected 61,780 MW o Peak hourly wind generation: 19,580 MW on January 21, 2019

o Peak hourly renewable penetration: 57.5 percent on November 26, 2019


Loss of Situational Awareness o Convergence rates for ERCOT’s State Estimator continued at high levels. o Telemetry availability rates remain stable at approximately 97 percent overall. o Telemetry accuracy metrics showed an improving trend.

Increasing Complexity in Protection and Control Systems o A favorable downward trend in the number of misoperations has continued each year

due to incorrect settings, communication failures, and relay failures. o There was a decrease in the overall Protection System Misoperation rate in 2019, to

6.3 percent for 2019 versus 7.3 percent for 2018.

Summary of Key Findings and Observations

Overall BPS reliability in the ERCOT region continues to perform within acceptable performance levels. The following are key findings:

Resource Adequacy o System resources met adequacy needs despite a peak reserve margin much lower

than the target 13.75 percent. Outage management, higher average winds than forecast and contributions from economic demand response were strong contributors as well as high DC tie imports.

o Gas generator EFOR rates are trending upward as the fuel mix shifts more to predominately gas-fired units.

o The MW-Weighted EFOR for 2019 was higher than the five-year moving average. Long term trends also indicate a gradual increase in EFOR rates.

o Two EEAs on August 13 and 15 were primarily due to slightly higher load conditions with lower wind output on August 13, and significantly lower wind output in addition to higher outages on August 15 (when compared to the summer peak day on August 12).

System Resilience o Protection system misoperations and wind generator ride-through issues continue to

be a main causal factor in system events. o 345 kV circuit outage rates increased in 2019, but remained within the five-year

average. o Customer impact from loss of load events reported on OE-417 reports increased in

2019 due to the multiple major storms affecting primarily distribution customers.

Changing Resource Mix o Over 1660 MW of coal and natural gas capacity was retired or mothballed in 2019. It

was replaced by over 2,700 MW of new capacity, 90 percent of which was renewable. o There is a long-term increasing trend in the maximum one-hour up ramp magnitudes

for net load and solar generation. Solar ramp rates are currently not accounted for in security-constrained economic dispatch.

o Average synchronous inertia across most operating hours continued to increase in 2019 when compared to previous years.

o Primary and secondary frequency response for the Region remains well above the NERC minimums.

Human Performance o Causal analysis of human errors in Protection System Misoperations shows repeated

issues symptomatic of lack of adequate error-checking processes and procedures. Bulk Power System Planning


o Planning reserve margins in the five-year planning horizon continue to show several years below the reference reserve margin level of 13.75 percent.

o Actual West Texas load growth continued to increase at rates of 5-7 percent per year. Transmission entities in the area are implementing upgrades to keep up with this growth rate.

o As of December 2019, ERCOT projections indicate utility-scale solar generation will increase 273 percent to over 8,500 MW and wind generation will increase 41 percent to more than 33,700 MW over the next two years (based on current signed generation interconnect agreements with financial security). During the same period, only 640 MW of new gas units are projected. The growth rate in renewable generation will continue to test ERCOT’s ability to maintain adequate system inertia and ancillary services.

o Distributed energy resources installations are growing in the Region but with limited visibility into performance and control.

Loss of Situational Awareness o A total of four (two Category 1) loss of SCADA or EMS events were reviewed in 2019.

Total duration was approximately six-and-a-half hours. Increasing Complexity in Protection and Control Systems

o Incorrect settings, logic, and design errors remained the largest cause of misoperations, accounting for 32 percent of misoperations in 2019.

o Multiple system events occurred in 2019 where Protection System Misoperations expanded the magnitude of the transmission elements outaged or caused loss of generation or load.

Recommended Focus Areas for 2020

Critical Infrastructure Interdependencies o Situational awareness of natural gas system operation, including pipeline status, gas

compressor station locations and failures, and deliverability issues o Electric-communication system relationships

Resource Adequacy o Continuing impact of generation unit retirements and resource mix changes o Assessing renewable resource impact on system ramping capability o Distributed energy resource effects on demand, ramping, and voltage control o Economic demand response programs triggering and effectiveness

Bulk Power System Planning o Planning reserve margins o Weak grid areas of the interconnection o Modeling of distributed energy resources o Scenario analysis for wide-area variability in renewable output

Changing Resource Mix o Continued inertia monitoring o Increasing use of stability-related generic transmission constraints due to renewable

resources o Accurate dynamic and steady-state modeling of renewable generation, in particular:

Low voltage and low frequency characteristics of inverter-based resources Stability modeling challenges with inverter-based resources

Resilience During Major System Events


o Analysis of weather-related equipment forced outages and recovery

Cyber and Physical Security o Promotion of information sharing with the Electricity Information Sharing and Analysis

Center (E-ISAC) and participation in programs

Human Performance o Development of BPS entity internal programs and workshops aimed at identified

issues


Introduction

Texas RE is the Federal Energy Regulatory Commission (FERC)-approved Regional Entity for the Texas Interconnection, as authorized by the Energy Policy Act of 2005. Through its Delegation Agreement with the North American Electric Reliability Corporation (NERC), Texas RE is authorized in the Texas Interconnection to:

Develop, monitor, assess and enforce compliance with NERC Reliability Standards.

Assess and periodically report on the reliability and adequacy of the BPS.

The Texas Interconnection (also referred to as the ERCOT region), is a separate electric interconnection located entirely within the state of Texas and operates as a single Balancing Authority (BA) and Reliability Coordinator (RC) area. It provides power to more than 25 million Texas customers—representing 90 percent of the state's electric load—and covers approximately 200,000 square miles. The ERCOT BPS connects more than 46,500 miles of transmission lines and 600+ generation units. The ERCOT region is projected to have more than 82,000 MW of expected generation capacity for the 2020 summer peak demand. Installed renewable generation capacity totals more than 23,800 MW of wind and 2,280 MW of solar. Texas Interconnection members include consumers, cooperatives, generators, power marketers, retail electric providers, investor-owned electric utilities (transmission and distribution providers), and municipal-owned electric utilities.

Figure 1 – ERCOT Region Map and 2019 Texas RE Registered Entities

Texas RE collects reliability data from multiple sources in its role as the Regional Entity. Data sources include, but are not limited to, the following:

Transmission Availability Data System (TADS) (NERC Rules of Procedure (ROP) Section 1600)

Generation Availability Data System (GADS) (NERC ROP Section 1600)

Reliability Function Texas RE

Balancing Authority 1

Distribution Provider 33

Distribution Provider-UFLS 8

Generator Operator 147

Generator Owner 173

Planning Authority 1

Reliability Coordinator 1

Resource Planner 1

Transmission Operator 18

Transmission Owner 29

Transmission Planner 26

Transmission Service Provider 1

Unique Entities 239


Demand Response Availability Data System (DADS) (NERC ROP Section 1600)

Misoperation Information Data Analysis System (MIDAS) (NERC ROP Section 1600)

Event Reports (NERC Reliability Standards and NERC Events Analysis Process)

Frequency Control Performance and Primary Frequency Response (NERC Reliability Standards and ERCOT Operating Guides)

Texas RE continually evaluates risks to system reliability within the ERCOT region through long-term and seasonal reliability assessments, event analyses, situational awareness, tracking reliability indicators, real-time performance monitoring, and planning observations. Texas RE developed the 2019 Assessment of Reliability Performance report to provide a high-level overview of the data collected in the region. This report is intended to provide:

2019 data at a high level;

Associated historical data;

Analysis of 2019 and other historical data as an indicator of the current state of the Texas Interconnection; and

Observations that help connect the current state of system reliability to the future.

This report describes Texas RE’s assessment of reliability data and historical trends in the following focus areas:

1. Resource Adequacy and Performance

2. System Resilience and Extreme Natural Events

3. Changing Resource Mix

4. Human Performance

5. Bulk Power System Planning

6. Loss of Situational Awareness

7. Increasing Complexity in Protection and Control Systems

8. Physical and Cyber Security

Each section provides a brief description of the data that is collected and the reliability area being addressed, historical trends, analysis and observations of the historical data, and conclusions.


ERCOT Region Bulk Electric System – By the Numbers


50,000

55,000

60,000

65,000

70,000

75,000

80,000

250,000

275,000

300,000

325,000

350,000

375,000

400,000

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Annual Energy Peak DemandGWH

Summer all-time peak demand:

74,533 MW on 8/12/2019

Record energy usage for 2019

of over 376,800 GWH

All-time peak hourly wind generation:

19,580 MW on 1/21/2019

Peak hourly renewable

penetration: 57.5% on

11/26/2019

Renewable energy served 21.1% of total

energy for 2019

Natural Gas 181,770

48%

Coal77,857

20%

Nuclear41,314

11%

Wind76,708

20%

Water9560%

Solar4,398

1%

Other1,056

0%

2019 Energy (GWH) by Fuel Type

Demand and Energy

0%

5%

10%

15%

20%

25%

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Renewable Energy % of Total Energy

Wind Solar


Control Performance

Standard 1 (CPS1): 174.8

for 2019 versus 175.7 for 2018

Primary Frequency Response: 918 MW/0.1 Hz for

2019 versus NERC obligation of 381

MW/0.1 Hz

TADS 345 kV circuit automatic outage rate per

100 miles: 2.97 for 2019 versus 1.98

for 2018

Protection system misoperation rate:

6.3% for 2019 versus 7.3% for

2018

GADS EFOR (MW Weighted): 7.6% for 2019 versus 6.1% for 2018

Reliability

0

200

400

600

800

1000

1200

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

NERC Primary Frequency Response Per BAL-003 (MW per 0.1 Hz)

0

0.5

1

1.5

2

2.5

3

3.5

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

345kV Ckts Outages / circuit 345kV Ckts Outages/100 mi-year

138kV Ckts Sustained Outages / circuit 138kV Ckts Sustained Outages/100 mi-year

345kV Xmfr Outages / Element 5 per. Mov. Avg. (345kV Ckts Outages / circuit)

5 per. Mov. Avg. (345kV Ckts Outages/100 mi-year)

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

2014 2015 2016 2017 2018 2019

GADS MW Weighted EFOR


Key Performance Indicators

Texas RE utilizes key performance indicators to evaluate how effectively the region is meeting important electric reliability objectives. The table below describes these indicators, how they are measured, the target values, and an assessment of the current state of each.

Key Performance Area

How is it measured? Why is it important? Target(s) Trend (Based on a statistical five-year linear regression)

Resource Adequacy

Summer planning reserve margin compared to reference target value

Number of events with loss of load due to resource adequacy

Count of Energy Emergency Alerts

Indicator of potential resource adequacy issues

Reserve margin > 13.75%

0 events with load loss

0 EEAs

Improved reserve margin in 2020, but below reference margin level for two of next five years

Transmission 345 kV outage rate per element

IROL exceedances

Indicator of potential transmission performance issues

Annual rate < 1.0. Declining trend in annual rate.

0 IROL exceedances

345 kV outage rates increased in 2019. 0 IROL exceedances

Generation EFOR

Primary frequency response

Balancing Contingency Events > Most Severe Single Contingency

Indicator of potential resource adequacy issue and issues related to changing resource mix

Annual rate < 7.0%. Declining trend in annual rate.

> 381 MW / 0.1 Hz

0 Balancing Contingency events > MSSC

Increasing trend in gas generation EFOR rates. Positive trend in frequency response.

System Inertia

Calculated from unit on-line status, MW rating, and inertia constant

Indicator of potential changing resource mix issues

Measured hourly inertia value above calculated critical inertia value

Stable to positive trend during all hours.

Protection System Misoperations

BES Protection System Misoperations rate

Indicator of protection system issues

Annual rate < 7.0%. Declining trend in annual rate.

Decreased rate in 2019. Five year


overall declining trend.

Human Performance

Transmission outage rate caused by human error

Generator outage rate caused by human error

Misoperation rate caused by human error

Indicator of human performance issues

< 2.0%. Declining trend in annual rate.



Stable trend in transmission and misoperation HP errors. Improved trend in generation HP errors.

Table 1 – Key Performance Indicator Trends


I. Event Analysis Review

Summary

The event analysis program is used to conduct root cause analysis of disruption events occurring on the BPS. The event analysis program begins with Texas RE’s situational awareness program to monitor real-time conditions and potential events. Detailed information is collected for larger and more impactful events. Review and analysis of this information helps identify potential reliability risks and emerging threats, or identify lessons learned that can be shared with industry. The analysis process involves identifying what happened, why it happened, and what can be done to prevent reoccurrence.

Situational Awareness (SA)

Texas RE situational awareness analyzes real-time information, collects data on system disturbances and other incidents impacting the BPS, and disseminates this information to internal departments, NERC, and government agencies as necessary. SA also monitors the Reliability Coordinator Information System (RCIS), ongoing storms, natural disasters, and other events that may potential impact the BPS.

Figure 2 – Bulk Power System Awareness - By the Numbers

31 OE-417

6 EOP-004

55 FNET

PI d

ata

RC

IS

We

ath

er

255 Daily

Reports

14 EA rpts

Maco

mb

er

12 Monthly

Reports

Quarterly

MRC Reports


Event Analysis (EA)

In 2019, Texas RE EA evaluated 84 qualified and non-qualified events. The majority of the non-qualified events consisted of frequency disturbances from large generator trips and weather-related OE-417 outage reports. Eleven events were qualified Category 1 events under the NERC EA process.

Figure 3 – 2019 Events by Category

Event Trends

In 2019, Texas RE analyzed 84 BPS events—on par with the events reported per year during the preceding four years. In total, 433 events were reviewed between 2015 and 2019. Of the 302 root and contributing causes identified, the “Equipment/Material” category occurred most frequently with 36 percent of all identified causes. “Design/Engineering” was second with 15 percent, followed closely by “Management/Organization” with 14 percent. The number of Category 1 events has been stable over the last five years.

73

11

0

0

0

Category

0

Category

1

Category

2

Category

3

Category

4 and 5

7 – 3 or more BES facilities lost (1a)

2 – Unintended loss of generation (1g)

2 – Loss of situational awareness/EMS (1h)

13 – Weather

6 – Physical threat/theft/intrusion

50 – Frequency disturbances/Generator trips

4 – Other


2015-2019 Event Analysis Trends

302 Root and Contributing Causes

Identified

Qualified event

attributes dominated by Generation

Loss, Protection System

Misoperations, and Loss of Situational

Awareness

0

20

40

60

80

100

120

2015 2016 2017 2018 2019

EventsGenerator Trips Cat 4 & 5 Cat 3 Cat 2 Cat 1 Cat 0

Over 270 Event Reports

433 Events Analyzed

A1-Design/Engineering

15%

A2-Equipment/Material

36%

A3-Indiv Human Performance

3%

A4-Management/Organization

14%

A5-Communication3%

A6-Training1%

A7-Other8%

AX-Overall Configuration

4%

AZ-Info to Determine Cause

LTA16%

Identified Root and Contributing Causes

0

10

20

30

40

50

60

70

80

90

Load Loss Generation Loss Misoperations Occurred UFLS/UVLS Operation Emergency Actions/Re-dispatch

SPS Operation Loss of SituationalAwareness

Attributes of Texas RE Events: Jan 2011 - Present

Cat 0 Cat 1 Cat 2 Cat 3 Cat 4-5


Figure 4 – Summary of 2015-2019 Event Analysis Trends


II. Resource Adequacy and Performance

Summary

Resource adequacy and performance encompasses an operational review of generation resources. The resource mix is constantly changing and is being affected by the integration of new technologies with different operating characteristics than traditional synchronous generation.

2019 highlights from the analysis of resource adequacy and performance include:

Sufficient operating reserves were maintained during summer and winter peak hours, with a new all-time record set for summer peak. Conditions were tight, as reflected by the number of capacity advisories, watches and two level 1 EEAs issued in August 2019.

Demand response played a key role during the two EEA level 1 events in August 2019. Primary Frequency Response achieved a median value of 918 MW/0.1 Hz versus the

NERC obligation of 381 MW/0.1 Hz. GADS EFOR (MW Weighted) was 7.6 percent for 2019 versus 6.1 percent for 2018. No Reportable Balancing Contingency event (800 MW loss or greater) recovery failures

and no Reportable Balancing Contingency events greater than the MSSC (1,375 MW) occurred in 2019.

The cumulative generation capacity impacted by natural gas fuel curtailments decreased in 2019 when compared to 2018. This was primarily due to a milder winter with relatively few severe cold weather days and in part due to regulatory requests from state agencies for better coordination of outages between the gas and electric industries.

Areas of concern include:

Gas generator EFOR rates are indicating an increasing trend with the shift to a predominately gas-fired mix.

The MW-Weighted EFOR for 2019 was higher than the five-year moving average. Long term trends also indicate gradually increasing EFOR rates.

Metrics and Data Associated with This Area

1. Analysis of planned versus actual seasonal operating reserves 2. Primary frequency response metrics 3. Generator performance metrics 4. Reportable Balancing Contingency event failures 5. Reportable Balancing Contingency events greater than the MSSC 6. Fuel constraints 7. Demand response performance


Primary frequency

response continues tobe maintained at high levels.

Resource Outages

Demand

Resource Capacity

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

Typical Case ExtremeLoad Case

Actual8/12/2019

Actual8/13/2019

Actual8/15/2019

4,226 4,226 3,756 4,174 4,879

74,85378,156

74,533 74,18170,855

78,929 78,929 80,341 79,66377,075

Summer 2019

Resource Outages Demand Resource Capacity

-4.4%

Reserve Margin

2.6%

Reserve Margin

-0.2% Reserve Margin

Projected Peak: 74,853 MW

Actual Hourly Peak: 74,533 MWWind % at Peak: 11.9%Max Hourly Wind: 19,580 MW

Max Hourly Wind %: 57.5%Advisories (PRC<3,000): 29

Watches (PRC<2,500): 2EEA (PRC<2,300): 2

Gas curtailments decreased in

2019 due to the mild winter but show an increasing trend over time

0%

5%

10%

15%

20%

25%

30%

Total Coal, Lignite,Fluidized Bed

GG - Gas Nuclear Gas Turbine/JetEngine (Simple

Cycle Operation)

Combined CycleBlock

CC GT units CC steam units

2014 2015 2016 2017 2018 2019MW Weighted EFOR

EFOR rates for gas fleet is showing a long-term increasing trend

0

200

400

600

800

1000

1200

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019


0

5,000

10,000

15,000

20,000

25,000

2012 2013 2014 2015 2016 2017 2018 2019

Cumulative MW of Gas Curtailments by Year


Figure 5 – Resource Adequacy and Performance - By the Numbers

Detailed Analysis

A. Analysis of Planned versus Actual Seasonal Operating Reserves

This focuses on the Summer of 2019 as the other seasons did not experience tight capacity conditions. For the Summer of 2019, peak demand was within 300 MW of the typical scenario but lower than the extreme forecast. An all-time record peak load was set on August 12, 2019. Actual reserve margin was approximately 2.6 percent. Sufficient operating reserves were maintained during the summer peak hours.

Figure 6 – Summer 2019 SARA versus Actual at Peak

Resource Outages

Demand

Resource Capacity

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

Typical Case ExtremeLoad Case

Actual8/12/2019

Actual8/13/2019

Actual8/15/2019

4,226 4,226 3,756 4,174 4,879

74,85378,156

74,533 74,18170,855

78,929 78,929 80,341 79,66377,075

Summer 2019

Resource Outages Demand Resource Capacity

-4.4%

Reserve Margin

2.6%

Reserve Margin

-0.2% Reserve Margin


Figure 7 – August 12, 2019 Capacity, Demand, and Reserves

During the week of August 12-16, 2019, ERCOT had unprecedented loads due to the hot weather conditions. A new all-time peak was set on Monday, August 12 . On Tuesday, August 13 (with similar load conditions) ERCOT declared an EEA level 1 when Physical Responsive Capability (PRC) levels dropped below 2,300 MW. On Thursday, August 15, with load levels approximately 4,000 MW less, ERCOT again declared an EEA level 1 when PRC levels dropped below 2,300 MW. A comparison of the system conditions on these days is shown in the table below.

At the time of the lowest PRC 8/12/2019 8/13/2019 8/15/2019

Time 14:45 15:15 15:15

Load 73,804 74371 70858

Wind 5859 4514 2024

Solar 1538 1338 1278

Outages 3765 3282 4916

PRC 2670 2171 2188

Non-renewable capacity 68750 69966 69076

Net Load 66407 68518 67555

Table 2 – System Conditions, August 12, 13, and 15, 2019

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,00012

-Au

g-19

00:

00:0

0

12-A

ug-

19 0

0:30

:00

12-A

ug-

19 0

1:00

:00

12-A

ug-

19 0

1:30

:00

12-A

ug-

19 0

2:00

:00

12-A

ug-

19 0

2:30

:00

12-A

ug-

19 0

3:00

:00

12-A

ug-

19 0

3:30

:00

12-A

ug-

19 0

4:00

:00

12-A

ug-

19 0

4:30

:00

12-A

ug-

19 0

5:00

:00

12-A

ug-

19 0

5:30

:00

12-A

ug-

19 0

6:00

:00

12-A

ug-

19 0

6:30

:00

12-A

ug-

19 0

7:00

:00

12-A

ug-

19 0

7:30

:00

12-A

ug-

19 0

8:00

:00

12-A

ug-

19 0

8:30

:00

12-A

ug-

19 0

9:00

:00

12-A

ug-

19 0

9:30

:00

12-A

ug-

19 1

0:00

:00

12-A

ug-

19 1

0:30

:00

12-A

ug-

19 1

1:00

:00

12-A

ug-

19 1

1:30

:00

12-A

ug-

19 1

2:00

:00

12-A

ug-

19 1

2:30

:00

12-A

ug-

19 1

3:00

:00

12-A

ug-

19 1

3:30

:00

12-A

ug-

19 1

4:00

:00

12-A

ug-

19 1

4:30

:00

12-A

ug-

19 1

5:00

:00

12-A

ug-

19 1

5:30

:00

12-A

ug-

19 1

6:00

:00

12-A

ug-

19 1

6:30

:00

12-A

ug-

19 1

7:00

:00

12-A

ug-

19 1

7:30

:00

12-A

ug-

19 1

8:00

:00

12-A

ug-

19 1

8:30

:00

12-A

ug-

19 1

9:00

:00

12-A

ug-

19 1

9:30

:00

12-A

ug-

19 2

0:00

:00

12-A

ug-

19 2

0:30

:00

12-A

ug-

19 2

1:00

:00

12-A

ug-

19 2

1:30

:00

12-A

ug-

19 2

2:00

:00

12-A

ug-

19 2

2:30

:00

12-A

ug-

19 2

3:00

:00

12-A

ug-

19 2

3:30

:00

Hourly Average Demand, Capacity, and Reserves - 8/12/2019

Non-renewable HSL (1) Wind HSL (2) Quick Start (3) On-line Non-Spin (4)

Off-line Non-Spin (5) Load (6) Net Load (7) PRC (8)

Wind (9) Solar (10) PRC=2300 (11)


The difference between the peak day of August 12 and the EEA days of August 13 and 15 can be focused down to two primary issues. August 13 had slightly higher load conditions but lower wind output. August 15 had significantly lower wind output (with much lower load) in addition to higher outages. For Summer 2019, the average scheduled generation outage was 306 MW with a maximum of 1,495 MW. This is in comparison to the final ERCOT Seasonal Assessment of Resource Adequacy (SARA) for Summer 2019 (released May 2019) that estimated typical maintenance outages of 381 MW. For Summer 2019, the average forced generation outage was 4,432 MW with a maximum of 8,629 MW. The final Summer 2019 SARA estimated typical forced outages of 3,845 MW with an extreme case of 6,510 MW.

Figure 8 – Summer 2019 Generation Scheduled and Forced Outages

B. Primary Frequency Response

Primary frequency response is defined as the immediate proportional increase or decrease in real power output provided by generating units/generating facilities and the natural real power dampening response provided by Load in response to system Frequency Deviations. This response is in the direction that stabilizes frequency. The following figure shows a typical frequency disturbance broken down into several periods.

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

Summer 2019 Generation Scheduled and Forced Outages

Sched Outage MW Forced Outage MW Total Outage MW

Typ Sched Outage MW Typ Forced Outage MW


Figure 9 – Typical Frequency Disturbance

Each of the periods of the frequency disturbance is analyzed by different metrics and performance indicators. Two of the key performance indicators are based on requirements in the BAL-002 and BAL-003 Standards. These are recovery of the Area Control Error (ACE) within 15 minutes following a Reportable Balancing Contingency Event, and maintaining the interconnection frequency response at or above the Interconnection Frequency Response Obligation (IFRO).

Period Time Frame Reliability Requirement Metric(s)

Arrest Period T0 to T+6 seconds

Arrest C-point at or above 59.3 Hz for loss of 2750 MW

(BAL-003)

- RoCoF/MW Loss - T0 to Tc - Nadir Frequency

Margin

Rebound/Stabilizing Period

T+6 to T+60 seconds

Achieve Interconnection frequency response at or above IFRO (381 MW per

0.1 Hz) (BAL-003)

- Primary Frequency Response

Recovery Period T+1 to T+15

minutes Recover ACE within 15

minutes (BAL-002) - Event recovery

time

Table 3 – Frequency Event Requirements and Metrics

Rotating turbine generators and motors synchronously interconnected to the system store kinetic energy during contingency events that is released to the system (also called inertial response). Inertial response provides an important contribution in the initial moments following a generation or load trip event and determines the initial rate of change of frequency (RoCoF). In response to a sudden loss of generation, kinetic energy will automatically be extracted from


the rotating synchronized machines on the interconnection, causing them to slow down and frequency to decline. The amount of inertia depends on the number and size of generators and motors synchronized to the system, and it determines the rate of frequency decline. Greater inertia reduces the rate of change of frequency, giving more time for primary frequency response to fully deploy and arrest frequency decay above under-frequency load shed set points. Therefore, with potential wide variations in inertia conditions with increasing use of inverter-based generation resources, there is a need to monitor and trend inertia and initial rate of change of frequency.

The Nadir, or C-Point frequency, is an indicator of the system imbalance created by the unit trip and is a combination of synchronous inertial response and governor response. Normalizing the unit MW loss by inertia can provide insight into how the Nadir can vary under different inertia conditions for the same MW loss value. The figure below shows the Nadir plotted against the generation MW loss value normalized for inertia and shows the inverse relationship for how historic performance for how the Nadir was affected by different MW loss and inertia conditions.

Figure 10 – Frequency Disturbance Nadir versus Gen Loss MW/Inertia, 2015-2019

y = 0.0021x2 - 0.0498x + 59.982R² = 0.7228

59.65

59.7

59.75

59.8

59.85

59.9

59.95

60

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00

C-Point Frequency vs Gen Loss MW/Inertia, 2015-2019C-Point Frequency

Gen Loss (MW)/Inertia (GW-sec)


The RoCoF during the initial frequency decline in the first 0.5 sec is largely driven by system inertia, therefore it is prudent to use the same analysis technique to plot the RoCoF against the generation MW loss normalized by system inertia. The figure below shows this relationship, with a straight line approximation.

Figure 11 – Rate of Change of Frequency versus Normalized Generation Loss, 2015-

2019

y = 16.875x + 9.9314R² = 0.7514

0

20

40

60

80

100

120

140

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00

RoCoF (mHz/sec) vs Gen Loss MW/Inertia, 2015-2019

Gen Loss MW/Inertia

RoCoF


Similar to the RoCoF, the time between the start of the frequency disturbance (T0) and the Nadir point (Tc) is also an indicator of system inertia. The following figure shows this metric for 2015-2019. Average value for this time continues to be in range of 6 seconds.

Figure 12 – Time (sec) between T0 (Start of Frequency Disturbance) to Nadir

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

2015 2016 2017 2018 2019

To-Tc


The following figure shows the trend in primary frequency response for the ERCOT region. In 2019, the average frequency response was 912 MW per 0.1 Hz and the median frequency response was 918 MW per 0.1 Hz as calculated per NERC Standard BAL-003 for the events that were evaluated during the period. The following graph shows the annualized primary frequency response trend per NERC Reliability Standard BAL-003. The green lines on the figure indicate Interconnection Frequency Response Obligation (IFRO) as calculated according to NERC Standard BAL-003.

Figure 13 – Annual Primary Frequency Response Trend for ERCOT Region

0

200

400

600

800

1000

1200

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019



C. Secondary Frequency Response

NERC Reliability Standards require a maximum Area Control Error (ACE) recovery time of 15 minutes for reportable disturbances. Average recovery time from generation loss events was 5.1 minutes in 2019 versus 6.2 minutes for calendar year 2018. The average event recovery time (see Figure 14) continues to show a long-term gradual upward trend since 2012.

Figure 14 – Event Recovery Time 2012-2019

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

2012 2013 2014 2015 2016 2017 2018 2019

Event Recovery Time (Minutes) - Reportable Balancing Contingency Events


D. 2019 Fossil-fueled Generator Performance Metrics

GADS provides various metrics to compare unit performance. Two of these methods are unweighted (time-based) and weighted (based on unit MW size). A summary of key unweighted performance metrics for the ERCOT generation fleet for 2019 is provided in the following table.

ERCOT Region GADS Data Metric

2015 2016 2017 2018 2019 5-Yr Avg

Unweighted Unweighted Unweighted Unweighted Unweighted Unweighted

# Units Reporting 413 409 415 407 402 409

Total Unit-Months 4899 4908 4860 4768 4803 4848

Net Capacity Factor (NCF)

45.5% 44.2% 43.3% 46.7% 46.8% 45.3%

Service Factor (SF) 49.8% 48.4% 46.1% 50.9% 51.7% 49.4%

Equivalent Availability Factor (EAF)

85.8% 87.0% 85.3% 85.2% 86.1% 85.9%

Scheduled Outage Factor (SOF)

9.1% 8.2% 8.3% 8.7% 9.3% 8.7%

Forced Outage Factor (FOF)

3.1% 2.9% 4.1% 3.9% 4.2% 3.6%

EFOR 7.0% 6.7% 9.6% 7.9% 8.3% 7.9%

Equivalent Forced Outage Rate Demand (EFORd)

5.5% 5.0% 6.6% 5.7% 6.1% 5.8%

Table 4 – ERCOT Generation Performance Metrics 2015 through 2019

Net Capacity Factor: Percent of maximum net energy produced for the period

Service Factor: Percent of time on line

Equivalent Availability Factor: Percent of time available without outages or derates

Scheduled Outage Factor: Percent of time on scheduled outage or derate

Forced Outage Factor: Percent of time on forced outage or derate

Equivalent Forced Outage Rate: Probability of being on a forced outage or derate

Equivalent Forced Outage Rate Demand: Probability that units will not meet generating requirements for demand periods due to forced outages or derates.

Figures 15 and 16 summarize MW weighted EFOR by fuel type over several years and monthly and yearly for aggregate units in ERCOT. The MW-Weighted EFOR for 2019 was higher than the five-year moving average. Long term trends also indicate gradually increasing EFOR rates.


Figure 15 – MW-Weighted EFOR Metric by Fuel Type and Year

Figure 16 – Time Trend for MW-Weighted EFOR

0%

5%

10%

15%

20%

25%

30%

Total Coal, Lignite,Fluidized Bed

GG - Gas Nuclear Gas Turbine/JetEngine (Simple

Cycle Operation)

Combined CycleBlock


2014 2015 2016 2017 2018 2019MW Weighted EFOR

0%

2%

4%

6%

8%

10%

12%

Jan

-15

Fe

b-1

5

Ma

r-1

5

Ap

r-1

5

Ma

y-1

5

Jun

-15

Jul-

15

Au

g-1

5

Se

p-1

5

Oct-

15

No

v-1

5

De

c-1

5

Jan

-16

Fe

b-1

6

Ma

r-1

6

Ap

r-1

6

Ma

y-1

6

Jun

-16

Jul-

16

Au

g-1

6

Se

p-1

6

Oct-

16

No

v-1

6

De

c-1

6

Jan

-17

Fe

b-1

7

Ma

r-1

7

Ap

r-1

7

Ma

y-1

7

Jun

-17

Jul-

17

Au

g-1

7

Se

p-1

7

Oct-

17

No

v-1

7

De

c-1

7

Jan

-18

Fe

b-1

8

Ma

r-1

8

Ap

r-1

8

Ma

y-1

8

Jun

-18

Jul-

18

Au

g-1

8

Se

p-1

8

Oct-

18

No

v-1

8

De

c-1

8

Jan

-19

Fe

b-1

9

Ma

r-1

9

Ap

r-1

9

Ma

y-1

9

Jun

-19

Jul-

19

Au

g-1

9

Se

p-1

9

Oct-

19

No

v-1

9

De

c-1

9

MW Weighted EFOR

Monthly Weighted EFOR Annual Weighted EFOR 5-year Average Weighted EFOR


2019 Fossil-fueled Generator Outages and De-rates

Table 5 provides a summary of immediate de-rates and forced outages for conventional generation from January 2019 through December 2019. The 2,122 immediate forced outage events is about five percent higher than 2018, with a median capacity of 171 MW per event nearly identical to last year’s, as were the top three systems affected.

The majority of the immediate de-rate events occurred due to low BTU or wet coal, pulverizer feeder and mill issues, and induced draft fan issues. Reference the following tables and graphics.

2019 Immediate De-Rates Immediate Forced Outages

Number of Events 2,272 2,122

Total Duration (hrs) 156,553.2 122,471.5

Total Capacity 230,272.7 405,724.6

Avg Duration per Event (hrs) 68.9 57.2

Median Duration per Event (hrs) 4.2 4.7

Avg Capacity per Event 101.4 191.2

Median Capacity per Event 52.0 171.0

Table 5 – Generator Immediate De-rate and Forced Outage Data (Jan. – Dec. 2019)

The cause of the immediate forced outage events can also be further broken down into major categories based on the GADS data.

Major System

Number of Forced Outage Events

Total Duration (hours)

Total Capacity

(MW)

Median Duration per Event (hours)

Avg Duration per Event (hours)

Median Capacity per Event

(MW)

Avg Capacity per Event

(MW)

Boiler System 209 9,444.4 70,313.6 8.6 45.2 245.0 336.4

Balance of Plant 457 21,419.3 93,630.5 5.8 46.9 175.0 204.9

Steam Turbine/Generator 1161 67,588.7 191,988.1 4.1 58.2 170.0 165.4

Heat Recovery Steam Generator 74 3,668.3 13,840.9 23.0 49.6 184.0 187.0

Pollution Control Equipment 39 428.4 5,068.3 5.7 11.0 88.0 130.0

External 107 16,995.8 15,990.6 9.0 158.8 102.7 149.4

Regulatory, Safety, Environmental 15 299.0 1,403.4 2.4 19.9 50.0 93.6

Personnel/ Procedure Errors 54 1,965.4 13,295.1 1.5 36.4 168.0 246.2

Other 6 662 194 3.7 110.4 36.0 32.3

Table 6 – 2019 Major Category Cause of Immediate Forced Outage Events from GADS


Figure 17 – 2019 Average Forced Outages per Unit

Figure 18 – 2019 Average Unavailability from Forced Outages per Unit

0

1

2

3

4

5

6

7

8

9

10

Coal, Lignite,Fluidized Bed

Combined CycleBlock


Average Forced Outages per Unit

2012 2013 2014 2015 2016 2017 2018 2019

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

Coal, Lignite,Fluidized Bed

Combined CycleBlock


Average Unavailability from Forced Outages

2012 2013 2014 2015 2016 2017 2018 2019


E. Renewable Generator Performance Metrics

The analysis of performance metrics for renewable generation is contained in Section IV, Changing Resource Mix.

F. Balancing Contingency Event Performance

Texas RE tracks the number of Balancing Contingency events and recovery time within the region to provide any potential adverse reliability indications. Per the NERC BAL-002-2 Disturbance Control Standard, a Reportable Disturbance is defined as any event which causes a change in area control error greater than or equal to 800 MW. Note that the BAL-002 definition for a Reportable Balancing Contingency Event changed from 1,100 MW to 800 MW for ERCOT in January 2018 when BAL-002-2 went into effect.

As part of the Event Analysis process, Texas RE investigates the cause and relative effect on reliability of Balancing Contingency events within the region. Balancing Contingency events greater than the MSSC (1,375 MW) typically do not create a significant reliability problem for the ERCOT region since ERCOT carries contingency reserves greater than the MSSC; however, these events warrant special consideration for review of system frequency response and recovery.

Figure 19 – Reportable Balancing Contingency Events by Year

0

1

2

3

4

5

6

7

8

9

10

11

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Reportable Balancing Contingency Events Balancing Events > MSSCBalancing Contingency Events


G. Fuel Constraints

There was a significant decrease in the unavailable generation capacity due to natural gas fuel curtailments in 2019 (compared to 2018) due to the much milder winter. There was also a decrease in gas curtailments in the summer, most likely due to regulatory efforts to minimize gas interruptions to generation.

Figure 20 – Cumulative Unavailable MW Due to Natural Gas Curtailments By Season

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

Winter Spring Summer Fall

Sum of Unavailable MW due to Gas Curtailments

2012 2013 2014 2015 2016 2017 2018 2019


Figure 21 – Cumulative Unavailable MW Due to Natural Gas Curtailments By Year

H. Demand Response

Three types of demand response are employed in the ERCOT region.

1. Load Resources (LR) providing Responsive Reserve Service (RRS) that are automatically interrupted by underfrequency relays when system frequency decreases to 59.7 Hz or below. These resources can also be manually deployed within 10 minutes by ERCOT in response to energy emergencies.

2. Emergency Response Service (ERS) is a service designed to be deployed by ERCOT as an operational tool under an EEA. ERS is designed to decrease the likelihood of ERCOT operating reserve depletion and the need for ERCOT to direct firm Load shedding. Two types of ERS are procured, ERS-10 (ERS with a 10 minute ramp period) and ERS-30 (ERS with a 30 minute ramp period).

3. Economic demand response that is employed by non-opt-in entities (NOIEs), such as municipalities, for economic purposes in the form of commercial-industrial programs, smart thermostat programs, peak shaving programs, etc.

4. Two of these Demand response programs (ERS and Economic) played a key role during the two EEAs in August 2019. ERCOT deployed over 1,700 MW of ERS on August 13 and August 15, 2019, in response to generation scarcity conditions and high energy prices.

0

5,000

10,000

15,000

20,000

25,000

2012 2013 2014 2015 2016 2017 2018 2019

Cumulative MW of Gas Curtailments by Year


Figure 22 – History of Demand Response Deployed by ERCOT

Figure 23 – Cumulative MW of Economic Demand Response Deployments

0

1

2

3

4

5

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

2011

Q2

2011

Q3

2011

Q4

2012

Q1

2012

Q2

2012

Q3

2012

Q4

2013

Q1

2013

Q2

2013

Q3

2013

Q4

2014

Q1

2014

Q2

2014

Q3

2014

Q4

2015

Q1

2015

Q2

2015

Q3

2015

Q4

2016

Q1

2016

Q2

2016

Q3

2016

Q4

2017

Q1

2017

Q2

2017

Q3

2017

Q4

2018

Q1

2018

Q2

2018

Q3

2018

Q4

2019

Q1

2019

Q2

2019

Q3

2019

Q4

DR MW Deployed - ERCOT # DR Deployment EventsMW # Events

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

Q1 Q2 Q3 Q4

2012 2013 2014 2015 2016 2017 2018 2019

MW Economic Demand Response Deployments


III. System Resilience

Summary

The BPS is becoming more dependent on other industrial sectors such as telecommunications for visibility and control. Coordination between sectors should be enhanced to mitigate vulnerabilities that significantly impact the reliability and resilience of the BPS; therefore system resilience not only includes the electric infrastructure, but also fuel sources and fuel delivery infrastructure, data and voice communications systems, water supplies, and potentially other critical industries. Resilience can be defined as, “[t]he ability to withstand and reduce the magnitude and/or duration of disruptive events, which includes the capability to anticipate, absorb, adapt to, and/or rapidly recover from such an event.” 2019 highlights from the analysis of resilience include:

The BPS was able to withstand several events in 2019 that involved the loss of multiple transmission and generation facilities.

There were no reported Interconnection Reliability Operating Limit (IROL) exceedances in 2019.

138 kV circuit outages rates per element remained stable in 2019 when compared to previous years and the five-year average rates.

Transmission system performance on extreme days improved in 2019 when compared to extreme days in 2017-2018.

For the 345 kV circuit outages in 2019, 20 percent of the sustained automatic outage events and 70 percent of the sustained outage duration involved failed circuit equipment.

For the 138 kV circuit outages in 2019, failed substation equipment and failed transmission circuit equipment dominated the sustained outages, accounting for 30 percent of the outage events and 75 percent of the outage duration.


Protection system misoperations and wind generator ride-through issues continue to be a main causal factor in system events.

345 kV circuit outage rates increased in 2019, but remained within the five-year average.

Customer impact from loss of load events reported on OE-417 reports increased in 2019 due to the multiple major storms affecting primarily distribution customers.


1. Event analyses 2. Transmission circuit outage rates 3. Multiple element outages 4. Outages initiated by failed substation equipment 5. Outages initiated by failed circuit equipment 6. Loss of load events 7. System Operating Limit (SOL) and Interconnection Reliability Operating Limit (IROL)

Performance


0

0.5

1

1.5

2

2.5

3

3.5

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

345kV Ckts Outages / circuit 345kV Ckts Outages/100 mi-year138kV Ckts Sustained Outages / circuit 138kV Ckts Sustained Outages/100 mi-year345kV Xmfr Outages / Element 5 per. Mov. Avg. (345kV Ckts Outages / circuit)5 per. Mov. Avg. (345kV Ckts Outages/100 mi-year)

Major storms caused an increase in distribution customer outages in 2019.

No category 2 or

higher events in 2019.

Equipment failure and weather continue to drive major events

487 momentary and sustained forced outages w ere reported on 16,382 circuit miles of 345 kV transmission.

374 sustained forced outages w ere reported on 23,224 circuit miles of 138 kV transmission.

36 sustained forced outages w ere reported on 223 345 kV transformers

0

2

4

6

8

10

12

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Total Customer Impact # Reported Events

24%

32%

4%

14%

16%

8%

2%

2011-2019 Event Cause Equipment Failure

Weather

Human Error/Other

EMS/Cyber

Sabotage/Vandalism

Relaying Issues

Natural Disaster/ForeignInterference

0

20

40

60

80

100

120

2015 2016 2017 2018 2019

EventsGenerator Trips Cat 4 & 5 Cat 3 Cat 2 Cat 1 Cat 0


Figure 24 – System Resilience - By the Numbers

Detailed Analysis

A. Event analyses

(1) Multiple wind unit loss event on January 22, 2019: Two separate 138 kV transmission line faults resulted in the loss of 636 MW and 382 MW of wind generation respectively. Five of the seven wind units affected had non-consequential MW losses, meaning the units affected were not directly connected to the faulted transmission circuit.

(2) Generation loss event on April 11, 2019: A single line-to-ground fault occurred on a 345 kV transmission line. An adjacent 345 kV transmission line also tripped due to misoperation of its protection system. The combined loss of both 345 kV circuits caused the loss of 664 MW of generation.

(3) Multiple transmission element loss event on May 9, 2019: A lightning strike in the proximity of a 345 kV substation caused the remote terminal unit (RTU) to open four 345 kV circuit breakers and 14 138 kV circuit breakers, due to DC transients damaging the RTU. Two 345 kV transmission lines, eight 138 kV transmission lines, and two 345/138 kV auto transformers were interrupted.

(4) Multiple wind unit loss event on May 20, 2019: A 138 kV bus fault resulted in the loss of 444 MW of wind generation. All of the wind unit MW loss was non-consequential.

(5) Generation loss event on August 10, 2019: A protection relay misoperation resulted in the loss of 1,244 MW of generation.

(6) Generation loss event on October 6, 2019: A protection relay misoperation resulted in delayed fault clearing, causing the loss of 1,100 MW of generation.

Historical Disturbance Data: In 2019, the number of events reported decreased slightly when compared to average number of events between 2015 through 2018.

Event

Category1

2015 2016 2017 2018 2019 5-Yr Avg

Non-Qualified 90 65 52 78 73 72

1 9 5 11 13 11 10

2 1 0 0 0 0 0

3 1 2 0 0 0 1

4 and 5 0 0 1 0 0 0

Total 101 72 64 91 84 83

Table 7 – Summary of Event Analyses

1 Link to NERC Events Analysis Process with category definitions:

https://www.nerc.com/pa/rrm/ea/ERO_EAP_Documents%20DL/ERO_EAP_v4.0_final.pdf


Figure 25 – Events Reported by Quarter

Figure 26 – 2011-2019 Event Cause Summary

Extreme Event Periods

For transmission, “extreme days” are based on the most impactful days as determined by the number of transmission line and transformer outages as well as duration of outages. For generation, “extreme days” are based on the most impactful days as determined by the number of generation immediate forced outages, de-rates, as well as the cumulative MW

2

14

109

4

8 8

5

9

2

4

8

14

65

45

14

98

0

5

10

15

20

25

30

0

5

10

15

20

25

30

2015 1stQtr

20152nd Qtr

2015 3rdQtr

2015 4thQtr

2016 1stQtr

20162nd Qtr

2016 3rdQtr

2016 4thQtr

2017 1stQtr

20172nd Qtr

2017 3rdQtr

2017 4thQtr

2018 1stQtr

20182nd Qtr

2018 3rdQtr

2018 4thQtr

2019 1stQtr

20192nd Qtr

2019 3rdQtr

2019 4thQtr

EventsTotal Events Cat 4 & 5 Cat 3 Cat 2 Cat 1 Cat 0 Generator Trips >450MW

24%

32%4%

14%

16%

8%

2%

2011-2019 Event CauseEquipment Failure

Weather

Human Error/Other

EMS/Cyber

Sabotage/Vandalism

Relaying Issues

NaturalDisaster/ForeignInterference


impact of the outages. The following tables shows a comparison of the extreme transmission event days and extreme generation event days for 2017-2019.

Date Number of Sustained

Transmission Outage Events on Extreme Day

Leading Causes

for Extreme

Day

Average Sustained

Outage Duration on

Extreme Day

Longest Sustained Outage on

Extreme Day

Average Sustained

Outage Duration for Year

Longest Sustained

Outage Duration for

Year

8/26/2017 40 Weather 80 hours 257 hours 54 hours 7,594 hours

1/16/2018 50 Weather 10 Hours 72 hours 53 hours 6,403 hours

5/18/2019 19 Weather 85 hours 332 hours 31 hours 1,657 hours

Table 8 – Extreme Transmission Event Day Analyses

Date Number of Generation

Outage Events on Extreme Day

Leading Causes for Extreme

Day

Cumulative Outage

Duration on Extreme Day

Cumulative MW Impact on Extreme

Day

Cumulative GWH Impact on Extreme

Day

8/27/2017 41 Weather 22,798 hours 10,107 MW 2,917.5 GWH

1/16/2018 84 Balance of Plant/Fuel

2,891 hours 11,893 MW 517.8 GWH

5/11/2019 36 Turbine Generator

1,626 hours 6,449 MW 282.5 GWH

Table 9 – Extreme Generation Event Day Analyses

Registered entities are required to report loss of load to 50,000 customers or more for one hour or more to the Department of Energy using OE-417 reports. 2019 showed a sharp increase in the customer impact from these events due to a number of major thunderstorm events that primarily affected distribution customers. The trend in these reports is included in the following figure.


Figure 27 – OE-417 Reports of Lost Load

0

2

4

6

8

10

12

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Total Customer Impact # Reported Events

OE-417 Reportable Disturbances


Event severity is determined by a number of key attributes, including load lost, MW of generation lost, protection system misoperations, emergency actions taken, etc. The following figure shows the breakdown of event attributes by event category.

Figure 28 – Event Attributes by Category

B. Transmission circuit outage rates

Tables 10 and 11 along with Figures 29-33 provide a summary of reported BPS transmission related outages. Compared to 2018 data, 2019 automatic outage rates per 100 miles of line per year for the 345 kV system increased by roughly 50 percent after four years of declining values. 138kV system sustained automatic outages decreased somewhat and were below the five-year average. The total outage duration from automatic outages for the 345 kV system also increased from 3,447 hours to 3,875 hours.

Momentary Outages Sustained Outages

Voltage Range Per Circuit Per 100 Miles Per Circuit Per 100 Miles

300-399 kV 0.79 2.30 0.23 0.67

100-199 kV Not reportable Not reportable 0.19 1.58

Table 10 – 2019 Momentary and Sustained Outages

Long-term trends are indicating an increasing trend in outage rates per circuit and per 100 miles of line for the 345 kV system. See the following figure and table. Note that the dashed lines represent a five-year rolling average for the respective 345 kV outage metrics.

0

10

20

30

40

50

60

70

80

90

Load Loss Generation Loss Misoperations Occurred UFLS/UVLS Operation Emergency Actions/Re-dispatch

SPS Operation Loss of SituationalAwareness

Attributes of Texas RE Events: Jan 2011 - Present

Cat 0 Cat 1 Cat 2 Cat 3 Cat 4-5


Figure 29 – 2010-2019 345 kV Automatic Outage Metrics

Voltage Class Name Metric 2015 2016 2017 2018 2019 5-Yr Avg

AC Circuit 300-399 kV Automatic Outages per Circuit

1.08 0.99 0.95 0.66 1.02 0.94

AC Circuit 300-399 kV Automatic Outages per 100 miles

3.00 2.78 2.68 1.98 2.97 2.68

AC Circuit 100-199 kV Sustained Automatic Outages per Circuit

0.25 0.19 0.22 0.22 0.19 0.21

AC Circuit 100-199 kV Sustained Automatic Outages per 100 miles

2.11 1.55 1.90 1.87 1.65 1.82

Transformer 300-399 kV

Automatic Outages per Element

0.18 0.11 0.14 0.13 0.16 0.14

Table 11 – TADS Circuit and Automatic Outage Historical Data for ERCOT Region

Automatic Outage Data

For 345 kV transmission circuits, predominant causes for momentary outages in 2019 were lightning, foreign interference, and unknown, representing 70 percent of the total momentary outages. Predominant causes for sustained outages in 2019 were lightning, failed circuit equipment, and weather, representing 62 percent of the total sustained outages. Failed transmission circuit equipment dominated the sustained outage duration, accounting for 70 percent of the outage duration.

0

0.5

1

1.5

2

2.5

3

3.5

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

345kV Ckts Outages / circuit 345kV Ckts Outages/100 mi-year

138kV Ckts Sustained Outages / circuit 138kV Ckts Sustained Outages/100 mi-year

345kV Xmfr Outages / Element 5 per. Mov. Avg. (345kV Ckts Outages / circuit)

5 per. Mov. Avg. (345kV Ckts Outages/100 mi-year)


For 138 kV transmission circuits, predominant causes for sustained outages in 2019 were weather, lightning, and failed substation/circuit equipment, representing 65 percent of the total sustained outages. Failed substation/transmission circuit equipment dominated the sustained outage duration, accounting for 75 percent of the outage duration.

Figure 30 – 2019 Automatic Outages by Month

Figure 31 – 2019 Automatic Outage Duration by Month

0

10

20

30

40

50

60

70

80

90

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

138kV Sustained Outages 345kV Sustained Outages

2019 Automatic Outages By Month

0

500

1000

1500

2000

2500


138kV Outage Duration (hrs) 345kV Outage Duration (hrs)

2019 Automatic Outage Duration By Month


Figure 32 – 2019 345 kV Sustained Outage Cause versus Duration

23%

19%

0%4%4%1%8%

0%1%

20%

0%

7%

13%

2019 345 kV Sustained Outage CauseWeather, ExcludingLightningLightning

Environmental

Contamination

Foreign Interference

Fire

Failed AC SubstationEquipmentFailed AC/DC TerminalEquipmentFailed Protection SystemEquipmentFailed AC Circuit Equipment

Power System Condition

Human Error

Unknown

1%

1…0% 3%

0%0%

16%

2%

1%

70%

0%0% 3%

3%

2019 345 kV Sustained Outage DurationWeather, ExcludingLightningLightning

Environmental

Contamination


Fire

Failed AC SubstationEquipmentFailed AC/DC TerminalEquipmentFailed ProtectionSystem EquipmentFailed AC CircuitEquipmentPower SystemConditionHuman Error

Unknown


Figure 33 – 2019 138 kV Sustained Outage Cause versus Duration

15%

20%

0%

0%

8%

0%

1%12%

0%

3%

18%

1%1%

6%

11%

3%

2019 138 kV Sustained Outage Cause Weather, excludinglightningLightning

Environmental

Contamination


Vandalism, Terrorism, orMalicious ActsFire

Failed AC SubstationEquipmentFailed AC/DC TerminalEquipmentFailed Protection SystemEquipmentFailed AC Circuit Equipment

Vegetation


Human Error

2% 0%

0%

0%

13%

0%

0%

23%

0%

1%

52%

1% 1%1%

5%2%

2019 138 kV Sustained Outage DurationWeather, excludinglightningLightning

Environmental

Contamination


Vandalism, Terrorism, orMalicious ActsFire

Failed AC SubstationEquipmentFailed AC/DC TerminalEquipmentFailed Protection SystemEquipmentFailed AC CircuitEquipmentVegetation


Human Error


C. Multiple Element Outages

For 345 kV circuits in 2019, 58 of the 487 reported automatic outage events involved two or more circuit elements. Dependent Mode outages (defined as an automatic outage of an element that occurred as a result of another outage) and Common Mode outages (defined as two or more automatic outages with the same initiating cause and occurring nearly simultaneously) represented 12 percent of all outages and 67 percent of sustained outage duration for the 345 kV system.

For 138 kV circuits in 2019, 144 of the 374 reported automatic sustained outage events involved two or more circuit elements. Dependent Mode and Common Mode outages represented 39 percent of all sustained outages and 45 percent of sustained outage duration.

Over the five year period from 2015-2019, multiple element outages represented 31 percent of sustained outages and 58 percent of the sustained outage duration for the 345 kV system.


Figure 34 – 2015-2019 345 kV Sustained Outages by Event Type

3%

2%

69%

5%

6%

5%2%

1% 6% 1%

5- Single Bus fault

6 - Internal Bkr fault

11 - Single Elem outage

13-Two or more Elem outage

31-Two or more Elem onCommon Str

49-Outages not covered by 5-31

60-Breaker failure

61-Protection Sys failure

62-Protection Sysmisoperation

90-Outages not covered by60-62

2015-2019 345 kV Sustained Automatic Outages by Event Type

2%0%

42%

7%

34%

8%

6% 0%1%0%

5- Single Bus fault

6 - Internal Bkr fault

11 - Single Elem outage

13-Two or more Elem outage

31-Two or more Elem onCommon Str


60-Breaker failure

61-Protection Sys failure

62-Protection Sysmisoperation


2015-2019 345 kV Automatic Outage Duration by Event Type


D. System Operating Limit (SOL) and Interconnection Reliability Operating Limit (IROL) Performance

A System Operating Limit is the value (such as MW, MVar, amperes, frequency, or voltage) that satisfies the most limiting of the prescribed operating criteria for a specified system configuration to ensure operation within acceptable reliability criteria. SOLs are based upon certain operating criteria. These include, but are not limited to:

Facility ratings (applicable pre- and post-contingency equipment or facility ratings)

Transient stability ratings (applicable pre- and post-contingency stability limits)

Voltage stability ratings (applicable pre- and post-contingency voltage stability)

System voltage limits (applicable pre- and post-contingency voltage limits)

An IROL is an SOL that, if violated, could lead to instability, uncontrolled separation, or cascading outages. There is currently one defined IROL in the ERCOT region, the North-Houston stability limit.

In the past five years, there were no reported exceedances of the North-Houston stability limit.

ERCOT uses Constraint Management Plans (CMPs) as a set of pre-defined actions executed in response to system conditions to prevent or resolve one or more thermal or non-thermal transmission security violations that may be SOLs. CMPs include, but are not limited to the following:

Re-dispatch of generation from Security-Constrained Economic Dispatch (SCED)

Remedial Action Plans (RAPs)

Pre-Contingency Action Plans (PCAPs)

Temporary Outage Action Plans (TOAPs)

Mitigation Plans (MPs)

When developing CMPs, ERCOT typically uses the 15-Minute Rating of the impacted transmission facility(ies), if available, as the limit. The following figures and tables show the detailed data for transmission facility constraints where the thermal rating of the facility was exceeded post-contingency (i.e., an SOL exceedance).

Voltage stability limits, transient and control stability limits, and stability issues for interfaces or in areas with low weight short circuit ratios are monitored and managed using Generic Transmission Limits (GTLs). The Panhandle interface is the most active by far when it comes to the need to constrain flows across the interface to maintain the system within stability limits.

In the past three years, there has been an increase in the number of GTLs, mainly due to localized stability issues from wind and solar generation.


Figure 35 – Interface Operation Minutes Greater Than 90 percent of GTL

In 2019, there were 14,707 base case exceedances on elements > 100 kV for at least one SCED 5-minute interval where the element load exceeded 100 percent of the limit (normal rating). There were approximately 84,811 post-contingency exceedances on elements > 100 kV for at least one SCED 5-minute interval where the element post-contingency calculated load exceeded 100 percent of the limit (15-minute rating). Table 9 shows the list of the top constraints for 2019 by duration based on these criteria. Note that only four of these ten were on the 2018 list; however, the exceedances on the top two increased substantially in 2019.

Constraint (Limiting Element) 2019 Approx. Number of Hours

2018 Approx. Number of Hours

Interface: Panhandle 700.7 294.3

Hamilton Road - Maverick 138 kV 589.9 434.6

Rio Pecos - Woodward 2 138 kV 555.2 47.5

Dollarhide - No Trees Switch 138 kV 547.2 4.3

Sonora 138/69 kV A1 525.3 43.2

Bowie Axfmr1 138/69 kV 296.4

Bruni 138_69_1 138/69 kV 263.4 218.7

Cico - Comfort 138 kV 250.9 96.8

Horse Hollow - Omega 345 kV 248.4 517.7

North Laredo Switch - Piloncillo 138 kV 235.7 30.7

Table 12 – 2019 Top Constraints by Duration

0 0 0

18,067

0 0 0

28,157

0 0 0

16,831

0 0 0

38,993

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

North-Houston McCamey Valley Import Panhandle

Minutes > 90% of GTL

2019 Q1 2019 Q2 2019 Q3 2019 Q4


Figure 36 – Constraints by Month for 2019

ERCOT also posts a Chronic Congestion Summary report each month. This report provides the following:

(1) All security violations that were 125 percent or greater of the Emergency Rating for a single SCED interval or greater than 100 percent of the Emergency Rating for a duration of 30 minutes or more during the prior reporting month and the number of occurrences and congestion cost associated with each of the constraints causing the security violations on a rolling 12-month basis.

(2) Operating conditions on the ERCOT System that contributed to each security violation reported in paragraph (1) above.

The table below shows a summary of the Chronic Congestion for 2019 by the cause of the congestion. Overall numbers are similar to last year. Total estimated congestion rent for 2019 exceeded $1 billion. Double circuit contingencies were responsible for 41 percent ($436 million) of the total congestion rent.

Chronic Constraint Cause Sum of # Intervals > 100% for 30 Min or more

Sum of Congestion Rent ($)

Area Load/Gen pattern 186 $51,545,561

Multiple outages in the area 212 $45,064,667

Planned outages in the area 984 $288,830,139

High North-Houston import 7 $161,438

Redundant 432 $17,171

0

5,000

10,000

15,000

20,000

25,000

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Constraints by Month for 2019

138kV Lines 138kV Xfmrs 345kV Lines 345kV Xfmrs 69kV Lines 69kV Xfmrs Interfaces


Incorrect rating or telemetry 28 $578,895

TOAP available 179 $82,198,480

Forced outages in the area 220 $104,285,570

PCAP/RAP/RAS/MP available 1131 $268,291,470

High DC export 9 $17,094,278

High/low generation in the area 1056 $204,086,308

Remedial switching available 25 $2,840,416

Area phase-shifter adjusted 28 $3,173,611

High load in the area 14 $4,959,805

2019 Total 4511 $1,073,127,809

2018 Total 4233 $1,026,886,535

Table 13 – 2019 Chronic Constraint Causes

Figure 37 – 2019 Chronic Constraint Causes by Total Congestion Rent

Area Load/Gen pattern

4%

Multiple outages in the area5%

Planned outages in the area

22%

High North-

Houston import

0%

Redundant

10%

Incorrect rating or telemetry

1%TOAP available

4%Forced outages in the area

5%

PCAP/RAP/RAS/MP available

25%

High DC export0%

High/low generation in the area

23%

Remedial

generation

available0%

Area phase-shifter

adjusted

1%High load in

the area0%

2019 Chronic Constraint Causes by Duration


E. Reliability Unit Commitments

The Reliability Unit Commitment (RUC) process ensures that there is adequate Resource capacity and Ancillary Services capacity committed in the proper locations to serve ERCOT forecasted load. Day-ahead RUC (DRUC) commitments are made for the next operating day. Hour-ahead (HRUC) commitments are made for a specific operating hour(s) after the DRUC process is completed.

HRUC commitments totaled 29 units for 241 commitment hours, a significant decrease from 2018. The primary reason for HRUC commitments was to relieve local congestion or constraints on the transmission system, which accounted for approximately 71 percent of all HRUC hours.

Figure 38 – 2019 Hourly Reliability Unit Commitments by Month and Cause

0

5

10

15

20

25

0

10

20

30

40

50

60

70

80

90

100


Reliability Unit Committments - By Month & Cause

Constraint/Congestion Capacity Short Start Units

Ho

urs

Un

its


IV. Changing Resource Mix

Summary

Today’s resource mix continues to evolve with the addition of emerging technologies like inverter-based resources, improving storage techniques, and federal, state, and provincial policies favoring renewable generation. Transmission Planners, Balancing Authorities, asset owners, and system operators of the BPS may not have sufficient time to develop and deploy plans addressing reliability considerations that result from the rapidly evolving resource mix.

The integration of new technologies and distributed energy resources (DERs) are affecting availability as well as the ability of operators to see and control resources within their area.

Coal unit retirements are creating concerns with lack of fuel diversity and greater reliance on natural gas. Natural gas is beginning to serve as the baseload fuel, heightening the need for greater focus on gas infrastructure in order to identify potential risks to the BPS.

Finally, the minimum requirements surrounding frequency response, voltage, and ramping resulting from the acceleration of the changing resource mix need to be continually evaluated.

2019 highlights from the analysis of changing resource mix include:

Average synchronous inertia across most operating hours continued to increase in 2019 when compared to previous years. This is in spite of the retirement of additional large coal units.

Primary Frequency Response achieved a median value of 918 MW/0.1 Hz versus the NERC obligation of 381 MW/0.1 Hz.


Over 1,660 MW of coal and natural gas capacity was retired or mothballed in 2019. It was replaced by over 2,700 nameplate MW, over 90% of which was renewable generation capacity.

There is a long-term increasing trend in the maximum one-hour up ramp magnitudes for net load and solar generation.

Low voltage ride-through issues for wind turbines continue to occur during transient voltage disturbances on the transmission system.


1. Unit additions and retirements 2. Fuel mix analysis 3. Inertia analysis 4. Ramping analysis 5. Fossil generation capacity factor metrics 6. Renewable generation capacity factor metrics


Gas66,040

59%Coal/Lignite21,470

19%

Solar3000%

Wind17,130

15%

Hydro1,010

1%

Nuclear5,250

5%

Other9301%

2015 Generation Nameplate Capacity

Gas67,530

55%

Coal/Lignite16,250

13%

Solar3,180

3%

Wind28,630

23%

Hydro6101%

Nuclear5,250

4%

Other8701%

2019 Generation Nameplate Capacity (no Non-Modeled Units)

Since 2015, wind and solar capacity have increased from 15% to 26% of generation capacity while coal has reduced from 19% to 13%

0

1000

2000

3000

4000

5000

6000

7000

8000

Load Ramp Wind Ramp Net Load Solar

Max 1-Hour Up-Ramps2013 2014 2015 2016 2017 2018 2019

There is a long-term

increasing trend in the maximum one-hour up ramp magnitudes for

net load and solar generation.

Inertia data shows risk of

approaching critical inertia level is less than 50 hours per year.

In 2019, 1665 MW of fossil

generation was retired while 2463 MW of renewable generation was

approved for commercial operation.

90,000

100,000

110,000

120,000

130,000

140,000

150,000

160,000

170,000

180,000

0 50 100 150 200 250 300

Inertia Duration Curve

2015 2016 2017 2018 2019

Iner

tia

(MW

-sec

)

Number of Hours


Figure 39 – Changing Resource Mix - By the Numbers

Detailed Analysis

A. Unit Additions and Retirements

Retirements and Mothball Status – 1,745 MW

Unit Date Status MW Fuel Type

Southwest Mesa Nov 2019 Retired 80 Wind

Gregory Power Partners Mothballed 365 Gas

JT Deely #1 Mothballed 415 Coal

JT Deely #2 Mothballed 415 Coal

Gibbons Creek G1 Oct 2019 Retired 470 Coal

New Resources Approved for Commercial Operation – 2,718 MW

Unit Date MW Fuel Type

Tahoka Wind 3/1/2019 300 Wind

Oak Grove repower 6/13/2019 45 Coal

Odessa Ector uprate 6/13/2019 80 Gas

S_Hills Wind 8/7/2019 30 Wind

Midway Wind 8/15/2019 163 Wind

VictPort 9/30/2019 100 Gas

Lockett Wind 9/19/2019 184 Wind

Rio Nogales upgrade 11/20/2019 19 Gas

Torrecillas Wind 11/7/2019 300 Wind

Phoebe Solar 11/26/2019 250 Solar

Blue Bell Solar 11/7/2019 30 Solar

Foard City 12/5/2019 350 Wind

Levee 12/4/2019 11 Gas

Cabezon Wind 12/17/2019 238 Wind

Canadian Breaks Wind 12/17/2019 210 Wind

West of Pecos Solar 12/20/2019 100 Solar

Karankawa Wind 12/30/2019 207 Wind

Karankawa 2 Wind 12/30/2019 101 Wind

Additional coal units were retired or mothballed in 2019. Another coal unit is slated to retire in 2020. As the coal fleet and older gas units continue to age, it is expected that the pace of retirement of these units will increase in favor of more efficient generation. B. Fuel Mix Analysis

At the end of 2019, the reported nameplate capacity was 122,321 MW of all modeled generators. Wind comprised more than 23 percent of the total nameplate capacity, while coal dropped to 13 percent of the total nameplate capacity.


Figure 40 – 2019 Generation Nameplate Capacity

The portion of total energy supplied by natural gas increased in 2019 to 48 percent (up from 44 percent in 2018). As expected, the portion of total energy supplied by coal decreased from 25 percent in 2018 to 20 percent in 2019. This downward trend is likely to continue in 2020 with the retirement of additional coal units and greater energy supply from renewable generation.

Gas67,530

55%

Coal/Lignite16,250

13%

Solar3,180

3%

Wind28,630

23%

Hydro6101%

Nuclear5,250

4%

Other8701%

2019 Generation Nameplate Capacity (MW) (Modeled Units)


Figure 41 – 2019 Energy by Fuel Type

Natural Gas 181,770

48%

Coal77,857

20%

Nuclear41,314

11%

Wind76,708

20%

Water9560%

Solar4,398

1%

Other1,056

0%

2019 Energy (GWH) by Fuel Type

0%

10%

20%

30%

40%

50%

60%

Jan

-10

Apr

-10

Jul-

10

Oct

-10

Jan

-11

Apr

-11

Jul-

11

Oct

-11

Jan

-12

Apr

-12

Jul-

12

Oct

-12

Jan

-13

Apr

-13

Jul-

13

Oct

-13

Jan

-14

Apr

-14

Jul-

14

Oct

-14

Jan

-15

Apr

-15

Jul-

15

Oct

-15

Jan

-16

Apr

-16

Jul-

16

Oct

-16

Jan

-17

Apr

-17

Jul-

17

Oct

-17

Jan

-18

Apr

-18

Jul-

18

Oct

-18

Jan

-19

Apr

-19

Jul-

19

Oct

-19

Fuel Type as % of Total Energy

Natural Gas Coal Nuclear Renewable

12 per. Mov. Avg. (Natural Gas) 12 per. Mov. Avg. (Coal) 12 per. Mov. Avg. (Nuclear) 12 per. Mov. Avg. (Renewable)


Figure 42 – Energy by Fuel Type Trend

Figure 43 – Planned Generation with Signed Interconnect Agreements and Meeting

Planning Guide Requirements

C. Synchronous Inertia

ERCOT began calculating synchronous inertia in July 2014 in order to better understand the system impact of increased renewable generation. The time-based trend in inertia levels is shown in the following figure.

ERCOT calculated that the critical inertia level for the Interconnection is approximately 94 Gigawatt-seconds (GW-s). ERCOT uses a critical inertia level of 100 GW-s for its operating procedures and in particular its forward projections for ancillary services procurement of responsive reserves in the day-ahead market.


Figure 44 – Historical System Inertia Type

The calculated synchronous inertia versus the system net load for 2019 continues to show a strong linear relationship. There was a noticeable shift upward in the regression lines for 2018 versus 2017, indicating an overall average increase in synchronous inertia. 2019 inertia continued this trend. This is most likely due to higher inertia gas units replacing the retired coal generation. Finally, the heat map graph of 2019 inertia levels shows the weakest inertia time periods are HE 01, 02, 03, and 04 during the shoulder months of February, March, April, and November. This is essentially unchanged from 2018.

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

Jan-

15

Mar

-15

May

-15

Jul-1

5

Sep

-15

No

v-15

Jan-

16

Mar

-16

May

-16

Jul-1

6

Sep

-16

No

v-16

Jan-

17

Mar

-17

May

-17

Jul-1

7

Sep

-17

No

v-17

Jan-

18

Mar

-18

May

-18

Jul-1

8

Sep

-18

No

v-18

Jan-

19

Mar

-19

May

-19

Jul-1

9

Sep

-19

No

v-19

ERCOT System Inertia


Figure 45 – Inertia versus Net Load, 2017-2019

The number and types of units on-line and providing synchronous inertia is a significant factor in the overall inertia of the Interconnection. ERCOT is calculating system inertia in real-time based on the number and type of generation resources that are on-line. Each different resource type has a different inertia constant.

2019 (red trend line):y = 4.5945x + 88059

R² = 0.8676

2018 (blue trend line):

y = 4.767x + 92027R² = 0.8573

2017 (green trend line):y = 4.8213x + 73882

R² = 0.8885

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

10,000 20,000 30,000 40,000 50,000 60,000 70,000

Inertia vs Net Load, 2017-2019System Inertia (MW-sec)

Net Load (MW)


Figure 46 – 2019 Inertia by Month and Operating Hour

The following figure shows the calculated synchronous inertia versus the percentage of load served by intermittent renewable resources (IRR), i.e., wind and solar generation. This figure indicates a fairly linear relationship between inertia and the IRR percentage. The regression lines also show an overall increase in the average inertia for 2019 when compared to 2018.

Hour Jan Feb Mar Apr May Jun July Aug Sept Oct Nov Dec

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

Inertia (MW-Sec)

360,000

352,000

344,000

336,000

328,000

320,000

312,000

304,000

296,000

288,000

280,000

272,000

264,000

256,000

248,000

240,000

232,000

224,000

216,000

208,000

200,000

192,000

184,000

176,000

168,000

160,000

152,000

144,000

136,000


Figure 47 – Inertia versus Percentage of Load Served by IRRs

The minimum hourly inertia level in 2019 was 134.6 GW-s on March 27, 2019 at 1:00 a.m., when the IRR penetration level was 50.2 percent and system load was 29,426 MW (net load of 14,645 MW).

Year Minimum Inertia (GW-s) Load (MW) Net Load (MW) IRR %

2015 130.3 27,798 20,569 26.1%

2016 138.4 26,839 14,797 44.9%

2017 130.0 28,443 13,178 53.7%

2018 128.8 28,412 13,452 52.7%

2019 134.6 29,426 14,645 50.2%

Table 14 – Minimum Inertia for 2015-2019

The following figure shows the inertia duration curve for 2015-2019 comparing the number of hours during the lowest inertia levels. This graph shows two key points. The first is the proximity of the actual inertia to the calculated critical inertia level of 94 GW-sec. The second is the number of hours, i.e. level of risk, for how long ERCOT operates at these lower inertia levels. Note that the 30 GW-sec of inertia margin equates to over 6,000 MW of combined cycle generation.

2019 (Purple):y = -298598x + 315119

R² = 0.4482

2018 (Blue):y = -319976x + 319616

R² = 0.4248

2017 (Red):y = -329478x + 298528

R² = 0.4775

2016 (Green):y = -299132x + 292539

R² = 0.353

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

0% 10% 20% 30% 40% 50% 60%

IRR % vs Inertia (MW-sec)


Additional information on inertia and its relationship to primary frequency response can be found in Section I, Resource Adequacy and Performance.

Figure 48 – Inertia Duration Curve

D. Net Demand Ramping Variability

Changes in the amount of non-dispatchable resources, system constraints, load behaviors and the generation mix can affect the ramp rates needed to keep the system in balance. Conventional resources must have sufficient ramping capability to maintain the generation-load balance when intermittent renewables have large up or down ramps. ERCOT calculates the system ramp capability in real-time to ensure that this ramping variability can be met. If insufficient ramping capability is not available, ERCOT will bring additional quick start resources on line.

Ramping Variability Load Wind Gen Solar Gen Net Load

Maximum One-Hour Increase 5,215 MW 4,218 MW 1,141 6,899

Maximum One-Hour Decrease -4,663 MW -3,607 MW -1,092 -6,150

Maximum Three-Hour Increase 13,304 MW 7,056 MW 1,158 MW 15,136 MW

Maximum Three-Hour Decrease -11,853 MW -6,565 MW -1,333 MW -15,292 MW

Table 15 – Maximum and Minimum Load, Wind, Solar, and Net-Load Ramps for 2019

120,000

130,000

140,000

150,000

160,000

170,000

180,000

0 50 100 150 200 250 300 350 400 450 500

Inertia Duration Curve

2015 2016 2017 2018 2019

Iner

tia

(MW

-sec

)

Number of Hours


There is a long-term increasing trend in the maximum one-hour up ramps for net load and solar. The following figure shows a comparison of the maximum one-hour load, net load, and wind ramps for 2019 compared to previous years.


Figure 49 – Maximum One-Hour Ramps for 2013-2019

0

1000

2000

3000

4000

5000

6000

7000

8000


Max 1-Hour Up-Ramps2013 2014 2015 2016 2017 2018 2019

-12000

-10000

-8000

-6000

-4000

-2000

0


Max 1-Hour Down-Ramps2013 2014 2015 2016 2017 2018 2019


E. Fossil generation capacity factor metrics

The average net capacity factor (NCF) from GADS data for the coal fleet decreased in 2019 when compared to 2018, but still remains near the five-year average. The net capacity factor for the combined cycle fleet remained steady in 2019 and slightly above the five-year average. With the retirement or mothball of ten coal units between 2018 and 2019, it is interesting to note that the NCF and SF for the remaining in-service coal units have not changed significantly.


2015 2016 2017 2018 2019 5-Yr Avg


Coal/Lignite Capacity Factor (NCF)

57.1% 58.1% 64.6% 66.3% 60.3% 61.3%

Coal/Lignite Service Factor (SF)

79.2% 76.2% 83.2% 84.2% 80.7% 80.7%

Combined Cycle Capacity Factor (NCF)

56.5% 52.0% 42.9% 52.3% 52.2% 51.2%

Combined Cycle Service Factor (SF)

68.2% 67.4% 57.4% 59.2% 58.2% 62.1%

Fossil fleet – All (NCF) 45.5% 44.2% 43.3% 46.7% 46.8% 45.3%

Fossil fleet – All (SF) 49.8% 48.4% 46.1% 50.9% 51.7% 49.4%

Table 16 – Fossil Generation Capacity/Service Factor Metrics

Figure 50 – 2015-2019 Fossil Generation Net Capacity Factors

F. Renewable generation capacity factor metrics

Wind generation produced a total of 76,708 GWH in 2019, an increase of 10 percent from 2018, equal to the approximate increase in wind nameplate capacity of 10 percent. Wind generation, as a percentage of total ERCOT energy produced, increased to 20 percent in 2019, up from 18.5 percent in 2018. In 2019, hourly wind generation reached a maximum of

0%

10%

20%

30%

40%

50%

60%

70%

Coal/Lignite Combined Cycle Fossil fleet

Fossil Fleet Net Capacity Factors, 2015-2019

2015 2016 2017 2018 2019

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Coal/Lignite Combined Cycle Fossil fleet

Fossil Fleet Service Factors, 2015-2019

2015 2016 2017 2018 2019


19,580 MW on January 21, 2019, at 7:00 p.m., and hourly wind generation served a maximum of 57.5 percent of system demand on November 26, 2019, at 3:00 a.m. The following graphs show the historical trends for wind generation growth in the region. The blue bars represent the wind generation for the month and the black line represents the 12-month moving average.

Figure 51 – 2008-2019 Wind Generation as a Percentage of ERCOT Total Energy

Figure 52 – 2008-2019 Wind Generation as Percentage of ERCOT Total Energy by Month

0%

5%

10%

15%

20%

25%

30%

Jan

-08

May

-08

Sep

-08

Jan

-09

May

-09

Sep

-09

Jan

-10

May

-10

Sep

-10

Jan

-11

May

-11

Sep

-11

Jan

-12

May

-12

Sep

-12

Jan

-13

May

-13

Sep

-13

Jan

-14

May

-14

Sep

-14

Jan

-15

May

-15

Sep

-15

Jan

-16

May

-16

Sep

-16

Jan

-17

May

-17

Sep

-17

Jan

-18

May

-18

Sep

-18

Jan

-19

May

-19

Sep

-19

Wind % of Total Energy

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Wind % of Total Energy


The following graph shows the distribution of capacity factor for all wind generation for the summer peak hours-ending of 15:00-19:00 for 2019.

Figure 53 – 2019 Wind Capacity Factor for Summer Peak Hours

Wind facilities greater than 200 MW began mandatory reporting in GADS-Wind in 2018. Wind facilities greater than 100 MW began mandatory reporting in GADS-Wind in 2019. GADS-Wind provides similar metrics as GADS to compare unit-level and fleet-level performance. Two of these methods provide resource-level and equipment-level performance rates. In 2019, 113 ERCOT wind facilities submitted a total of 1,986 unit-months of data in GADS-Wind. Resource-level metrics look at the resource as a whole. Pooled equipment metrics provide a mechanism to look at sub-group performance of turbines of similar capacity. A summary of key performance metrics based on resource versus pooled equipment values for the ERCOT wind generators for 2019 is provided in the following table.

Metric ERCOT Region GADS-Wind Data 2018

ERCOT Region GADS-Wind Data 2019

Resource Equipment Resource Equipment

Net Capacity Factor (PRNCF and PENCF)

37.6% 39.7% 37.5% 39.7%

Equivalent Forced Outage Rate (PREFOR and PEEFOR)

10.3% 4.2% 12.1% 5.8%

Equivalent Scheduled Outage Rate (RESOR and PEESOR)

1.2% 1.1% 1.6% 1.5%

0

5

10

15

20

25

0% 4% 8% 12% 16% 20% 24% 28% 32% 36% 40% 44% 48% 52% 56% 60% 64% 68% 72% 76% 80%

Wind Capacity Factor, Jun-Aug 2019, HE 1500-1900Hours


Equivalent Availability Factor (REAF and PEEAF)

88.8% 94.0% 87.0% 91.8%

Table 17 – ERCOT Wind Generation Performance Metrics, 2019

Pooled Resource Equivalent Forced Outage Rate (PREFOR): Probability of forced plant downtime when needed for load.

Resource Equivalent Scheduled Outage Rate (RESOR): Probability of maintenance or planned plant downtime when needed for load.

Resource Equivalent Availability Factor (REAF): Percent of time the plant was available.

Pooled Resource Net Capacity Factor (PRNCF): Percent of actual plant generation.

Pooled Equipment Equivalent Forced Outage Rate (PEEFOR): Probability of forced WTG equipment downtime when needed for load.

Pooled Equipment Equivalent Scheduled Outage Rate (PEESOR): Probability of maintenance or planned WTG equipment downtime when needed for load.

Pooled Equipment Net Capacity Factor (PENCF): Percent of actual WTG equipment generation while on line.

Pooled Equipment Equivalent Availability Factor (PEEAF): Percent of time the WTG equipment was available.

Figure 54 – 2019 Wind GADS Equivalent Availability Factors

75%

80%

85%

90%

95%

100%

All Summer Winter Shoulder

Wind Generation - Equivalent Availability FactorsResource Pooled Equipment


Figure 55 – 2019 GADS-Wind Equivalent Outage Rates

Figure 56 – 2019 GADS-Wind Net Capacity Factors

0%

2%

4%

6%

8%

10%

12%

14%

16%

Coastal North Panhandle South West

Wind GenerationResource Equiv Forced Outage Rate Resource Equiv Scheduled Outage Rate

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

All Summer Winter Shoulder

Wind Generation - Average Net Capacity Factors, All HoursResource Pooled Equipment


Utility-scale solar generation within the region continued its significant growth in 2019. The amount of energy provided by solar generation increased 34 percent versus 2018. The following graphs show the historical trends for solar generation growth in the region. The blue bars represent the solar generation for the month and the black line represents the 12-month moving average.

Figure 57 – 2015-2019 Solar Generation MWH

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

500,000

Jan

-15

Mar

-15

May

-15

Jul-

15

Sep

-15

No

v-15

Jan

-16

Mar

-16

May

-16

Jul-

16

Sep

-16

No

v-16

Jan

-17

Mar

-17

May

-17

Jul-

17

Sep

-17

No

v-17

Jan

-18

Mar

-18

May

-18

Jul-

18

Sep

-18

No

v-18

Jan

-19

Mar

-19

May

-19

Jul-

19

Sep

-19

No

v-19

Solar Energy (MWH)


Figure 58 – 2015-2019 Solar Generation as a Percentage of ERCOT Total Energy

Figure 59 – 2019 Solar Capacity Factor for Summer Hours

0.00%

0.20%

0.40%

0.60%

0.80%

1.00%

1.20%

1.40%

1.60%Ja

n-1

5

Mar

-15

May

-15

Jul-

15

Sep

-15

No

v-15

Jan

-16

Mar

-16

May

-16

Jul-

16

Sep

-16

No

v-16

Jan

-17

Mar

-17

May

-17

Jul-

17

Sep

-17

No

v-17

Jan

-18

Mar

-18

May

-18

Jul-

18

Sep

-18

No

v-18

Jan

-19

Mar

-19

May

-19

Jul-

19

Sep

-19

No

v-19

Solar % of Total Energy

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

2019 Summer Solar Capacity Factor by Operating Hour


V. Human Performance

Summary

The BPS is becoming more complex and as the industry faces turnover in technical expertise, it will have difficulty staffing and maintaining appropriately skilled workers. New controls needed to maintain compliance with reliability and CIP compliance requirements have added complexity to many positions. In addition, inadequate human performance (HP) makes the grid more susceptible to both active and latent errors, negatively impacting reliability. HP weaknesses may hamper an organization’s ability to identify and address precursor conditions to promote effective mitigation and behavior management. There continues to be a need for skilled workers (such as protection engineers) to prevent both active and latent errors which negatively affect reliability.

2019 highlights from the analysis of human performance include:

Outage rates for Protection System Misoperations caused by human error are showing an improving, downward trend.

Generator forced outage rates caused by human error dropped significantly in 2019 compared to previous years.


Causal analysis of human errors in Protection System Misoperations shows repeated issues due to lack of adequate error-checking processes and procedures.


1. Transmission outages initiated by human error 2. Generator outages initiated by human error 3. Protection system misoperations caused by human performance 4. Analysis of system events with human performance cause codes

Detailed Analysis

A. Outages initiated by human error Outage rates for protection system misoperations caused by human error are showing an improving, downward trend. Outage rates for generator immediate forced outages caused by human error were significantly lower in 2019 when compared to previous years.


Element Type Metric 2015 2016 2017 2018 2019 5-Yr Avg

AC Circuit 300-399 kV Outages per Element Initiated by Human Error

1.2% 2.9% 0.7% 1.5% 1.2% 1.5%

AC Circuit 100-199 kV Outages per Element Initiated by Human Error

2.2% 1.6% 1.3% 1.1% 2.1% 1.7%

Transformer 300-399 kV

Outages per Element Initiated by Human Error

1.0% 1.0% 0.5% 0.5% 0.5% 0.7%

Generator Immediate Forced Outages Initiated by Human Error

4.1% 3.9% 4.2% 3.9% 2.4% 3.7%

Protection Systems Misoperation Rate Caused by Human Error

3.6% 2.9% 3.2% 2.9% 2.7% 3.1%

Table 18 – Outages Rates Caused by Human Error

Figure 60 – Outage Rates Caused by Human Error

Since 2015, there have been 462 generation immediate forced outages caused by human error in ERCOT. The breakdown of the causes is shown below.

0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

AC Circuit300-399 kV

AC Circuit100-199 kV

Transformer300-399 kV

GeneratorForced Outages

ProtectionSystem

Misoperations

Outage Rates Initiated by Human Error

2015 2016 2017 2018 2019


Figure 61 – Generator Forced Outage Human Error Issues

Human Performance in Protection System Misoperations

Human performance remains the primary causal factor in misoperations, primarily due to incorrect settings and/or as-left errors. The following list provides examples of common actual human error-related misoperations since 2015 in the ERCOT region. The number in parentheses represents the number of occurrences.

(21) Incorrectly wired CTs or PTs (polarity, ratio, etc.)

(10) CTs left in the shorted or open position

(25) Field settings did not match the issued settings

(18) Field wiring did not match the engineering drawings

(10) Zero-sequence polarization versus negative sequence polarization

( 9) Pilot relaying disabled on one end of the line

( 4) Cut-off switches left in wrong position

( 9) Modeling

( 3) Relay firmware versions not current, or different on each end of a line

Operator error

58%

Maintenance personnel

error

19%

Contractor error13%

Operating

procedure error

6%

Maintenance procdure error

3%

Contractor procedure

error1%

2015-2019 Generator Human Error - by Cause

0

100

200

300

400

500

600

700

800

0

20

40

60

80

100

120

2015 2016 2017 2018 2019

Lost Production Potential (GWH) # of HP Events

Lost Production for Human Error Generator Forced Outages

Events GWH


Figure 62 – Protection System Misoperations Trend Caused by Human Performance

B. Human performance in system events

The NERC Cause Code process provides a systematic approach to assigning cause code(s) after an event on the BPS is analyzed. Appropriate use of this method after event analysis will result in effective labeling, collection, and trending of causes. It will also will lead to the proper application of risk management procedures to develop and implement appropriate corrective and proactive actions.

Ninety-one events in ERCOT have been analyzed using this cause code process, with 503 root cause and contributing cause codes assigned. Approximately 55 percent of the assigned root and contributing cause codes are related to potential human performance issues (shown in red).

0%

10%

20%

30%

40%

50%

60%

70%

80%

2015Q1

2015Q2

2015Q3

2015Q4

2016Q1

2016Q2

2016Q3

2016Q4

2017Q1

2017Q2

2017Q3

2017Q4

2018Q1

2018Q2

2018Q3

2018Q4

2019Q1

2019Q2

2019Q3

2019Q4

Protection System Misoperations - Human Error %


Figure 63 – Event Analysis Human Performance Cause Coding

0

10

20

30

40

50

60

70

80

90D

esig

n in

put

Des

ign

ou

tpu

t

Des

ign

/do

cum

enta

tio

n

De

sign

/In

sta

llati

on

…

Ope

rab

ility

of…

Cal

ibra

tio

n f

or

inst

rum

en

ts

Peri

odi

c /c

orr

ecti

ve…

Insp

ecti

on

/tes

tin

g

Pro

cure

me

nt

con

tro

l

De

fect

ive

, fa

iled

, or…

Equ

ipm

ent

inte

ract

ions

Skill

-ba

sed

err

or

Rul

e-ba

sed

err

or

Man

agem

ent

met

ho

ds

Res

ourc

e m

anag

emen

t

Wo

rk o

rgan

izat

ion

/pla

nn

ing

Sup

erv

iso

ry m

eth

od

s

Ch

ange

man

agem

ent

Wri

tten

com

mun

icat

ion

…

Wri

tten

co

mm

un

icat

ion

…

Wri

tten

com

mun

icat

ion

not

…

Ver

bal

co

mm

un

icat

ion

No

tra

inin

g p

rovi

ded

Trai

nin

g m

eth

od

s

Trai

ning

mat

eria

l

Exte

rnal

ph

eno

men

a

Haz

ard

ous

mat

eria

l

Sup

plie

r p

rob

lem

Un

able

to

iden

tify

ro

ot c

ause

Failu

re m

od

e o

nly

Oth

er p

arti

es in

volv

ed

Event Analysis Cause Codes


VI. Bulk Power System Planning

Summary

BPS planning has transitioned from centrally planned and constructed resources based on forecasted load growth and reliability projects to more reactive, rather than proactive, planning based on the integration of new resources and technologies driven by policies and incentives. The speed with which these resources are being integrated in some areas results in planners not having sufficient time to update (or create) system models and scenarios of potential future states to identify system reliability needs. Planners are challenged to implement mitigation plans or reliability upgrades to address likely scenarios, driving the need for more real-time operating procedures.

The changing resource mix has made it challenging to evaluate BPS stability, including inertia and frequency response, voltage support (adequate dynamic and static reactive compensation), and ramping constraints.

2019 highlights from the analysis of bulk power system planning include:

There were no reported ERCOT IROL exceedances in 2019

Summer Peak: Actual 74,533 MW versus projected 74,853 MW

Winter Peak: Actual 60,646 MW versus projected 61,780 MW

Peak hourly wind generation: 19,580 MW on January 21, 2019

Peak hourly renewable penetration: 57.5% on November 26, 2019


Planning reserve margins in the five-year planning horizon continue to show several years below the reference reserve margin level of 13.75 percent.

West Texas load growth continues to grow at rates of five-seven percent per year. Transmission entities in the area are implementing upgrades to keep up with this growth rate.

As of December 2019, ERCOT projections indicate utility-scale solar generation will increase 273 percent to over 8,500 MW and wind generation will increase 41 percent to more than 33,700 MW during the next two years (based on current signed generation interconnect agreements with financial security). During the same period, only 640 MW of new gas units are projected. The growth rate in renewable generation will continue to test ERCOT’s ability to maintain adequate system inertia and ancillary services.


1. Net energy for load 2. Reserve margin 3. IROL exceedances 4. Transmission element availability metrics 5. EEAs 6. Distributed energy resources and non-modeled generation


Detailed Analysis

A. Net energy for load In 2019, total annual energy usage was 383.67 GWH, an increase of 1.8 percent over 2018. Peak hourly demand was 74,533 MW on August 12, 2019. The West Load Zone has seen the largest load energy usage increase (7.4 percent per year since 2015). The LCRA Load Zone experienced negative load energy increases over the same time frame.

Figure 64 – Annual Energy and Peak Demand

50,000

55,000

60,000

65,000

70,000

75,000

80,000

250,000

275,000

300,000

325,000

350,000

375,000

400,000

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Annual Energy Peak DemandGWH MW


Figure 65 – Energy by Load Zone

Figure 66 – Peak Demand by Load Zone

The weather zone with the largest load energy usage increase was the Far West (10.7 percent per year since 2015).

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

LZ_AEN LZ_CPS LZ_HOUSTON LZ_LCRA LZ_NORTH LZ_RAYBN LZ_SOUTH LZ_WEST

2011 2012 2013 2014 2015 2016 2017 2018 2019

Total Energy by Load Zone (GWH)

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

LZ_AEN LZ_CPS LZ_HOUSTON LZ_LCRA LZ_NORTH LZ_RAYBN LZ_SOUTH LZ_WEST

2011 2012 2013 2014 2015 2016 2017 2018 2019

Peak Demand by Load Zone (MW) Coincident with ERCOT Peak


Figure 67 – Energy by Weather Zone

Figure 68 – Peak Demand by Weather Zone

Overall energy growth rate has averaged 2.0 percent per year and demand growth rate has averaged 1.4 percent per year since 2015.

B. Reserve margin

NERC develops and publishes its Long Term Reliability Assessment (LTRA) each December to independently assess each region in an effort to identify trends, emerging

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

Coast East Far West North Central North South Central South West

2011 2012 2013 2014 2015 2016 2017 2018 2019

Total Energy by Weather Zone (GWH)

0

5,000

10,000

15,000

20,000

25,000

30,000

Coast East Far West North Central North South Central South West

2011 2012 2013 2014 2015 2016 2017 2018 2019Peak Demand by Weather Zone (MW) Coincident with ERCOT Peak


issues, and potential risks during the 10-year horizon. A key component of the LTRA is an evaluation of the peak demand and planning reserve margins, which are based on average weather conditions and the forecasted economic growth conditions at the time of the assessment.

ERCOT publishes its Capacity and Demand Report (CDR) twice each year, in December and May. The purpose of the CDR is to provide updates to the planning reserve margins based on current load forecasts and resource availability.

While both of these reports are focused on the long-term planning reserve margins, the results will differ due to data collection dates and forecasting of load.

In the LTRA, NERC uses a reference planning reserve margin of 13.75 percent, based on a one event in 10 year loss of load probability. The LTRA shows a planning reserve margin below the 13.75 percent target for four of the next five years. The ERCOT CDR shows a planning reserve margin below the 13.75 percent target for only two of the next five years.

The figure below shows the summer reserve margins calculated based on the December 2019 NERC LTRA and the December 2019 ERCOT CDR.

Figure 69 – Summer Peak Reserve Margins

50,000

55,000

60,000

65,000

70,000

75,000

80,000

85,000

90,000

95,000

100,000

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

2020 2021 2022 2023 2024

Summer Reserve EstimatesERCOT December 2019 CDR NERC December 2019 LTRA

Summer Capacity Summer Firm Load Forecast

Summer Peak Demand


C. Energy emergency alerts

There were two EEA level 1 alerts issued in August 2019. These events are discussed further in Section II. The figure below shows the historical EEA data.

Figure 70 – EEA Events by Year

D. Distributed energy resources and non-modeled generation

DERs include any non-BES resource located solely within the boundary of the distribution utility, such as:

Distribution and behind-the-meter generation

Energy storage facilities

Microgrids

Cogeneration

Stand-by or back-up generation

Increasing amounts of DER will change how the distribution system interacts with the BPS by transforming the distribution system into an active energy source. Currently, the aggregated effect of DER is not fully represented in BPS models or real-time operating tools. There are also differing expectations for DER performance between current PUCT rules and IEEE standards. Issues with DERs include:

0

1

2

3

4

5

6

7

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

EEA 1 EEA 2 EEA 3

EEA EVENTS


Modeling (both steady-state and dynamic)

Ramping and energy-load balance

Reactive power and voltage stability

Frequency ride-through

System protection and islanding protection

Visibility and control

Unanticipated power flows

Load forecast errors

Currently under ERCOT Protocols, distributed generation resources greater than 1 MW must register with ERCOT and provide resource registration data per Protocol 16.5(5) and Planning Guide 6.8.2. Additionally, P.U.C. SUBST. R. 25.211(n) requires every electric utility to file (by March 30 of each year) a distributed generation Interconnection report with the commission for the preceding calendar year that identifies each distributed generation facility interconnected with the utility’s distribution system, including ownership, capacity, and whether it is a renewable energy resource.

At the end of 2019, ERCOT had approximately 1,510 MW of non-modeled generation capacity and 638 MW of distributed generation resources (DGR) that has provided data for mapping capacity to their modeled loads. The data is broken down by fuel type below.


Figure 71 – Non-Modeled Generation Capacity by Fuel Type

0

100

200

300

400

500

600

700

800

900

1000

ERCOT Non-Modeled and Distributed Generation Capacity - 2019

Non-Modeled Generation Distributed Generation

0

100

200

300

400

500

600

700

800

ERCOT Non-Modeled Generation Capacity

2018 2019


VII. Loss of Situational Awareness

Summary

The loss of situational awareness can be a precursor or contributor to a BPS event. It also highlights emerging challenges with visibility into DER impacts on the grid. Loss of situational awareness can also occur when control room operators do not have sufficient information and visibility to manage the grid in real-time. Loss of situational awareness due to insufficient communication and data regarding neighboring entities’ operations is a risk as operators may act on incomplete information. The increased dependency on natural gas as a predominant fuel source presents challenges in real-time to system operators and situational awareness must now include gas sources, pipeline, gas storage, infrastructure maintenance, compressor station location and failures, and deliverability concerns.

2019 highlights from the analysis of loss of situational awareness include:

Convergence rates for ERCOT’s State Estimator continue to perform at high levels.

Telemetry availability rates remain stable at approximately 97 percent overall. Telemetry accuracy metrics are showing an improving trend.


A total of four (two Category 1) loss of SCADA or EMS events were reviewed in 2019. Total duration was approximately 6-and-a-half hours.

Metrics and Data Associated with this Area

1. Analysis of loss of EMS or SCADA events 2. State Estimator convergence metrics 3. Telemetry availability metrics

Detailed Analysis

A. Analysis of loss of EMS or SCADA events

Loss of EMS/SCADA events continue to be a focus point at the NERC and regional levels. Category 1 events include loss of operator ability to remotely monitor and control BES elements, loss of communications from SCADA Remote Terminal Units (RTU), unavailability of Inter-Control Center Communications Protocol (ICCP) links, loss of the ability to remotely monitor and control generating units via Automatic Generation Control (AGC), and unacceptable State Estimator or Contingency Analysis solutions for more than 30 minutes.

Loss of SCADA or EMS events reviewed in 2019 include the following:

A Transmission Operator (TOP) had a loss of SCADA visibility and control due to a hardware failure on a server. A software bug prevented the SCADA system from performing a successful failover to redundant servers.

A TOP lost monitoring and control capabilities while operating from its backup control center.


The figure below shows the historical ERCOT loss of SCADA or EMS events by year.

Figure 72 – Loss of EMS and SCADA Events by Year

The figure below shows the historical ERCOT loss of SCADA or EMS events by duration.

0

1

2

3

4

5

6

7

8

2013 2014 2015 2016 2017 2018 2019

Loss of EMS and SCADA Events Reported by Year


Figure 73 – Loss of EMS and SCADA Events by Duration

Figure 74 – Loss of EMS and SCADA Events by Type and Attribute

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

0-60 60-120 120-180 180-240 240-300 >300

Number of Cat 1h Events by Duration (minutes)

0

5

10

15

20

25

30

Loss of Ability toMonitor or Control

Loss of SE/RTCA Loss of ICCP Loss of RTU Loss of AGC

Category 1h Events by Type and Attributes

Communications Software Facility Maintenance


B. State Estimator convergence metrics

ERCOT’s goal for State Estimator convergence is 97 percent or higher. In 2019, the convergence rate was 99.99 percent.

Figure 75 – ERCOT State Estimator Convergence Rate

C. Telemetry availability metrics

ERCOT telemetry performance criteria states that 92 percent of all telemetry provided to ERCOT must achieve a quarterly availability of 80 percent. The following figure shows the telemetry availability metric per the ERCOT telemetry standard. For 2019, the total number of telemetry points failing the availability metric averaged 4,422 each month, or 3.9 percent of the total system telemetry points.

97.50

98.00

98.50

99.00

99.50

100.00

Jan-

15

Mar

-15

May

-15

Jul-1

5

Sep

-15

No

v-1

5

Jan-

16

Mar

-16

May

-16

Jul-1

6

Sep

-16

No

v-1

6

Jan-

17

Mar

-17

May

-17

Jul-1

7

Sep

-17

No

v-1

7

Jan-

18

Mar

-18

May

-18

Jul-1

8

Sep

-18

No

v-1

8

Jan-

19

Mar

-19

May

-19

Jul-1

9

Sep

-19

No

v-1

9

SE Convergence Rate


Figure 76 – ERCOT Telemetry System Availability

D. Telemetry accuracy metrics

ERCOT uses several processes to verify the accuracy of telemetry when compared to State Estimator solutions. These include:

1. Residual difference between telemetered value and State Estimator value on Transmission Elements over 100kV is <10 percent of emergency rating or < 10MW (whichever is greater) on 99.5 percent of all samples during a month period.

2. The sum of flows into any telemetered bus is less than the greater of five MW or five percent of the largest Normal line rating at each bus.

3. The telemetered bus voltage minus state estimator voltage shall be within the greater of two percent or the accuracy of the telemetered voltage measurement involved for at least 95 percent of samples measured.

The following figures show the historic performance for these metrics.

80,000

84,000

88,000

92,000

96,000

100,000

104,000

108,000

112,000

116,000

120,000

90.0%

91.0%

92.0%

93.0%

94.0%

95.0%

96.0%

97.0%

98.0%

99.0%

100.0%Ja

n-16

Mar

-16

May

-16

Jul-1

6

Sep

-16

No

v-16

Jan-

17

Mar

-17

May

-17

Jul-1

7

Sep

-17

No

v-17

Jan-

18

Mar

-18

May

-18

Jul-1

8

Sep

-18

No

v-18

Jan-

19

Mar

-19

May

-19

Jul-1

9

Sep

-19

No

v-19

% System Availability Total Telemetry Points


Figure 77 – State Estimator versus Telemetry Accuracy

Figure 78 – Bus Summation Telemetry Accuracy

0

200

400

600

800

1,000

1,200

State Estimator vs Transmission Telemetry AccuracyBPS Lines and Transformers with Values Out of Range

0

50

100

150

200

250

300

Bus Summation Telemetry AccuraryBPS Buses with Values Out of Range


Figure 79 – Bus Voltage Telemetry Accuracy

0

2

4

6

8

10

12

14

16

18

20

Important Bus Voltage Telemetry AccuracyNumber of Buses with Values Out of Range


VIII. Increasing Complexity in Protection and Control Systems

Summary

Failure to properly design, coordinate, commission, operate, maintain, prudently replace, and upgrade BPS control system assets could negatively impact system resilience. More frequent and wider-spread outages initiated or exacerbated by protection and control system misoperations or failures could occur as a result. Asset management strategies are evolving to include greater amounts of digital network based controls for substation assets that introduce cybersecurity risks.

2019 highlights from the analysis of protection system misoperations include:

There is a positive downward trend in the number of misoperations occurring each year due to incorrect settings, communication failures, and relay failures.

There was a decrease in the overall Protection System Misoperation rate in 2019, to 6.3 percent for 2019 versus 7.3 percent for 2018.


Incorrect settings, logic, and design errors remained the largest cause of misoperations, accounting for 32 percent of misoperations in 2019.

Multiple system events occurred in 2019 where Protection System Misoperations expanded the magnitude of the transmission elements outaged or caused loss of generation or load.


1. Protection system misoperation rate 2. Analysis of system events with protection system misoperations 3. Transmission outages initiated by failed protection system equipment

Detailed Analysis

A. Protection System Misoperation statistics

Since January 2015, the overall transmission system Protection System Misoperation rate has a slight downward trend, from 8.8 percent in 2015 to 6.3 percent in 2019.

138 kV 2015 2016 2017 2018 2019 5-Yr Avg

Number of Misoperations

155 97 120 101 115 118

Number of Events 1689 1815 1676 1639 1852 1734

Percentage of Misoperations

9.2% 5.3% 7.2% 6.2% 6.3% 6.8%

345 kV 2015 2016 2017 2018 2019 5-Yr Avg


29 32 30 48 40 36


Number of Events 673 584 606 548 715 625


4.3% 5.5% 4.9% 8.8% 5.6% 5.8%

< 100 kV 2015 2016 2017 2018 2019 5-Yr Avg


0 3 0 5 1 2

Number of Events 93 74 76 44 55 68


0.0% 4.0% 0.0% 11.4% 1.9% 2.6%

Table 19 – Protection System Misoperation Data

Figure 80 – Protection System Misoperation Trends

In 2019, three main categories account for 63 percent of the total misoperations: incorrect settings/logic/design (32 percent), relay failures (18 percent), and other (14 percent).

There is a positive downward trend in the number of misoperations occurring each year due to incorrect settings and relay failures. Misoperations due to communications failures and logic errors are also showing a positive downward trend.

However, AC system errors, other/explainable errors, and as-left personnel errors are showing negative upward trends.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

2015Q1

2015Q2

2015Q3

2015Q4

2016Q1

2016Q2

2016Q3

2016Q4

2017Q1

2017Q2

2017Q3

2017Q4

2018Q1

2018Q2

2018Q3

2018Q4

2019Q1

2019Q2

2019Q3

2019Q4

Overall 345kV 138kV 4 per. Mov. Avg. (Overall)% Misoperation Rate


Figure 81 – Protection System Misoperation Count by Cause 2011-2019

The following figures show a comparison of protection system misoperation rates between a sample of different ERCOT Transmission Owners compared to the aggregated region performance and a comparison of protection system misoperation rates between the different NERC regions for the period of January 2015 through December 2019.

Figure 82 – Protection System Misoperation Rates by Entity 2015-2019

0

10

20

30

40

50

60

70

2011 2012 2013 2014 2015 2016 2017 2018 2019

0%

5%

10%

15%

20%

25%

30%

Entity #1 Entity #2 Entity #3 Entity #4 Entity #5 Entity #6 Entity #7 Entity #8 Entity #9 Entity #10 Entity #11 Entity #12 Entity #13 Entity #14 ERCOT total

Overall % Misoperation Rate by Entity 2015-2019


Figure 83 – Protection System Misoperation Rates by Region 2015-2019

B. Analysis of system events with protection system misoperations

Protection System Misoperations were a contributing factor in several key system events in 2019. In all of these events, the misoperations expanded the magnitude of the transmission elements outaged or caused loss of generation or load.

1) On April 11, 2019, a 345 kV line protective relay misoperated for a fault due to an incorrect setting of a ground overcurrent protective function and caused the loss of a generation facility, resulting in the loss of 664 MW of generation.

2) On August 10, 2019, a 345 kV line protective relay failed and misoperated for a fault and caused the loss of a generation facility, resulting in the loss of 1,244 MW of generation.

3) On August 21, 2019, a fault occurred on 345 kV line. The misoperation occurred at a

nearby wind generation facility due to an incorrect setting of a ground distance protective function, causing the loss of 160 MW of generation. A second misoperation occurred on a nearby 138 kV line due to an incorrect setting of a ground overcurrent function. Mutual coupling effects were not properly accounted for in the relay setting calculation.

4) On October 6, 2019, a fault occurred on a 138 kV capacitor bank. The capacitor bank

protection failed to trip due to an incorrect setting of an overcurrent element. The delayed clearing time caused low voltages throughout the area and resulted in the loss of multiple generation facilities, with 1,100 MW of generation.

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

MRO NPCC RF SERC TRE WECC NERC

Regional Misoperation Rates

2015 Q1 - 2015 Q4 2016 Q1 - 2016 Q4 2017 Q1 - 2017 Q4 2018 Q1 - 2018 Q4 2019 Q1 - 2019 Q4


C. Transmission outages initiated by failed protection system equipment

From TADS data, there was a significant decrease in 2019 in the outage rate per element initiated by failed protection system equipment for 345 kV transmission circuits. The outage rates per element initiated by failed protection system equipment for 138 kV circuits and 345 kV transformers remained flat.

Figure 84 – Outage Rates Caused by Failed Protection Equipment

0

0.005

0.01

0.015

0.02

0.025

0.03

345 kV Circuits 138 kV Circuits 345 kV Xfmrs

Outage Rate per Element Initiated by Failed Protection Equipment

2015 2016 2017 2018 2019


IX. Physical and Cyber Security

Summary

Intentional damage, destruction, or disruption to facilities can cause localized to extensive Interconnection-wide BPS disruption—potentially for an extended period. Exploitation of cybersecurity vulnerabilities can potentially result in loss of control or damage to BPS-related voice communications, data, monitoring, protection and control systems, or tools. Successful exploitation can damage equipment, causing loss of situational awareness and—in extreme cases—can result in degradation of reliable operations to the BPS, including loss of load.

Texas RE monitors infrastructure protection issues as part of its situational awareness effort. These issues primarily consist of substation intrusions and copper theft that are typically dealt with by local law enforcement. However, if the issue involves critical infrastructure, cyber intrusions, or possible sabotage, then it is elevated to NERC and the Department of Energy under the reporting requirements in NERC Reliability Standard EOP-004. (Note that almost all cyber-related incidents are not attributed to attacks on the BPS and are reported outside of these reports, provided through CIP-008 incident reporting or direct communications to the E-ISAC).

2019 highlights from the analysis of protection system misoperations include:

Substation intrusions reported through ERCOT channels are low risk, with vandalism and copper theft being the most common.


Many routine occurrences of theft or vandalism involving BPS facilities continue to appear to go unreported to ERCOT as the Reliability Coordinator, given that some large population centers have few events.

Historical Data and Trends

Since January 2014, substation intrusions and copper theft have ranged from three to 12 in any one month, averaging four per month. For the purposes of this figure, physical/cyber security issues include bomb threats, sabotage, and cyber security issues.


Figure 85 – ERCOT Trend in Substation Intrusions/Copper Theft/Cyber Security Issues

0

5

10

15

20

25

30

35

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

2016Q1

2016Q2

2016Q3

2016Q4

2017Q1

2017Q2

2017Q3

2017Q4

2018Q1

2018Q2

2018Q3

2018Q4

2019Q1

2019Q2

2019Q3

2019Q4

Physical/Cyber Security Intrusion/Copper Theft

Infrastructure Protection


Figure 86 – Locations for Reported Intrusions/Copper Theft/Cyber Security Issues

Appendix A – Transmission Availability Analysis

TADS Element and Outage Data

A summary of the aggregated ERCOT TADS elements, circuit miles, and outage data is shown in the following tables.

Year Circuits (300-399 kV) Circuit Miles (300-399 kV) Transformers (300-399 kV)

2010 279 9,245.7

2011 299 9,490.3

2012 302 9,653.1

2013 359 12,890.7

2014 394 13,976,4

2015 394 14,615.8 206

2016 422 14,811.1 213

2017 431 15,049.6 217

2018 464 15,540.2 221

2019 480 16,382.2 223

Table A.1 – 2010-2019 End of Year Circuit Data

Automatic

Non-Automatic Operational

Outage Information

Count Duration (hours)

Count Duration (hours)

2010 195 1,090.0 24 1,167.9

2011 276 1,908.6 66 7,096.1

2012 226 682.6 45 4,264.6

2013 197 1,935.6 32 7,877.4

2014 276 2,917.3 69 6,001.3

20152 477 10,806.9 44 2,821.8

2016 436 6,446.1 43 3,645.6

2017 438 18,657.8 18 345.9

2018 334 22,619.0 27 3,472.9

2019 523 7398.8 82 14,591.1

5-Yr Average 442 13,185.7 43 4,975.5

Table A.2 – 2010-2019 345 kV Circuit and Transformer Outage Data

Automatic Outage Data

For 2015-2019 for 345 kV circuits, Failed AC Circuit Equipment represented 12 percent of sustained outage cause and 59 percent of sustained outage duration.

2 Outage count and duration for 2015-2019 includes 345 kV transformers which began reporting in 2015


Figure A.1 – 345 kV Circuit Automatic Outages by Month

Figure A.2 – Multi-Year Comparison of TADS Outages and Duration by Month (> 200 kV)

0

10

20

30

40

50

60

70

80

90


2015 2016 2017 2018 2019

345 kV Circuit Automatic Outage Count by Month

0

500

1,000

1,500

2,000

2,500

3,000

3,500


2015 2016 2017 2018 2019

345 kV Circuit Automatic Outage Duration by Month (Hours)


Figure A.3 – 345 kV Circuit Momentary Outage Count by Cause

Figure A.4 – 345 kV Circuit Sustained Outage Count by Cause

0

20

40

60

80

100

120

1402015 2016 2017 2018 2019

345 kV Circuit Momentary Outage Count by Cause

0

10

20

30

40

50

602015 2016 2017 2018 2019345 kV Circuit Sustained Outage Count by Cause


Figure A.5 – 345 kV Circuit Sustained Outage Duration (Hours) by Cause

The following table shows the 345 kV circuit sustained outage data by average duration for 2015-2019 combined, 138 kV circuit sustained outage data for 2015-2019 combined, and the 345 kV transformer sustained outage data for 2015-2019 combined.

345 kV Circuits Sustained Cause Code

Number of Sustained Outages

Average Outage

Duration (Hours)

Median Outage Duration (Hours)

Failed AC Circuit Equipment 108 114.5 17.8

Weather, Excluding Lightning 73 70.3 4.4

Failed AC Substation Equipment 80 23.6 4.3

Foreign Interference 22 7.2 0.2

Lightning 50 10.0 0.2

Power System Condition 26 8.7 3.9

Contamination 29 8.4 2.2

Fire 7 10.9 9.5

Unknown 53 5.4 0.5

Failed Protection System Equipment 22 6.8 2.1

Other 50 8.2 0.1

Human Error 37 2.5 1.5

Vegetation 2 0.03 0.03

0

1,000

2,000

3,000

4,000

5,000

6,0002015 2016 2017 2018 2019

345 kV Circuit Sustained Outage Duration (Hours) by Cause


2015-2019 345 kV Circuits 563 38.4 3.6

138 kV Circuits Sustained Cause Code


Average Outage

Duration (Hours)


Fire 16 71.1 8.3



Environmental 17 14.1 13.5

Vegetation 54 12.9 5.8


Lightning 153 8.4 0.1





Other 114 3.5 0.7

Unknown 129 3.4 0.4


2015-2019 138 kV Circuits 1824 21.0 2.6

345 kV Transformers Sustained Cause Code


Average Outage

Duration (Hours)








Other 12 15.2 10.6



Lightning 8 5.4 5.6

Environmental 1 3.5 3.5

Unknown 2 2.0 2.0

Fire 1 0.1 0.1

2015-2019 345 kV Transformers 154 260.0 10.8

Table A.3 – Sustained Outage Data by Average Outage Duration


Figure A.6 – 138 kV Circuit Sustained Outage Counts by Month

0

10

20

30

40

50

60

70

80


138 kV Circuit Sustained Outage Count by Month

2015 2016 2017 2018 2019

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000


138 kV Circuit Sustained Outage Duration (Hours) by Month

2015 2016 2017 2018 2019


Figure A.7 – 138 kV Circuit Sustained Outage Duration (Hours) by Month

Figure A.8 – 138 kV Circuit Sustained Outage Count by Cause

Figure A.9 – 138 kV Circuit Sustained Outage Duration by Cause

0

10

20

30

40

50

60

70

80

90

100138 kV Circuit Sustained Outage Count by Cause

2015 2016 2017 2018 2019

0

2,000

4,000

6,000

8,000

10,000

12,000138 kV Circuit Sustained Outage Duration (Hours) by Cause

2015 2016 2017 2018 2019


Appendix B – Generation Performance Analysis

GADS Element and Outage Data

A number of generators reporting ERCOT GADS and GADS-Wind data is shown in the following tables.

Units Reporting 2015 2016 2017 2018 2019

Total 412 409 415 407 402

Coal/Lignite 29 29 29 26 21

Gas 58 51 48 45 43

Nuclear 4 4 4 4 4

Gas Turbine/Jet Engine 84 89 85 87 90

Hydro 8 8 8 8 8

Fluidized Bed 5 6 6 5 5

Combined Cycle (Block) 28 27 18 18 18

Combined Cycle GT 140 140 151 149 149

Combined Cycle ST 58 58 62 61 61

Other 1 1 3 3 3

Wind (>200 MW) 47 55

Wind (100<MW<200) 35 73

Wind (< 100 MW) 58 80

Number of Wind Turbines 9,466 14,132

Table B.1 – 2015-2019 GADS and GADS-Wind Units Reporting

The following figure uses GADS data to plot fleet capacity by age and fuel type. It shows two important characteristics of the fossil fuel fleet: (1) there is an age bubble around 35–39 years driven by coal and some gas units; and (2) there is a significant age bubble around 12–17 years comprised almost exclusively of combined cycle units. As the coal fleet and older gas units continue to age, it is expected that the pace of retirement of these units will increase in favor of more efficient generation.


Figure B.1 – GADS Fossil Generation in ERCOT by Age and Fuel Type

Figure B.2 – 2019 GADS Metrics by Age

0

5

10

15

20

25

30

35

40

45

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75

Age of ERCOT Units Reporting GADS# Units

Age

49.82%

7.14%

52.51%

5.63%

60.73%

3.85%

60.13%

5.31%

28.09%

14.84%

7.14%

29.51%

0%

10%

20%

30%

40%

50%

60%

70%

Net Capacity Factor (NCF) Equivalent Forced Outage Rate (EFOR)

0-9 10-19 20-29 30-39 40-49 >50

2019 Metrics by Age


Figure B.3 – 2015-2019 GADS EFOR by Age

GADS provides various metrics to compare unit performance. Two of these methods are unweighted (time-based) and weighted (based on unit MW size). Table 4 in Section II showed the fleet level unweighted metrics. The following table shows the same metrics by fuel type.


Coal/Lignite Gas Jet Engine CC Block CC GT CC ST


# Units Reporting 21 43 90 18 149 61

Total Unit-Months 245 516 1066 216 1788 732

Net Capacity Factor (NCF)

60.3% 10.0% 13.7% 52.2% 55.5% 45.1%

Service Factor (SF) 80.7% 25.3% 18.0% 58.2% 67.4% 68.6%

Equivalent Availability Factor (EAF)

85.1% 73.5% 89.4% 87.8% 85.7% 86.8%

Scheduled Outage Factor (SOF)

8.0% 17.7% 6.3% 9.3% 9.4% 9.1%

Forced Outage Factor (FOF)

4.3% 6.1% 3.4% 2.6% 4.3% 3.7%

EFOR 7.7% 24.9% 16.3% 4.5% 6.1% 7.2%

EFORd 5.1% 16.0% 6.8% 3.4% 5.2% 4.8%

Table B.2 – ERCOT Generation Performance Metrics by Fuel Type for 2019

0%

5%

10%

15%

20%

25%

30%

35%

0-9 10-19 20-29 30-39 40-49 >50

2016 2017 2018 2019

MW Weighted EFOR by Age


Figure B.4 – 2015-2019 Count of Immediate Forced Outage Events by Month

0

50

100

150

200

250

300


Forced Outage Events 2015 Forced Outage Events 2016 Forced Outage Events 2017

Forced Outage Events 2018 Forced Outage Events 2019


Figure B.5 – 2015-2019 Count of Immediate Forced De-rate Events by Month

0

50

100

150

200

250

300

350

400


Forced Derate Events 2015 Forced Derate Events 2016 Forced Derate Events 2017

Forced Derate Events 2018 Forced Derate Events 2019


Appendix C – Frequency Control Performance Analysis

Observations

Control Performance Standard 1 (CPS1): 174.8 for calendar year 2019 versus 175.7 for calendar year 2018

Balancing Authority ACE Limit (BAAL) exceedances: 16 clock-minutes for calendar year 2019 versus 16 clock-minutes for 2018

Average recovery time from generation loss events: 5.1 minutes for calendar year 2019 versus 6.2 minutes for calendar year 2018

Historical Data and Trends

A. CPS1 Performance

NERC Reliability Standard BAL-001-2 requires each BA to operate such that the 12-month rolling average of the clock-minute Area Control Error (ACE) divided by the clock-minute average BA Frequency Bias times the corresponding clock-minute average frequency error is less than a specific limit. This is referred to as Control Performance Standard 1 (CPS1). The NERC CPS1 Standard requires rolling 12-month average performance of at least 100 percent. The following figure shows the ERCOT region CPS1 trend since January 2010. For 2019, the annualized CPS1 score was 174.8.

Figure C.1 – CPS1 Average January 2010 to December 2019

100

110

120

130

140

150

160

170

180

190

200

100

110

120

130

140

150

160

170

180

190

200

CP

S1

Av

era

ge

CPS1 12-Month Rolling Avg Poly. (12-Month Rolling Avg)


Figure C.2 – ERCOT CPS1 Annual Trend since January 2008

As of October 2015, all generation units are required to set their governor deadbands at 0.017 Hz per Regional Standard BAL-001-TRE.

Figure C.3 shows bell curves of the ERCOT frequency profile, comparing 2011 through 2019. The shape of the bell curve in 2019 was virtually identical to 2018.

The blue dashed lines on the figure represent the Epsilon-1 (ε1) value of 0.030 Hz which is used for calculation of the CPS-1 score. The red dashed lines represent governor deadband settings of 0.017 Hz. The purple dashed lines represent three times the ε1 value which is used for BAAL exceedances per NERC Standard BAL-001-2.

127.5126.4

141.0

150.8

148.9

162.1

166.2 163.3

174.3

176.6 174.9

175.7

174.8

100

110

120

130

140

150

160

170

180

January2008

January2009

January2010

January2011

January2012

January2013

January2014

January2015

January2016

January2017

January2018

January2019

January2020

CPS1 Average


Figure C.3 – Frequency Profile Comparison

The following figure shows the 2019 CPS1 scores by operating hour compared to previous years.

The CPS1 score by operating hour continues to indicate possible issues for hour-ending (HE) 06:00 and HE 07:00. These issues are related to the load ramps during these hours and procedures used by generation resource entities during unit startup and shutdown.

The daily RMS1 figure shows the average root-mean-square of the frequency error based on one-minute frequency data. The long-term trend continues to show excellent control of frequency error. The red dashed line on the figure shows the 17 mHz governor deadband required by BAL-001-TRE in relation to the daily RMS1.

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

59.9

59.9

05

59

.91

59.9

15

59

.92

59.9

25

59

.93

59.9

35

59

.94

59.9

45

59

.95

59.9

55

59

.96

59.9

65

59

.97

59.9

75

59

.98

59.9

85

59

.99

59.9

95 60

60.0

05

60

.01

60.0

15

60

.02

60.0

25

60

.03

60.0

35

60

.04

60.0

45

60

.05

60.0

55

60

.06

60.0

65

60

.07

60.0

75

60

.08

60.0

85

60

.09

60.0

95

60.1

2011 2012 2013 2014 2015 2016 2017 2018 2019

ERCOT Frequency Profile

0.017 Hz deadband


Figure C.4 – CPS1 Score by Operating Hour for 2015 through 2019

Figure C.5 – Daily RMS1 for 2015 through 2019

B. Time Error Correction Performance

In 2019, there were no manual Time Error Corrections. In December 2016, ERCOT added an Area Control Error (ACE) Integral term to the Generation-To-Be-Dispatched (GTBD) calculation. This term corrected longer-term errors in generation basepoint deviation rather

140

145

150

155

160

165

170

175

180

185

190

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

CPS1 by Operating Hour

2015 2016 2017 2018 2019

0

5

10

15

20

25

Jan

-15

Feb

-15

Mar

-15

Ap

r-15

May

-15

Jun

-15

Jul-

15A

ug-

15

Sep

-15

Oct

-15

No

v-1

5D

ec-

15

Jan

-16

Feb

-16

Mar

-16

Ap

r-16

May

-16

Jun

-16

Jul-

16A

ug-

16

Sep

-16

Oct

-16

No

v-1

6D

ec-

16

Jan

-17

Feb

-17

Mar

-17

Ap

r-17

May

-17

Jun

-17

Jul-

17A

ug-

17

Sep

-17

Oct

-17

No

v-1

7D

ec-

17

Jan

-18

Feb

-18

Mar

-18

Ap

r-18

May

-18

Jun

-18

Jul-

18A

ug-

18

Sep

-18

Oct

-18

No

v-1

8D

ec-

18

Jan

-19

Feb

-19

Mar

-19

Ap

r-19

May

-19

Jun

-19

Jul-

19A

ug-

19

Sep

-19

Oct

-19

No

v-1

9D

ec-

19

RMS-1 (mHz)


than depending on regulation. Since implementation of the ACE Integral into the GTBD, ERCOT is controlling frequency to zero average time error.

C. Balancing Authority ACE Limit (BAAL) Performance

The Frequency Trigger Limits (FTLs) are defined as ranges for the Balancing Authority ACE Limit high and low values per NERC Standard BAL-001-2 which became enforceable in July 2016. The FTL-Low value is calculated as 60 Hz – 3 x Epsilon-1 (ε1) value of 0.030 Hz, or 59.910 Hz for the ERCOT region. The FTL-High value is calculated as 60 Hz + 3 x Epsilon-1 (ε1) value, or 60.090 Hz for the ERCOT region.

The following table shows the total one-minute intervals where frequency was above the FTL-High alarm level or below the FTL-Low alarm level.

All low BAAL exceedance minutes in 2019 were associated with large generation unit trips.

High/Low Frequency

2015 Total Minutes

2016 Total Minutes

2017 Total Minutes

2018 Total Minutes

2019 Total Minutes

five-year Avg

Low (<59.91 Hz)

13 26 18 17 16 18

High (>60.09 Hz)

1 0 0 0 0 < 1

Table C.1 – Frequency Trigger Limit Performance