2013 Assessment of Reliability Performance Services/2015 Texas … · The goals of the 2015...

2015 Assessment of Reliability Performance of the Electric Reliability Council of Texas, Inc. (ERCOT) Region

By Texas Reliability Entity, Inc. April 2016

2015 ASSESSMENT OF RELIABILITY PERFORMANCE PAGE 2 OF 117 APRIL 2016

Table of Contents

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

Introduction ................................................................................................................................ 7

2015 At A Glance ....................................................................................................................... 8

Summary of Key Findings and Observations ............................................................................10

Recommended Focus Areas for 2016 .......................................................................................12

I. Disturbances and Events ...................................................................................................15

II. Transmission .....................................................................................................................22

III. Generation .........................................................................................................................38

IV. Load and Demand Response ............................................................................................52

V. Frequency Control and Primary Frequency Response .......................................................56

VI. Protection System Performance ........................................................................................73

VII. Infrastructure Protection .....................................................................................................79

VIII. Emerging Reliability Issues ................................................................................................80

Appendix A – References .........................................................................................................85

Appendix B – Disturbance Events Analysis ...............................................................................86

Appendix C – Transmission Availability Analysis .......................................................................94

Appendix D – Generation Availability Analysis ........................................................................ 102

Appendix E – Demand Response Historical Data ................................................................... 106

Appendix F – Frequency Control Performance Analysis ......................................................... 107

Appendix G – Protection System Misoperations Analysis ....................................................... 112

Table of Figures and Tables

Figure 1 – ERCOT Region Map ................................................................................................. 7

Figure 2 – Annual Energy and Peak Demand ............................................................................ 9

Figure 3 – Wind Percentage of Total Energy .............................................................................. 9

Figure 4 – 2015 Energy by Fuel Type .......................................................................................10

Table 1 – Summary of Events Analysis .....................................................................................16

Figure 5 – Events Reported by Quarter .....................................................................................17

Figure 6 – 2011-2015 Event Cause Summary ..........................................................................17

Figure 7 – OE-417 Reports of Lost Load ...................................................................................18

Table 2 – Major Causes of Generator Trips > 450 MW .............................................................19

Table 3 – 2015 Events with Multiple Generator Trips ................................................................20

Table 4 – 2015 Momentary and Sustained Outages .................................................................23

Figure 8 – 2008-2015 345 kV Automatic Outage Metrics ..........................................................23

Table 5 – TADS Circuit and Automatic Outage Historical Data for ERCOT Region ...................24

Figure 9 – 2015 345 kV Momentary Outage Cause ..................................................................25

Figure 10 – 2015 Automatic Outages by Month ........................................................................25

Figure 11 – 2015 Automatic Outage Duration by Month ............................................................26

Figure 12 – 2015 345 kV Sustained Outage Cause ..................................................................26

Figure 13 – 2015 138 kV Sustained Outage Cause ..................................................................27

Figure 14 – 2015 345 kV Sustained Outage Duration ...............................................................27


Figure 15 – 2015 138 kV Sustained Outage Duration ...............................................................28

Figure 16 – 2015 345 kV Momentary and Sustained Outages by Cause per Month ..................28

Figure 17 – 2015 345 kV Automatic Outage Data by Duration ..................................................29

Figure 18 – 2015 138 kV Sustained Outage Data by Duration ..................................................29

Figure 19 – 2015 345 kV Outage Rates by Entity ......................................................................30

Figure 20 – 2015 345 kV Outage Cause and Duration by Outage Mode ...................................30

Figure 21 – 2015 138 kV Outage Cause and Duration by Outage Mode ...................................31

Table 6 – 2015 Top Constraints ................................................................................................33

Figure 22 – Quarterly Trend in Transmission Line Constraints ..................................................33

Figure 23 – Generation Bus Voltage Control Chart for August 2015 .........................................34

Figure 24 – 345 kV Bus Voltage Chart for Important State Estimator Buses 2015 ....................35

Figure 25 – SPS and Operating Procedure Trends ...................................................................36

Figure 26 – Special Protection System Operation Trend ...........................................................37

Figure 27 – 2015 Generation Nameplate Capacity ....................................................................39

Figure 28 – 2015 Energy by Fuel Type .....................................................................................40

Figure 29 – Energy by Fuel Type Trend ....................................................................................40

Table 7 – ERCOT Generation Performance Metrics January through December 2015 .............41

Figure 30 – GADS Generation Performance Metrics by Fuel Type and Year ............................42

Table 8 – Average GADS Generation Unit Outage Hours .........................................................42

Table 9 – Generator Immediate De-rate and Forced Outage Data (Jan. – Dec. 2015) ..............43

Table 10 – 2015 Major Category Cause of Immediate De-rate Events from GADS ...................43

Figure 31 – Summer 2015 Generation Scheduled and Forced Outages ...................................44

Figure 32 – Winter 2014-2015 Generation Scheduled and Forced Outages .............................44

Figure 33 – 2015 Immediate De-rate and Forced Outage Events by Fuel Type ........................45

Figure 34 – 2015 Immediate De-rate/Forced Outage Duration (Hours) by Fuel Type ................46

Figure 35 – 2015 Generator Immediate De-rate Events/Duration by Cause ..............................46

Table 11 – 2015 Major Category Cause of Immediate Forced Outage Events from GADS .......47

Figure 36 – 2015 Generator Immediate Forced Outage Events/Duration by Cause ..................47

Figure 37 – 2008-2015 Wind Generation MWh .........................................................................48

Figure 38 – 2008-2015 Wind Generation as a Percentage of ERCOT Total Energy .................49

Figure 39 – 2008-2015 Wind Generation as Percentage of ERCOT Total Energy by Month .....49

Figure 40 – 2015 Wind Capacity Factors for Summer Peak Hours ............................................50

Figure 41 – August 2015 Coastal/Non-Coastal Wind and Solar Generation Capacity Factors ..51

Figure 42 – Annual Energy and Peak Demand .........................................................................53

Figure 43 –Energy by Area .......................................................................................................53

Figure 44 – Demand Response Availability ...............................................................................54

Figure 45 – ERCOT Region Demand Response Availability and Deployments .........................55

Table 12 – Demand Response Deployments by Non-Opt-In Entities for 2015 ..........................55

Table 13 – Demand Response Deployments in 2015 ...............................................................55

Figure 46 – CPS1 Average January 2008 to December 2015 ...................................................57

Figure 47 – ERCOT CPS1 Annual Trend since January 2008 ..................................................57

Figure 48 – Frequency Profile Comparison ...............................................................................58

Figure 49 – CPS1 Score by Hour for 2013 through 2015 ..........................................................59


Figure 50 – Time Error Corrections 2012-2015 .........................................................................60

Table 14 – 2015 Time Error Correction Summary .....................................................................60

Figure 51 – Average Net Load Change vs. Time Error by Hour of Day .....................................61

Figure 52 – Average Short-Term Load Forecast Error vs. Time Error by Hour of Day ...............62

Table 15 – Regulation Exhaustion Rates ..................................................................................63

Figure 53 – 2015 Regulation Exhaustion Rates by Operating Hour ...........................................63

Figure 54 – 2015 Net Regulation Deployments by Operating Hour ...........................................64

Figure 55 – 2015 Net Regulation Deployment vs. Basepoint Deviation by Operating Hour .......65

Figure 56 – 2015 Time Error vs. Basepoint Deviation by Operating Hour..................................66

Figure 57 – 2015 Basepoint Deviation vs. Net Regulation 7-day Moving Average ....................67

Figure 58 – CPS2 Monthly Average – June 2008 to December 2015 ........................................68

Table 16 – Frequency Trigger Limit Performance .....................................................................69

Figure 59 – Annual Primary Frequency Response Trend for ERCOT Region ...........................69

Figure 60 – Histogram of ERCOT Frequency Response 2012-2015 .........................................70

Figure 61 – Primary Frequency Response Trend for ERCOT Region .......................................70

Figure 62 – Primary Frequency Response versus Physical Response Capability .....................71

Figure 63 – ACE Recovery Time versus Regulation Deployed ..................................................72

Table 17 – Failures of PDCWG Metrics by Unit Type ................................................................72

Table 18 – Protection System Misoperation Data ......................................................................74

Figure 64 – Protection System Misoperation Trends .................................................................74

Figure 65 – Protection System Misoperations by Cause 2011-2015 .........................................75

Figure 66 – Protection System Misoperation Rates by Entity 2012-2015 ..................................76

Figure 67 – Protection System Misoperation Rates by Region 2012 Q4-2015 Q3 .....................76

Figure 68 – Protection System Misoperations Trend Caused by Human Performance..............78

Figure 69 – ERCOT Trend in Substation Intrusions/Copper Theft/Cyber Security Issues ..........79

Figure 70 – Inertia versus Net Load ..........................................................................................80

Figure 71 – Inertia versus Percentage of Load served by IRRs .................................................81

Table 19 – Maximum and Minimum One-Hour Load and Wind Ramp for 2015 .........................81

Figure 72 – Average Net Load Ramp by Season and Operating Hour for 2015 ........................82

Figure 73 – ERCOT FIDVR Event, Summer 2015 .....................................................................83

Figure 74 – ERCOT FIDVR Load Impact, Summer 2015 ..........................................................84

Figure B.1 – 2011-2015 ERCOT Events by Category ...............................................................86

Figure B.2 – 2011-2015 ERCOT Events by Cause ...................................................................86

Figure B.3 – DCS Events by Year .............................................................................................87

Figure B.4 – DCS and EEA Events by Year ..............................................................................88

Table B.1 – EEA Event Magnitude and Duration .......................................................................88

Table B.2 – Historical Events with Multiple Generator Trips ......................................................90

Figure B.5 – ERCOT Telemetry System Availability ..................................................................91

Figure B.6 – Bus Summation Telemetry Accuracy ....................................................................92

Figure B.7 – State Estimator vs. Telemetry on Major Transmission Elements ...........................92

Figure B.8 – State Estimator vs. Telemetry on Congested Transmission Elements ..................93

Figure B.9 – State Estimator vs. Telemetry on Top 20 Critical Buses ........................................93

Table C.1 – 2008-2015 End of Year Circuit Data ......................................................................94


Table C.2 – 2010-2015 345 kV Automatic and Non-Automatic Outage Data .............................94

Figure C.1 – 345 kV Automatic Outages by Month ....................................................................95

Figure C.2 – Multi-Year Comparison of TADS Outages and Duration by Month (> 200 kV) ......95

Figure C.3 – 345 kV Momentary Outages by Cause .................................................................96

Figure C.4 – 345 kV Sustained Outages by Cause ...................................................................96

Figure C.5 – 345 kV Sustained Outage Duration (hours) by Cause...........................................97

Figure C.6 – 345 kV Average AC Circuit Momentary Automatic Outage Mode..........................98

Figure C.7 – 345 kV Average AC Circuit Sustained Automatic Outage Mode ...........................98

Figure C.8 – 345 kV Momentary Outage Modes Comparison ...................................................99

Figure C.9 – 345 kV Sustained Outage Modes Comparison .....................................................99

Figure C.10 – 2010-2015 345 kV Common Mode/Dependent Mode Outages by Cause ......... 100

Figure C.11 – 2010-2015 345 kV Common Mode/Dependent Mode Outages by Duration ...... 100

Figure C.12 – 2010-2015 AC Circuit Outage Mode (> 200 kV) ................................................ 101

Figure D.1 – 2015 Net and Gross Capacity Factors for the ERCOT Generation Fleet ............. 102

Figure D.2 – 2015 Scheduled/Forced Outage Factors for the ERCOT Generators ................. 103

Figure D.3 – 2013-2015 Lost MWH from Forced Outages ...................................................... 104

Figure D.4 – 2012-2015 Cumulative Events by Operating Hour - Summer .............................. 104

Figure D.5 – 2012-2015 Cumulative Events by Operating Hour - Winter ................................. 105

Table E.1 – Demand Response Deployments since 1/1/2010 ................................................. 106

Figure F.1 – Comparison of 2011-2015 Regulation-Up Exhaustion Rates by Month ............... 107

Figure F.2 – Comparison of 2011-2015 Regulation-Down Exhaustion Rates by Month ........... 108

Figure F.3 – Comparison of 2012-2015 Reg-Up Exhaustion Rates by Operating Hour ........... 108

Figure F.4 – Comparison of 2012-2015 Reg-Down Exhaustion Rates by Operating Hour ....... 109

Figure F.5 – Comparison of 2012-2015 Average Regulation Deployed by Operating Hour ..... 109

Figure F.6 – Non-Spin Reserve Service Deployment History .................................................. 110

Figure F.6 – Hourly Ancillary Service Supply Responsibility Failures ...................................... 111

Figure F.7 – Responsive Reserve Service Deployment History .............................................. 111

Figure G.1 – Protection System Misoperation Data for 2011-2015 by Voltage ........................ 112

Figure G.2 – Protection System Misoperation Data for 2011-2015 by Category ...................... 112

Figure G.3 – Protection System Misoperation Data for 2011-2015 by Relay System Type ..... 113

Figure G.4 – Protection System Misoperation Data for 2011-2015 by Equipment Protected ... 113

Figure G.5 – Protection System Misoperation Data for 2011-2015 by Cause .......................... 114

Figure G.6 – Protection System Misoperations 2011-2015 Relay Failures by System Type .... 114

Figure G.7 – Protection System Misoperations by Year by Cause 2011-2015 ........................ 115

Figure G.8 – Protection System Misoperations by Misoperation Type 2011-2015 ................... 115

Figure G.9 – Protection System Misoperation Data by Cause and Element ............................ 116

Figure G.10 – Protection System Misoperation Data by Cause 2011-2015 ............................. 117

Figure G.11 – Protection System Misoperations by Category 2011-2015................................ 117


Executive Summary

The goals of the 2015 Assessment of Reliability Performance report for the Electric Reliability Council of Texas (ERCOT) bulk power system (BPS) are to illuminate the historical and overall BPS reliability picture, to help identify risk areas, and to prioritize and create actionable results for reliability improvement.

This report represents an ongoing effort by Texas Reliability Entity, Inc. (Texas RE) to provide a view of risks to reliability based on historic performance. 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 reporting arising from multiple sources, thereby providing efficient data and information transparency. The key findings and observations can serve as inputs to process improvements, event analysis, reliability assessments, and critical infrastructure protection.

For 2015, the overall BPS reliability performed within the defined acceptable performance metrics. The following are the key observations were made for 2015:

Transmission availability remained high; outages per circuit and outages per 100 miles of line increased in 2015 primarily due to a significant increase in lightning-related outages.

ERCOT-wide generation performance metrics continue to exceed NERC-wide averages.

The portion of total energy supplied by natural gas continued to trend upward. The portion of energy supplied by coal continued to be displaced by natural gas and renewable generation.

Frequency control and primary frequency response metrics continue to be maintained at high levels.

Protection System Misoperation rates remained stable. However, misoperations from incorrect settings remains a key issue to be addressed.

Major focus areas for 2016 are:

Overall growth and continued integration of renewable generation;

Emerging system issues; o System inertia changes with resource mix o System short-circuit strength o Net demand ramping variability o Reactive performance

Protection system misoperations caused by incorrect settings;

Critical Infrastructure Protection;

Ancillary service changes;

Voltage control; and

SCADA, EMS, and telemetry performance.


Introduction

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

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

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

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 covers over 24 million Texas customers—representing 90% 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 550 generation units. The ERCOT region will have more than 77,000 MW of expected generation capacity for summer peak demand. Installed wind generation capacity totals more than 16,000 MW. In August 2015, the grid had an all-time peak of 69,877 MW. ERCOT region 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

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)

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

Protection System Misoperation Reports (NERC Reliability Standards and ERCOT Operating Guides)

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, events analysis, situational awareness, tracking reliability indicators, real-time performance monitoring, and planning observations. Texas RE developed the 2015 Assessment of Reliability Performance report to provide a high-level overview of the data collected in the ERCOT region. This report is intended to provide:

2015 data at a high level;

Associated historical data;

An analysis of the 2015 and other historical data as an indicator of the current state of the ERCOT region; and

Observations that help connect the state of the region today to the future; and

This report provides Texas RE’s assessment of reliability data and historical trends in eight focus areas:

1. Disturbances and Events

2. Transmission

3. Generation

4. Load and Demand Response

5. Frequency Control and Primary Frequency Response

6. Protection System Performance

7. Infrastructure Protection

8. Emerging Reliability Issues

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.

2015 At A Glance

Record peak demand: 69,877 MW on August 10, 2015 Record wind generation: 13,883 MW on December 20, 2015 at 11:07 am Record wind penetration: 44.7% of total energy on December 20, 2015 at 3:05 am Control Performance Standard 1 (CPS1): 174.3 for calendar year 2015 vs 163.3 for

calendar year 2014 Frequency Response: 720.43 MW/0.1 Hz vs. NERC obligation of 471 MW/0.1 Hz Protection system misoperation rate: 7.2% for 2015 vs. 8.7% for 2014 TADS 345 kV automatic outage rate per 100 miles: 2.99 for 2015 vs. 1.97 for 2014 GADS EFORd: 5.5% for 2015 vs. 5.4 % for 2014


Figure 2 – Annual Energy and Peak Demand

Figure 3 – Wind Percentage of Total Energy

50,000

55,000

60,000

65,000

70,000

75,000

250,000

260,000

270,000

280,000

290,000

300,000

310,000

320,000

330,000

340,000

350,000

2008 2009 2010 2011 2012 2013 2014 2015

Annual Energy Peak DemandGWH MW

0%

2%

4%

6%

8%

10%

12%

14%

2008 2009 2010 2011 2012 2013 2014 2015

Wind % of Total Energy


Figure 4 – 2015 Energy by Fuel Type

Summary of Key Findings and Observations

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

1. Frequency control metrics continue to be maintained at high levels.

The frequency Control Performance Standard (CPS) metrics for the ERCOT region continue to be among the highest of all the NERC regions. Minor issues have been noted with regulation and generation basepoint deviations which are being addressed by ERCOT region working groups.

2. Primary frequency response metrics continue to be maintained at high levels.

The implementation phase of Regional Standard BAL-001-TRE is essentially complete. In 2016, the focus will shift to review of compliance with the standard.

3. Transmission outage rates increased in 2015 but overall performance remains high. The outage rate per circuit and outage rate per 100 miles of line increased in 2015 primarily due to a significant increase in lightning-related outages. For the 345 kV system,

Natural Gas 48%

Coal28%

Nuclear11%

Wind12%

Other1%

2015 Energy (GWH) by Fuel Type


failed substation equipment and failed transmission circuit equipment dominated the sustained outages, accounting for 77% of the outage duration. For the 345 kV system, approximately 21% of transmission outages in 2015 were due to unknown causes. Identifying the causes of unknown outages may help entities detect and address system issues that could have a larger impact on reliability Approximately 16% of 345 kV transmission outage events (43% of outage duration) involve multiple elements. These events, which may go beyond recognized planning criteria, represent a tangible threat to reliability. Mandatory TADS reporting of sustained outage data for circuits operated at less than 200 kV began in 2015. With only one year of data, it is difficult to determine meaningful trends or performance comparisons. Initial data for the 138 kV system shows failed substation equipment and failed transmission circuit equipment dominate the sustained outages, accounting for 81% of the outage duration.

Trending of System Operating Limit (SOL) exceedances should be a high priority to determine whether it could provide insights to reliability trends for the ERCOT region. As Competitive Renewable Energy Zone (CREZ) lines become fully subscribed, the probability of SOL exceedances may increase.

4. Generation unit performance compares well with NERC-wide generation fleet averages.

ERCOT region generators compare well with the NERC-wide GADS performance metrics.

Analysis of GADS event data shows increased probability of de-rates, startup failures, and forced outages during peak hours and ramp periods.

In 2015 there were seven occurrences of multiple generator trips at the same site or sympathetic trips at multiple sites. These events create frequency deviations on the system and temporary generation shortfalls which must be mitigated by the use of reserves. Further analysis is needed to determine common failure modes and share lessons learned to reduce these types of events.

The relatively mild winter in 2015 did not result in any major incidents of generator trips or de-rates due to cold weather.

5. Protection System Misoperation rates are showing a stable trend. However, misoperations from incorrect settings warrant further assessment. Incorrect settings, relay failures or malfunctions, and communication failures account for 71% of the reported Protection System Misoperations. Incorrect settings, logic, and design errors accounted for 42% of misoperations in 2015. Many incorrect settings were due to engineering errors in general protection applications or modeling, which are common to all three relay technologies (microprocessor, electromechanical, and solid state), meaning that the misoperation would have occurred regardless of the protective


relay technology used. The percentage of misoperations caused by “human performance” (i.e., settings errors, design errors, wiring errors, etc.) is approximately 50% of the total misoperations.

Recommended Focus Areas for 2016

The following are the recommended focus areas for the upcoming year:

1. Renewable Generation Growth Wind generation produced a total of 40,786 GWH in 2015, an increase of 13% from 2014. Wind generation, as a percentage of total energy produced, increased to 11.7% in 2015, up from 10.6% in 2014. In 2015, instantaneous output of wind generation reached a record of 13,883 MW on December 20, 2015, and wind generation served a record of 44.7% of system demand on December 20, 2015. 2015 also marked the first year that wind generation supplied a greater percentage of total energy than nuclear.

As of the summer of 2015, over 228 MW of solar capacity was generating into the ERCOT system. Instantaneous solar output reached 197 MW on December 3, 2015. Current projections indicate that solar generation will exceed 1,400 MW over the next two years. Texas RE will continue to monitor the solar generation growth as well as its use in short-term and long-term assessments of resource adequacy.

Current generation projects under study continue to indicate a significant expansion of wind resources in the Panhandle. Voltage stability limits and system short circuit strength is beginning to constrain wind power delivery from the Panhandle to the rest of the ERCOT region. In order to maximize wind generation export capacity, the Panhandle areas may need significant upgrades for voltage stability, short circuit strength, and high voltage ride-through capability. Texas RE will continue to monitor the ongoing studies in this area as well as monitor generation interconnections for actual implementation of wind projects in the Panhandle.

Additional renewable energy topics being monitored by Texas RE are: (1) Model requirements for renewable generation. (2) Calculating the capacity values attributable to renewable generation connected to

Transmission and Distribution facilities in planning assessments. (3) Requirements for renewable generation:

a. Voltage and frequency ride-through b. Reactive and real power control c. Inertial response criteria

(4) Revisions to planning and operational practices, procedures, and tools to integrate renewable generation.

(5) Reliability impacts to the transmission system if renewable generation connected to distribution facilities do not remain interconnected, stable, and operational during and after normally expected momentary system disturbances in the transmission system.

(6) Renewable generation forecasting requirements for operations.


2. Emerging Issues

The NERC Essential Reliability Services Task Force published its Measures Framework Report in December 20151. In the report, the Task Force report detailed important directional measures to help the industry understand and prepare for the increased deployment of variable energy resources, retirement of conventional coal units, advances in demand response technologies, and other changes to the traditional characteristics of generation and load resources. The recommendations focused on the broad areas of managing frequency, load ramping, voltage control, and dispatchability. Specific recommendations impacting the ERCOT region include development of industry practices and measures for synchronous inertia at the Interconnection level, frequency response at the interconnection level, real time inertial models, net demand ramping variability, system reactive capability and overall reactive performance, and system short circuit strength.

3. Ancillary Service Changes

Changes to the Ancillary Service products and methodology continue to be discussed among ERCOT region stakeholders and working groups. Texas RE will continue to monitor these discussions and the progress of changes to the Nodal Protocols as the ancillary service changes are implemented.2

4. Protection System Misoperations

The percentage of misoperations caused by “human performance” (i.e., settings errors, design errors, wiring errors, etc.) continue to be the cause of between 40% and 50% of the total misoperations. In many cases, setting errors appear to be associated with a weak understanding of a setting not available on an electromechanical or solid state relay, lack of compensation for tap loads, lack of coordination between forward looking distance elements and reverse looking blocking elements, settings made without considering various system configurations, instantaneous ground overcurrent elements that overreach a remote bus, and incorrect directionality-related settings. Inaccurate modeling has also been noted as a common cause of misoperations, with issues such as incorrect transformer connections, lack of consideration for mutual coupling, and out-of-date line impedances.

5. SCADA/EMS/Telemetry Performance

From 2013-2015, there were a total of 17 loss of EMS/SCADA events reported in the ERCOT region. Loss of EMS or SCADA events will continue to be of concern due to their impact on visibility and situational awareness for system operators. Accuracy and availability of telemetry is a key issue for situational awareness for system operators as

1 NERC Essential Reliability Services Task Force Measures Framework Report: http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/ERSTF%20Framework%20Report%20-%20Final.pdf 2 ERCOT Future Ancillary Services Team (FAST) website: http://www.ercot.com/committees/other/fast/index.html

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/ERSTF%20Framework%20Report%20-%20Final.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/ERSTF%20Framework%20Report%20-%20Final.pdf

http://www.ercot.com/committees/other/fast/index.html


well as real-time assessment tools. Texas RE is active on the NERC EMS task force and is facilitating a regional working group with stakeholders to discuss these issues.

6. Voltage Control

Discussions are ongoing with ERCOT stakeholders on revisions to the voltage profile process and real-time voltage control procedures within the Nodal Protocols and Operating Guides. A new ERCOT working group was created in 2015 to facilitate the voltage profile planning process. Texas RE will continue to monitor these discussions and the progress of changes to the Nodal Protocols as the changes are implemented.

7. Critical Infrastructure Protection

Critical Infrastructure Protection (CIP) will continue to remain a priority for NERC, the Department of Homeland Security, and Texas RE for the foreseeable future. The effectiveness of the NERC Physical Security Standard (CIP-014) implementation will continue to be evaluated in 2016.


I. Disturbances and Events

Introduction

While major outages of the Bulk Electric System are rare, minor events and outages are a common occurrence in a system as complex as ERCOT. Many factors contribute to these disturbances, including line exposure over large geographic areas, misoperations of protective devices, and the multitude of elements that are required to operate and monitor a system as complex as the electrical grid.

This section provides information about disturbances events in the region.

Additional data on events analysis is presented in Appendix B.

Observations

Weather is the number one cause of disturbances identified from the event analyses, followed by equipment failures, vandalism/sabotage/theft, and protection system misoperations.

The occurrences of multiple generator trips due to sympathetic tripping and unit trips at the same site due to a single point of failure are continued cause for concern. Review of these events has produced corrective actions at the specific sites, however further analysis is needed to determine common failure modes and share lessons learned to reduce these types of events.

Loss of EMS and SCADA events continue to be a concern due to their impact on visibility and situational awareness for system operators.

Prompt correction of identified causes and support for developing industry lessons learned should be continually monitored and improved upon. This will lead to a more rigorous self-analysis by registered entities, which are expected to conduct such analyses. Additionally, it will lead to the development of a stronger culture of reliability. Lessons learned as a result of an event analysis should be shared with the industry in a timely manner to serve as a learning initiative and enhance system reliability.

Historical Data and Trends

A. Key Events in 2015

Two notable system events occurred in 2015.

(1) Lower Rio Grande Valley event on July 3, 2015

On July 3, 2015 at 17:16, a horizontal post insulator failure on a 138 kV transmission line in the Lower Rio Grande Valley caused a multi-phase-to-ground fault. The line fell into the 138kV bus at the substation causing an A-

2015 Events in Brief

Events Reported: 101

Protection System Misoperations: 179

Generation Forced Outages: 1,856

345 kV Transmission Automatic

Outages: 477


phase-to-ground fault on the 138kV bus. The line also contacted 12 kV circuits at a nearby 69-12 kV substation. Multiple protection system misoperations and other sympathetic trips occurred to expand the scope of the outage. The most significant of these was the loss of a static VAR compensator (STATCOM) and capacitor bank at a nearby 138 kV substation. The loss of multiple elements created a low voltage condition in the local area on the transmission system which resulted in operation of Under-Voltage Load Shed (UVLS) relays in the area. Loss of firm load from the event was 95.5 MW of consequential load loss plus an additional 92 MW due to the UVLS relay operations. Additionally, 244 MW of wind generation tripped off after the initial fault. As different line energization attempts occurred over the next 15 minutes, multiple wind farms experienced additional turbine trips.

(2) Multiple generator loss event on July 29, 2015

On July 29, 2015 at 18:16:40, two combined cycle trains at a Central Texas generation facility tripped offline carrying a combined total of about 953 MW. A nearby coal unit also tripped carrying 554 MW. One of the 345 kV transmission lines serving the combined cycle facility also tripped due to a misoperation of the protection system. System frequency dropped to a minimum frequency of 59.723 Hz and recovered to 60 Hz in approximately 6 minutes, 6 seconds. After the loss of generation, an advisory was issued for Physical Responsive Capability (PRC) being below 3000 MW. The advisory was cancelled at 18:50.

B. Historical Disturbance Data

In 2015, the number of events reported continued to show a stable trend compared to 2012-2014.

Event Category3 2011 2012 2013 2014 2015

Non-Qualified 89 87 92 77 90

1 24 6 7 11 9

2 1 0 2 2 1

3 3 1 1 1 1

4 and 5 1 0 0 0 0

Total 118 94 102 91 101

Table 1 – Summary of Events Analysis

3 Link to NERC Events Analysis Process with category definitions:

http://www.nerc.com/pa/rrm/ea/EA%20Program%20Document%20Library/ERO_EAP_V3_final.pdf


Figure 5 – Events Reported by Quarter

Figure 6 – 2011-2015 Event Cause Summary

12

21

18

2

5

3 3

56

43

7

109

5

7

2

14

109

0

5

10

15

20

25

30

0

5

10

15

20

25

30

20111stQtr

20112ndQtr

20113rdQtr

20114thQtr

20121stQtr

20122ndQtr

20123rdQtr

20124thQtr

20131stQtr

20132ndQtr

20133rdQtr

20134thQtr

20141stQtr

20142ndQtr

20143rdQtr

20144thQtr

20151stQtr

20152ndQtr

20153rdQtr

20154thQtr

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

26%

3%

8%

14%

14%

1%

0%

34%

Equipment Failure

Natural Disaster/ForeignInterference

Protection SystemIssues

EMS

Sabotage/Vandalism

Cyber

Unknown

Weather

2011-2015 Event Cause


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. 2015 showed a sharp increase in the number and customer impact from these events versus 2014, primarily due to multiple severe weather events in the region. The trend in these reports is included in the following figure.

Figure 7 – OE-417 Reports of Lost Load

0

1

2

3

4

5

6

7

8

9

10

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1,400,000

1,600,000

1,800,000

2,000,000

2009 2010 2011 2012 2013 2014 2015

Total Customer Impact # Reportable Events


C. Generation Loss Events

Texas RE staff also review each loss of generation greater than 450 MW and frequency deviations less than 59.91 Hz within the ERCOT region. These events are reported to NERC Situational Awareness staff, but are not tracked or reported under the NERC Events Analysis Process. There were 65 in 2015 that were reviewed by Texas RE. Over the past five years, these events averaged 17 per quarter. The trend in generator unit trips is shown in Figure 5.

The unit trip events greater than 450 MW that occurred in 2015 can be broken down to indicate the major categories and causes of the unit trips based on GADS data.

Major System Number of Trips

Boiler System 16

Burner Management/Controls 4

Induced/Forced Draft Fans 5

Other 7

Balance of Plant 33

Feedwater System 4

Electrical/Instrumentation 20

Power Station Switchyard 0

Distributed Control System 1

Other 8

Steam Turbine/Generator 13

Valves/Piping/Lube Oil 4

Controls 5

Exciter 2

Generator 1

Pollution Control Equipment 0

External to Plant 2

Other 2

Table 2 – Major Causes of Generator Trips > 450 MW

Several key items of focus have been the identification of events where multiple generators trip at the same site, generator units trip due to faults or system conditions outside the generator’s protection system zones, or where it is not necessary to trip the unit in order to clear the fault condition. Tracking these types of events provides an indication of issues with generator protective relaying, generator excitation system issues, and control system issues. These events also go beyond normal single contingency planning criteria. In 2015, there were seven occurrences of multiple generator trips. Three of these events were due to protection system misoperations. Since 2011, other issues that have resulted in multiple unit trips include:

Loss of plant instrument air system which was common to multiple units at the same site

Protection system misoperations

Sympathetic trips on combined cycle units due to exhaust temperature spread

Fuel supply lines to gas plants (low gas pressure, control valve problems, etc.)


Failures in the plant auxiliary power system which was common to multiple units at the same site

Date Event Description

1/8/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to maintenance personnel error

1/15/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to a control system issue

4/27/2015 Two generation facilities tripped due to a nearby 345 kV system fault combined with a protection system misoperation

5/25/2015 Two natural gas units at the same site tripped due to a control system issue

6/26/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to generator instrument transformer issues

7/29/2015 Two 2x1 combined cycle trains at the same site tripped due to a HV substation protection system misoperation

12/30/2015 Two 2x1 combined cycle trains at the same site tripped due to a HV substation protection system misoperation

Table 3 – 2015 Events with Multiple Generator Trips

D. EMS/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 Bulk Electric System (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.

For 2013-2015, there were 17 loss of EMS/SCADA events that lasted 30 minutes or more reported in the ERCOT region. Events reported in 2015 include the following:

A Transmission Operator reported that it lost its State Estimator due to a software issue.

A Transmission Operator reported that its ICCP data to ERCOT ceased transmitting properly following an ICCP database update.

A Transmission Operator reported that it lost monitoring and control functionality due to a network firewall malfunction.

A Transmission Operator reported that it lost monitoring and control functionality due to a card failure.

Common themes include:

Improper sizing setting of parameters

Improper user/application permission issues

Incorrect application parameter settings

Incorrect database settings/configuration

Improper patch management

Incorrect recovery procedures


Missing documentation for programs, procedures, etc.

External program configuration issues (anti-virus, etc.)

Improper configuration of security tools

Improper server redundancy set up

Improper power sources redundancy

Improper Local/Wide Area Network (LAN/WAN) configuration

Incorrect communication network settings

Improper disk/memory sizing

Improper settings on routers, switches, etc.

Improper server clustering

Improper settings, procedures, or design requirements for failover

Inadequate testing


II. Transmission

Introduction

Texas RE collects transmission outage and inventory data annually each February from the 30 Transmission Owners throughout the ERCOT region for transmission elements operated at 100 kV and above using TADS. The outage data is separated into voltage classes, outage duration (momentary or sustained), and outage cause. These categories illustrate the types of outages occurring on the BES.

This section provides information summarizing the data collected from TADS as well as transmission performance data from other sources for the region. Additional data on TADS analysis is presented in Appendix C.

Observations

Mandatory TADS reporting of sustained outage data for circuits operated at less than 200 kV began in 2015.

There were no ERCOT IROL exceedances in 2015.

For the 345 kV system, approximately 21% of transmission outages in 2015 were due to unknown causes. Identifying and correcting the unknown outages may help entities detect and address system issues that could have a larger impact on reliability.

For the 345 kV system, failed substation equipment and failed transmission circuit equipment dominated the sustained outages, accounting for 77% of the outage duration. Approximately 58% of 345 kV sustained automatic outages lasted two hours or more.

For the 138 kV system, failed substation equipment and failed transmission circuit equipment also dominated the sustained outages, accounting for 81% of the outage duration. Approxmately 57% of 138 kV sustained automatic outages lasted two hours or more.

For the 345 kV system in 2015, 79 of the 477 (17%) reported automatic outage events involved two or more circuit elements. Dependent Mode outages (defined as an automatic outage of an element which occurred as a result of another outage) and Common Mode outages (defined as one or more automatic outages with the same initiating cause and occur nearly simultaneously) represented 14% of all momentary outages, 21% of all sustained outages, and 43% of sustained outage duration for the 345 kV system.

2015 Transmission Performance in Brief

345 kV Circuits: 413

345 kV Circuit miles: 14,832

345 kV Circuit Outages: 440

345 kV Transformer Outages: 37

345 kV Outage Duration: 8,617 hrs

138 kV Circuits: 1705

138 kV Circuit miles: 20,217

138 kV Circuit Sustained Outages: 425

138 kV Outage Duration: 13,310 hrs

Basecase constraint violations for at least

one SCED interval: 3571



A. 2015 TADS Metrics for ERCOT region

Compared to 2014 data, 2015 outage rates per 100 miles of line per year for the 345 kV system increased by 52% (from 1.97 to 2.99) and the total outage duration from automatic outages increased from 2,917 hours to 8,617 hours.

Momentary Outages Sustained Outages

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

300-399 kV 0.75 2.06 0.34 0.94

100-199 kV Not reportable Not reportable 0.25 2.11

Table 4 – 2015 Momentary and Sustained Outages

Though there was an increase in the outage rates from 2014 to 2015, long term trends show a stable trend in outage rates per circuit and per 100 miles of line. ERCOT region outage rates are also comparable to NERC-wide outage rates for the 300-399 kV overhead voltage class. See the following figure and table.

Figure 8 – 2008-2015 345 kV Automatic Outage Metrics

Voltage Class Name Metric 2011 2012 2013 2014 2015 5-Yr Avg

AC Circuit 300-399 kV

Automatic Outages per Circuit

0.94 0.75 0.56 0.71 1.08 0.81


Automatic Outages per 100 miles

2.89 2.41 1.63 1.97 2.99 2.38

0

0.5

1

1.5

2

2.5

3

3.5

2008 2009 2010 2011 2012 2013 2014 2015

ERCOT Outages / circuit ERCOT Outages/100 mi-year

NERC 300-399kV Outages per circuit NERC 300-399kV Outages per 100mi-year



Sustained Automatic Outages per Circuit

0.254


Sustained Automatic Outages per 100 miles

2.11

Table 5 – TADS Circuit and Automatic Outage Historical Data for ERCOT Region

B. Automatic Outage Data

For the 345 kV system, predominant causes for momentary outages in 2015 were lightning, foreign interference, and unknown, representing 74% of the total momentary outages. Predominant causes for sustained outages in 2015 were weather, lightning, failed substation/transmission equipment, and failed protection system equipment, representing 82% of the total sustained outages. Failed substation/transmission circuit equipment dominated the sustained outage duration, accounting for 77% of the outage duration. Approximately 58% of 345 kV sustained automatic outages lasted two hours or more.

For the 138 kV system, predominant causes for sustained outages in 2015 were weather, lightning, and failed circuit equipment, representing 51% of the total sustained outages. Failed substation/transmission circuit equipment dominated the sustained outage duration, accounting for 81% of the outage duration. Approximately 57% of 138 kV sustained automatic outages lasted two hours or more.

The number of reported momentary outages on elements operated at 200 kV and above increased by 77% between 2014 and 2015, from 180 to 319 primarily due to an increase in lightning-related outages. The number of reported sustained outages on elements operated at 200 kV and above also increased between 2014 and 2015, from 96 to 158.

4 2015 was the first year of TADS reporting for 138 kV circuits.


Figure 9 – 2015 345 kV Momentary Outage Cause

Figure 10 – 2015 Automatic Outages by Month

7%

40%

11%

13%

3%1%

1%

3%

1%

21%

2015 345kV Momentary Outage Cause Weather, excludinglightning

Lightning

Contamination

Foreign Interference

Fire

Failed AC SubstationEquipment

Failed Protection SystemEquipment

Failed AC CircuitEquipment

Human Error

Unknown

0

20

40

60

80

100

120

140

160

180

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

138kV Sustained Outages 345kV Sustained Outages345kV Momentary Outages

2015 Automatic Outages By Month


Figure 11 – 2015 Automatic Outage Duration by Month

Figure 12 – 2015 345 kV Sustained Outage Cause

0

1000

2000

3000

4000

5000

6000


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

2015 Automatic Outage Duration By Month

33%

26%1%2%

8%

1%

8%

4%

10%

1%3%

4%

2015 345 kV Sustained Outage Cause Weather, excludinglightningLightning

Environmental

Contamination


Fire

Failed AC SubstationEquipmentFailed Protection SystemEquipmentFailed AC CircuitEquipmentPower System Condition

Human Error


Figure 13 – 2015 138 kV Sustained Outage Cause

Figure 14 – 2015 345 kV Sustained Outage Duration

21%

13%

2%

7%1%

10%10%

17%

3%

0%

9%

5% 1%

2015 138kV Sustained Outage Cause Weather, excludinglightningLightning

Environmental


Fire

Failed AC SubstationEquipmentFailed Protection SystemEquipmentFailed AC CircuitEquipmentVegetation

Power System Condition

Human Error

Unknown

15%

3%

0%0%

1%

0%

26%

1%

52%

1% 1%

0%0%

2015 345kV Sustained Outage DurationWeather, excludinglightningLightning

Environmental

Contamination


Fire

Failed AC SubstationEquipmentFailed ProtectionSystem EquipmentFailed AC CircuitEquipmentPower SystemConditionHuman Error

Unknown

Other


Figure 15 – 2015 138 kV Sustained Outage Duration

Figure 16 – 2015 345 kV Momentary and Sustained Outages by Cause per Month

4% 7%

1%

1%

0%

13%

2%

68%

1%

0% 2%0%

1%

2015 138kV Sustained Outage DurationWeather, excludinglightningLightning

Environmental

Contamination


Vandalism, Terrorism, orMalicious ActsFire

Failed AC SubstationEquipmentFailed Protection SystemEquipmentFailed AC CircuitEquipmentVegetation


0

10

20

30

40

50

60

70

80

90

100

Other Unknown Human ErrorPower System Condition Failed AC Circuit Equipment Failed Protection System EquipmentFailed AC Substation Equipment Fire Foreign InterferenceContamination Lightning Weather, excluding lightning

2015 345 kV Momentary and Sustained Outages By Cause Per Month


Figure 17 – 2015 345 kV Automatic Outage Data by Duration

Figure 18 – 2015 138 kV Sustained Outage Data by Duration

The following figure shows a comparison of 2015 automatic outage rates per circuit and per 100 miles of line between different ERCOT Transmission Owners compared to the aggregated region performance.

24%

6%

6%

6%34%

7%

17%

2015 345 kV Automatic Outage Count by Duration

1-5 minutes

6-10 minutes

11-30 minutes

31-120 minutes

121 minutes to 24 hours

>24 hours to 48 hours

> 48 hours

17%

4%

5%

16%41%

7%

10%

2015 138 kV Automatic Outage Count by Duration

1-5 minutes

6-10 minutes

11-30 minutes

31-120 minutes

121 minutes to 24hours

>24 hours to 48hours

> 48 hours


Figure 19 – 2015 345 kV Outage Rates by Entity

C. Common Mode and Dependent Mode Outage Data

For the 345 kV system in 2015, 79 of the 477 reported automatic outage events involved two or more circuit elements. Dependent Mode outages (defined as an automatic outage of an element which occurred as a result of another outage) and Common Mode outages (defined as two or more automatic outages with the same initiating cause and occur nearly simultaneously) represented 14% of all momentary outages, 21% of all sustained outages, and 43% of sustained outage duration for the 345 kV system.

Figure 20 – 2015 345 kV Outage Cause and Duration by Outage Mode

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

ERCOTRegion

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

2015 345 kV Outage Rates per Circuit and per 100 miles - By Entity

Automatic Outages per Circuit Automatic Outages per 100 miles of Line

83%

0%

2%

12% 2%

Single Mode

Dependent ModeInitiating

Dependent Mode

Common Mode

Common ModeInitiating

2015 345 kV Momentary and Sustained Outage Mode

57%

0%

1%

41%

0%

Single Mode


Dependent Mode

Common Mode


2015 345 kV Outage Duration by Outage Mode


For 2010-2015 combined, Dependent Mode outages and Common Mode outages represented 10% of all momentary outages, 29% of all sustained outages and 49% of sustained outage duration for the 345 kV system.

When you break down the 345 kV Common Mode and Dependent Mode outages by Event Type, outages of two or more elements on common structures (Event Type 31) represented 20% of these outages and 67% of the outage duration.

For the 138 kV system in 2015, 134 of the 425 reported automatic sustained outage events in 2015 involved two or more circuit elements. Dependent Mode outages and Common Mode represented 31% of all sustained outages and 27% of sustained outage.

Figure 21 – 2015 138 kV Outage Cause and Duration by Outage Mode

When you break down the 138 kV Common Mode and Dependent Mode outages by Event Type, outages of two or more elements on common structures (Event Type 31) represented 6% of these outages and 55% of the outage duration. Protection system failures (Event Types 61 and 62) represented 37% of the Common Mode and Dependent Mode outages.

D. Vegetation Management

Conductor contact with trees has been an initiating trigger and a contributing factor in several major system disturbances since 1965, including the August 14, 2003, Blackout. Tree contact caused the loss of multiple transmission circuits in several of the outages, causing multiple contingencies and further weakening of the system.

NERC began collecting vegetation-related outage information in 2004 as a result of the August 14, 2003 Blackout. Initiatives to reduce vegetation-related outages include quarterly vegetation management reports and self-certification of vegetation-related outages by Transmission Owners (TO) through the enforcement of the FAC-003 Standard.

69%

4%

10%

16%1%

Single Mode


Dependent Mode

Common Mode


2015 138 kV Sustained Outage Mode

73%

2%4%

21%

0%

Single Mode


Dependent Mode

Common Mode


2015 138 kV Outage Duration by Outage Mode


In the ERCOT region in 2015, there were no vegetation-related outages reported in the TADS system for 345 kV circuits and 20 reported vegetation-related outages for 138 kV circuits.

E. System Operating Limit and Interconnection Reliability Operating Limit Performance

A System Operating Limit (SOL) 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 2015, there were no exceedances of the North-Houston stability limit.

ERCOT utilizes 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 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 utilizes the 15-minute rating of the impacted transmission facility(ies), if available. The following charts show the monthly trend in transmission facility constraints where the thermal rating of the facility was exceeded post-contingency (i.e., an SOL exceedance).

In 2015, there were 3,571 violations of basecase limits for at least one SCED interval that required ERCOT to take some sort of action. There were approximately 27,400 post-contingency constraint limit violations for at least one SCED interval. Since June 2012, 138 kV line constraints averaged 33 circuits each month with an average duration of 5.1 hours per circuit per month. 345 kV line constraints averaged 6 circuits each month with an average duration of 6.7 hours per circuit per month. Some, but not all, of these constraints had an associated CMP. Table 6 shows the list of the top constraints for 2015.


Constraint (Binding Element) Number of Days

Approx. Number of

Hours

Transmission Project

Bruni 138/69 kV autotransformer 123 195

Bosque Switch – Rogers Hill 138 kV 25 38

Carrollton NW – Lakepointe 138 kV 36 58 Yes

Coleto Creek – Kennedy Switch 138 kV 26 19

Gibbons Creek – Singleton 345 kV 16 80

Hamilton – Maverick 138 kV 39 36 Yes

Morgan Creek – Sun Switch 138 kV 34 56

Lon Hill – Smith 69 kV line 70 144 Yes

Singleton – Zenith 345 kV lines (2) 64 160 Yes

Twin Oak Switch – Jack Creek 345 kV line 60 99 Yes

Wolfgang – Rotan 69 kV line 75 51

San Angelo Power Station T1 138/69 kV 27 62

Zorillo – Ajo interface 122 210 Yes

Table 6 – 2015 Top Constraints

Figure 22 – Quarterly Trend in Transmission Line Constraints

F. Voltage Control

ERCOT Operating Guides require a generation resource to provide either leading or lagging reactive power up to the required capability of the unit upon request from a transmission operator or ERCOT ISO. The guides also require a generation resource to maintain the transmission system voltage at the point of interconnection with the transmission system within 2% of the voltage profile while operating at less than the

0

100

200

300

400

500

600

700

800

0

20

40

60

80

100

120

140

160

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

345kV Line Constraint Total Hrs 138kV Line Constraint Total Hrs

345kV Lines (# of Ckts w/ Post-Contingency Overload) 138kV Lines (# of Ckts w/ Post-Contingency Overload)# Ckts Hours


maximum reactive capability of the generation resource. ERCOT voltage control procedures also require the transmission operators to maintain bus voltages between 95% and 105% of nominal during normal operating conditions and between 90% and 110% of nominal post-contingency. The following chart shows an examples of a voltage control analysis for multiple generation buses. For this chart (and the following “whisker” plot chart), the top and bottom of the box represent the 75th and 25th percentiles, respectively. The “whiskers”, which extend above and below the box, indicate the minimum and maximum values in the data sample. This chart is based on one-minute telemetry data. The generation voltage control chart is normalized based on the voltage profile setpoint for each generation bus. The red lines indicate the +/- 2% range for the voltage profile currently specified by ERCOT Nodal Operating Guides.

Figure 23 – Generation Bus Voltage Control Chart for August 2015

The following chart shows the 2015 voltage control analysis for twenty 345 kV buses defined by ERCOT as being important for State Estimator to converge to a correct solution. This chart is based on one-hour telemetry data. The red lines indicate the +/- 5% operational voltage limits.

0.94

0.96

0.98

1.00

1.02

1.04

1.06

Bus 1345kV

Bus 2345kV

Bus 3138kV

Bus 4345kV

Bus 5345kV

Bus 6138kV

Bus 7345kV

Bus 8345kV

Bus 9345kV

Bus10

345kV

Bus11

345kV

Bus12

345kV

Bus13

345kV

Bus14

138kV

Bus15

345kV

Bus16

345kV

Bus17

345kV

Bus18

138kV

Bus19

345kV

Bus20

345kV

Bus21

345kV

Bus22

138kV

Bus23

345kV

Bus24

138kV

Bus25

138kV

Bus26

345kV

Bus27

138kV

Bus28

138kV

Bus29

345kV

Bus30

345kV

Bus31

345kV

Bus32

345kV

Bus33

345kV

Bus34

345kV

Bus35

345kV

Bus36

345kV

Bus37

138kV

Bus38

345kV

Bus39

345kV

Bus40

345kV

Bus41

345kV

Bus42

345kV

Bus43

69kV

Bus44

345kV

Bus45

138kV

Bus46

138kV

Bus47

345kV

Bus48

345kV

Bus49

345kV

Bus50

138kV

Vo

ltag

eP

rofi

le C

on

tro

l Po

int


Figure 24 – 345 kV Bus Voltage Chart for Important State Estimator Buses 2015

Distribution Service Provider (DSP) Responsibilities Related to Voltage Support ERCOT Nodal Protocols required DSPs to install sufficient static reactive power capability in substations and on the distribution voltage system to maintain at least a 0.97 lagging power factor for the maximum net active power measured in aggregate on the distribution system. DSPs provide power factor data as part of the Annual Load Data Request (ALDR) each year. Note that the ALDR data is measured on the high side of the distribution transformer, where the Protocol requirement for 0.97 power factor is at the distribution voltage level. A power factor of 0.96 was used for this analysis to allow for losses of approximately 1% through the distribution transformer. The following is a summary of the data for 2015 summer peak. Number of DSP’s reporting: 33 Total number of substations measured: 3,264 Total MW at summer peak for DSP’s reporting: 62,956 MW Aggregate power factor: 0.975 Number of substations with power factor less than 0.96: 594 Percentage of substations with power factor less than 0.96: 18.2%

G. Special Protection Systems

Special Protection Systems (SPSs) are protective relay systems designed to detect abnormal ERCOT system conditions such as transmission contingency overloads and take automatic pre-planned corrective actions to maintain a secure system. The following chart shows the trend in SPSs in service, as well as operating procedures and guides used congestion management, since 2011 reported by Texas RE to NERC. Operating procedures and guides include Remedial Action Plans, Mitigation Plans, and Pre-

0.94

0.96

0.98

1

1.02

1.04

1.06

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

Voltage Control Charts


Contingency Action Plans as defined in ERCOT Protocols and Operating Guides. For the purposes of this chart, “economic” SPSs were removed since these types of SPSs are not necessary to meet the NERC Transmission Planning standards.

Figure 25 – SPS and Operating Procedure Trends

ERCOT Operating Guides require owners of SPSs to report operations of these systems to Texas RE on a quarterly basis. The following figure shows the trend in arming/disarming operations and activation of the SPS systems. Since 2011 Q3, there has only been one reported misoperation of an SPS in the ERCOT region (shown by the yellow dot in 2012 Q4 on the chart).

0

10

20

30

40

50

60

2011 2012 2013 2014 2015 2016 (est)

SPS Operating Procedure/Guides


Figure 26 – Special Protection System Operation Trend

0

1

2

3

4

5

6

7

8

9

10

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

SPS Arm/Disarm Operations SPS Activations/Operations


III. Generation

Introduction

Texas RE began collecting generation outage information for units 50 MW and larger in January 2012 through the Generation Availability Data System (GADS). Starting in January 2013, units 20 MW and larger MW began mandatory reporting in GADS. The GADS data is used to calculate various generation metrics, including gross and net capacity factors, scheduled and forced outage rates, availability factors, seasonal de-rating factors, and starting reliability and average run times. ERCOT generators provided GADS data represent approximately 79% of the installed nameplate capacity within the region.

Additional data on GADS analysis is presented in Appendix D.

Observations

Record wind generation: 13,883 MW on December 20, 2015 Record wind penetration: 44.7% of total energy on December 20, 2015 GADS EFORd: 5.5% for 2015 vs. 5.4 % for 2014 The relatively mild winter in 2015 did not result in any major incidents of generator trips or

de-rates due to cold weather. The portion of total energy supplied by natural gas continued to trend upward in 2015

compared to previous years. The portion of total energy supplied by coal continued to decline as natural gas and wind supplied greater portions of total energy.

As of December 2015, ERCOT projections indicate solar generation will increase to over 1,400 MW and wind generation will increase to almost 21,000 MW over the next two years based on current signed generation interconnect agreements with financial security. Texas RE will continue to monitor the solar and wind generation growth as well as its use in short-term and long-term assessments of resource adequacy.

Changes to the Ancillary Service products and methodology continue to be discussed among ERCOT region stakeholders and working groups as the capacity of renewable generation increases.


A. Resource Portfolio

In 2015, the reported nameplate capacity was 112,120 MW of all types of generators. Coal, natural gas, and wind comprise more than 93% of the installed capacity.

2015 Generation Performance in Brief

Nameplate Capacity: 112,120 MW

Net Generation

From Nuclear: 39,384 GWH

From Wind: 40,786 GWH

From Natural Gas: 167,894 GWH

From Coal/Lignite: 97,655 GWH

From Other: 1,747 GWH

Forced Outages (GADS): 1,856

GADS EFORd: 5.5% for 2015


Figure 27 – 2015 Generation Nameplate Capacity

The portion of total energy supplied by natural gas continued to trend upward in 2015 compared to previous years. The portion of total energy supplied by coal continued to decline as natural gas and renewables supplied greater portions of total energy. 2015 also marked the first year that wind generation supplied a greater percentage of total energy than nuclear.

Natural Gas 66,040

59%

Coal21,470

19%

Nuclear5,250

5%

Wind17,130

15%

Hydro1,010

1%

Solar3000%

Other9301%

2015 Generation Nameplate Capacityby Fuel Type


Figure 28 – 2015 Energy by Fuel Type

Figure 29 – Energy by Fuel Type Trend

Natural Gas 167,982

48%

Coal97,655

28%

Nuclear39,384

11%

Wind40,786

12%

Other1,811

1%

2015 Energy (GWH) by Fuel Type

0%

10%

20%

30%

40%

50%

60%

Fuel Type as % of Total Energy

Natural Gas Coal Nuclear Wind

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


B. 2015 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 performance metrics based on unweighted versus weighted values for the ERCOT generation fleet for 2015 is provided in the following table.

Metric ERCOT region GADS Data 2015

NERC Fleet Average 2009-2013

Unweighted Weighted Unweighted Weighted

Net Capacity Factor (NCF) 45.51% 45.51% 36.81%

Service Factor (SF) 49.83% 62.46% 45.26% 49.61%

Equivalent Availability Factor (EAF)

85.82% 85.15% 84.99% 84.72%

Scheduled Outage Factor (SOF) 9.07% 9.46% 8.77% 9.76%

Forced Outage Factor (FOF) 3.12% 3.19% 4.34% 3.67%

Equivalent Forced Outage Rate (EFOR)

6.99% 6.51% 19.60% 15.41%

Equivalent Forced Outage Rate Demand (EFORd)

5.50% 2.65% 8.22%

Table 7 – ERCOT Generation Performance Metrics January through December 2015

Net Capacity Factor: NCF = Σ (Net Actual Generation) / Σ (NMC x PH)

Service Factor: SF = Σ SH / Σ PH

Availability Factor: AF = Σ AH / Σ PH

Scheduled Outage Factor: SOF = Σ (POH + MOH) / Σ PH

Forced Outage Factor: FOF = Σ FOH / Σ PH

Equivalent Forced Outage Rate: EFOR = Σ ( FOH + EFDH ) / Σ (FOH + SH + Synch Hours + Pump Hours + EFDH)

Where:

Forced Outage Hours (FOH) Equivalent Forced De-rate Hours (EFDH) Period Hours (PH) Planned Outage Hours (POH) Maintenance Outage Hours (MOH) Availability Hours (AH) Service Hours (SH) Net Maximum Capability (NMC) GADS metrics in Table 7 for the ERCOT region in 2015 were in-line or better than the NERC fleet average for 2009-2013, the latest years that data is available. The Equivalent Forced Outage Rate – Demand (EFORd), which measures the rate of forced outage events on generating units during periods of load demand, was lower for ERCOT region units in comparison to the NERC fleet average, indicating a lower risk that a unit may not be available to meet generating requirements due to forced outages or de‐ratings.


Figure 30 – GADS Generation Performance Metrics by Fuel Type and Year

During the period from January 2015 through December 2015, the average FOH was 263.9 hours, the average MOH (including extensions to MO) was 131.3 hours, and the average POH (with extensions of PO) was 634.7 hours.

Metric (per unit) ERCOT region 2012

ERCOT region 2013

ERCOT region 2014

ERCOT region 2015

4-Year Average

Avg Forced Outage Hours 282.7 256.2 248.9 263.9 262.9

Avg Maintenance Outage Hours

203.0 149.0 134.7 131.3 154.5

Avg Planned Outage Hours 516.9 511.6 594.7 634.7 564.5

Table 8 – Average GADS Generation Unit Outage Hours

C. 2015 Outages and Derates

From January 2015 through December 2015, there were a total of 1,856 immediate forced outage events, totaling 98,848 hours, with a total outage capacity of 401,354 MW, or an average of 216 MW per event. The majority of the immediate forced outage events occurred due to waterwall trips, control issues, blade path temperature spreads, high exhaust temperatures, and vibration issues.

During the same period, there were a total of 2,729 immediate de-rate events, totaling 116,826 hours, with a total de-rate capacity of 272,844 MW, or an average of 100 MW per de-rate event. The majority of the immediate de-rate events occurred due to low BTU coal and opacity issues. Reference the following chart and graphics.

0%

2%

4%

6%

8%

10%

12%

14%

Total Coal, Lignite,Fluidized Bed

GG - Gas Nuclear Gas Turbine/JetEngine (Simple

Cycle Operation)

Combined CycleBlock

CC GT units CC steam units

2012 2013 2014 2015EFORd


Immediate De-Rates Immediate Forced Outages

Number of Events 2,729 1,856

Total Duration 116,825.9 hours 98,847.7 hours

Total Capacity 272,844.1 MW 401,354.4 MW

Avg Duration per Event 42.8 hours 53.3 hours

Avg Capacity per Event 99.9 MW 216.3 MW

Table 9 – Generator Immediate De-rate and Forced Outage Data (Jan. – Dec. 2015)

The cause of the immediate de-rate events can be further broken down into major categories

based on the GADS data.

Major System Number

of De-rate Events

Total Duration (hours)

Total Capacity

(MW)

Average Duration per Event (hours)

Average Capacity per Event (MW)

Boiler System 942 92,353.8 28,504.1 98.0 30.3

Nuclear Reactor 0 0.0 0.0 0 0

Balance of Plant 308 33,091.4 21,264.7 107.4 69.0

Steam Turbine/Generator 452 52,105.5 33,418.3 115.3 73.9

Heat Recovery Steam Generator 55 7,991.1 1,137.1 145.3 20.7

Pollution Control Equipment 242 19,760.4 5,260.2 81.7 21.7

External 627 55,992.3 24,543.5 89.3 39.1

Regulatory, Safety, Environmental 50 3,502.3 1,057.9 70.0 21.2

Personnel/Procedure Errors 24 4,717.3 235.2 196.6 9.8

Table 10 – 2015 Major Category Cause of Immediate De-rate Events from GADS

The following charts show the generation MW out of service for scheduled and forced outages

for the Summer 2015 and the Winter 2014-2015.


Figure 31 – Summer 2015 Generation Scheduled and Forced Outages

Figure 32 – Winter 2014-2015 Generation Scheduled and Forced Outages

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Summer 2015 Generation Scheduled and Forced Outages

Sched Outage MW Forced Outage MW Total Outage MW

0

5000

10000

15000

20000

25000

30000

Winter 2014-2015 Generation Scheduled and Forced Outages

Sched Outage MW Forced Outage MW Total Outage MW


The following charts show the 2015 GADS data for immediate forced outages and immediate

de-rate events broken down by fuel type and cause.

Figure 33 – 2015 Immediate De-rate and Forced Outage Events by Fuel Type

0

500

1000

1500

2000

2500

Fossil Nuclear Gas Turbine/JetEngine (Simple Cycle

Operation)

CC Block CC GT CC ST

Immediate De-rate Events Immediate Forced Outage Events


Figure 34 – 2015 Immediate De-rate/Forced Outage Duration (Hours) by Fuel Type

Figure 35 – 2015 Generator Immediate De-rate Events/Duration by Cause

0

10000

20000

30000

40000

50000

60000

70000

Fossil Nuclear Gas Turbine/JetEngine (Simple

Cycle Operation)

CC Block CC GT CC ST

Immediate De-rate Duration (hours) Immediate Forced Outage Duration (hours)

0

50

100

150

200

250Immediate De-Rate Duration (Avg Hours Per Event) Immediate De-Rate Capacity (Avg MW per Event)


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)

Average Duration per Event (hours)

Average Capacity per Event (MW)

Boiler System 206 9,471.7 79,432.6 46.0 385.6

Nuclear Reactor 0 0.0 0.0 0 0

Balance of Plant 404 17,674.5 88,767.6 43.7 219.7

Steam Turbine/Generator 942 60,187.7 162,537.8 63.9 172.5

Heat Recovery Steam Generator 65 3,999.3 13,833.2 61.5 212.8

Pollution Control Equipment 23 268.7 2,678.4 11.7 116.5

External 90 3,069.3 25,349.8 34.1 281.7

Regulatory, Safety, Environmental 12 1,577.7 2,419.0 131.5 201.6

Personnel/Procedure Errors 74 257.1 17,420.0 3.5 235.4

Table 11 – 2015 Major Category Cause of Immediate Forced Outage Events from GADS

Figure 36 – 2015 Generator Immediate Forced Outage Events/Duration by Cause

0

50

100

150

200

250

300

350

400

450Immediate Forced Outage Duration (Avg Hours Per Event) Immediate Forced Outage Capacity (Avg MW Per Event)


D. Renewable Generation Wind generation produced a total of 40,786 GWH in 2015, an increase of 13% from 2015. Wind generation, as a percentage of total ERCOT energy produced, increased to 11.7% in 2015, up from 10.6% in 2014. In 2015, instantaneous output of wind generation reached a record of 13,883 MW on December 20, 2015, and wind generation served a record of 44.7% of system demand on December 20, 2015. 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 37 – 2008-2015 Wind Generation MWh

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

4,000,000

4,500,000

5,000,000

Jan

-08

Ap

r-0

8

Jul-

08

Oct

-08

Jan

-09

Ap

r-0

9

Jul-

09

Oct

-09

Jan

-10

Ap

r-1

0

Jul-

10

Oct

-10

Jan

-11

Ap

r-1

1

Jul-

11

Oct

-11

Jan

-12

Ap

r-1

2

Jul-

12

Oct

-12

Jan

-13

Ap

r-1

3

Jul-

13

Oct

-13

Jan

-14

Ap

r-1

4

Jul-

14

Oct

-14

Jan

-15

Ap

r-1

5

Jul-

15

Oct

-15

Wind Energy (MWH)


Figure 38 – 2008-2015 Wind Generation as a Percentage of ERCOT Total Energy

Figure 39 – 2008-2015 Wind Generation as Percentage of ERCOT Total Energy by Month

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%Ja

n-0

8

Ap

r-0

8

Jul-

08

Oct

-08

Jan

-09

Ap

r-0

9

Jul-

09

Oct

-09

Jan

-10

Ap

r-1

0

Jul-

10

Oct

-10

Jan

-11

Ap

r-1

1

Jul-

11

Oct

-11

Jan

-12

Ap

r-1

2

Jul-

12

Oct

-12

Jan

-13

Ap

r-1

3

Jul-

13

Oct

-13

Jan

-14

Ap

r-1

4

Jul-

14

Oct

-14

Jan

-15

Ap

r-1

5

Jul-

15

Oct

-15


0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

18.00%

20.00%

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

2008 2009 2010 2011 2012 2013 2014 2015



The capacity factor for wind generation and other renewable resources has been a topic for several years. The following graphs show the distribution of capacity factor for all wind generation for the summer peak hours-ending of 1400-1800 for 2015 and the distribution of coastal/non-coastal wind capacity factors and solar capacity factors across all operating hours for August 2015. The August 2015 data clearly shows the distinct differences between the capacity factors of the coastal and non-coastal wind during the peak and off-peak hours.

Figure 40 – 2015 Wind Capacity Factors for Summer Peak Hours

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

Wind Capacity Factor, Jun-Sept 2015, Hours 14-18


Figure 41 – August 2015 Coastal/Non-Coastal Wind and Solar Generation Capacity

Factors

0%

10%

20%

30%

40%

50%

60%

70%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Average Capacity Factor by Operating Hour, August 2015 Coastal Wind

Non-Coastal Wind

Solar


IV. Load and Demand Response

Introduction

Demand Response (DR) is one of many resources needed to manage the increasing demand for electricity and lack of resource adequacy. In addition to providing capacity for resource adequacy purposes, capacity and ancillary services provided by DR help to ensure resource adequacy while providing operators with additional flexibility in maintaining operating reliability.

The NERC DADS Working Group has specified statistics to quantify demand response performance and the data collection requirements. The goal of DADS is to collect Demand Response enrollment and event information to measure its actual performance, including its contribution to improved reliability. Ultimately, this analysis can provide industry with a basis for projecting contributions of dispatchable and non-dispatchable Demand Response to support forecast adequacy and operational reliability.

Observations

Record peak demand: 69,877 MW on August 10, 2015

Load Resources providing Responsive Reserve Service continue to be an invaluable tool to respond to significant frequency disturbances and energy emergencies. Emergency Response Service (ERS) has also proven effective during energy emergencies.


Total energy consumption increased by 2.2% in 2015 versus 2014, to over 347,000 GHW. Peak demand also increased in 2015 by 5.0% versus 2014, reaching an all-time record of 69,877 MW on August 10, 2015.

2015 Load and Demand Response in

Brief

Summer Peak Demand: 69,877

Winter hourly Peak Demand: 56,750

Total Energy GWH: 347,466

Load Resource Deployments: 1

Demand Response Deployments: 66


Figure 42 – Annual Energy and Peak Demand

Figure 43 –Energy by Area

50,000

55,000

60,000

65,000

70,000

75,000

250,000

260,000

270,000

280,000

290,000

300,000

310,000

320,000

330,000

340,000

350,000

2008 2009 2010 2011 2012 2013 2014 2015

Annual Energy Peak DemandGWH MW

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

Austin DFW Houston Laredo San Antonio Valley Other

2014 2015

Total Energy by Area (GWH)


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

1. Load Resources (LR) providing Responsive Reserve Service (RRS) 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. The total amount of LRs is limited to 50% of the total ERCOT RRS requirement, or a maximum of 1,400 MW, for any given hour.

2. 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. ERCOT may deploy ERS following an EEA Level 2 instruction. ERCOT may also deploy ERS Resources immediately following an EEA Level 3 instruction if ERCOT System conditions do not allow time for ERCOT to deploy ERS prior to 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. Demand response employed by non-opt-in entities (NOIEs) such as municipalities in the form of commercial-industrial programs, smart thermostat programs, peak shaving programs, etc.

The following chart shows the registered capacity and average hourly committed capacity for demand response since April 2011. The DR MW Deployed-ERCOT represents Load Resource and ERS deployments by ERCOT. The DR MW Deployed-Other represents demand response deployed by NOIEs.

Figure 44 – Demand Response Availability

0

1000

2000

3000

4000

5000

6000

7000

8000

Ap

r-1

1

Jun

-11

Au

g-1

1

Oct

-11

Dec

-11

Feb

-12

Ap

r-1

2

Jun

-12

Au

g-1

2

Oct

-12

Dec

-12

Feb

-13

Ap

r-1

3

Jun

-13

Au

g-1

3

Oct

-13

Dec

-13

Feb

-14

Ap

r-1

4

Jun

-14

Au

g-1

4

Oct

-14

Dec

-14

Feb

-15

Ap

r-1

5

Jun

-15

Au

g-1

5

Oct

-15

Dec

-15

Registered Capacity (MW) Average Hourly Committed DR Capacity (MW) 3 per. Mov. Avg. (Registered Capacity (MW))

Demand Response Availability Data


Figure 45 – ERCOT Region Demand Response Availability and Deployments

From June-September 2015, there were a total 67 deployments of DR by NOIEs, or an average of 44.5 MW dispatched per deployment. The average response rate was 84% based on the MW reduction divided by the MW dispatched, or 37.3 MW per deployment.

Month # of Events

Sum of MW Dispatched

Sum of Realized

MW

Average Event

Duration (Hours)

Response Rate Based on

Maximum Hour of Response

June 12 501.8 339.7 1.3 68%

July 27 1,190.6 1,019.1 1.5 86%

August 22 959.4 864.6 1.4 90%

September 6 287.9 260.3 1.5 90%

Total 67 2,939.7 2,483.8 1.4 84%

Table 12 – Demand Response Deployments by Non-Opt-In Entities for 2015

Load Resources and ERS have proven to be extremely effective in providing response to significant loss of generation events (discussed further in Primary Frequency Response section of this report). The following table provides a list of Load Resource deployments for 2015.


7/29/2015 Approximately 16 MW of Load Resources deployed automatically by underfrequency relay due to a large generation trip event.

Table 13 – Demand Response Deployments in 2015

0

10

20

30

40

50

60

70

80

90

100

0.00

500.00

1000.00

1500.00

2000.00

2500.00

3000.00

3500.00

4000.00

4500.00

5000.00

2011Q2

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

DR MW Deployed - ERCOT DR MW Deployed - Other # DR Deployment EventsMW # Events


V. Frequency Control and Primary Frequency Response

Introduction

The ERCOT Performance, Disturbance Compliance Working Group (PDCWG) is responsible for reviewing, analyzing, and evaluating the frequency control performance of the ERCOT region. On a monthly basis, the group reviews various metrics and trends, and makes recommendations as needed for improvements to ERCOT’s frequency control process.

Observations

CPS1: 174.3 for calendar year 2015 vs. 163.3 for calendar year 2014 CPS2: 99.4 for calendar year 2015 vs. 98.8 for calendar year 2014 (Note: For information

only, ERCOT has a waiver for CPS2) Frequency Response: 720.4 MW/0.1 Hz vs. NERC obligation of 471 MW/0.1 Hz Average recovery time from generation loss events: 4.8 minutes vs. 4.9 minutes for

calendar year 2014


A. CPS1 Performance

NERC Reliability Standard BAL-001-1 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%. The following figure shows the ERCOT region CPS1 trend since January 2008. Since the start of the Nodal Market in December 2010, the region has made steady improvement in the CPS1 trend. For 2015, the annualized CPS1 score was 174.3.

2015 Frequency Control in Brief

All-time high Annual CPS-1: 174.3

All-time high Annual CPS-2: 99.4

Frequency Response: 720.4 MW/0.1 Hz

Average recovery time: 4.8 minutes


Figure 46 – CPS1 Average January 2008 to December 2015

Figure 47 – ERCOT CPS1 Annual Trend since January 2008

100

110

120

130

140

150

160

170

180

100

110

120

130

140

150

160

170

180

CP

S1 A

vera

ge

CPS1 12-Month Rolling Avg

127.5126.4

141.0

150.8

148.9

162.1

166.2 163.3

174.3

100.00

110.00

120.00

130.00

140.00

150.00

160.00

170.00

180.00

January2008

January2009

January2010

January2011

January2012

January2013

January2014

January2015

January2016

CPS1 Average


Figure 48 shows Bell curves of the ERCOT frequency profile, comparing 2011 through 2015. Two items are of note: (1) The shape of the Bell curve continues to narrow since 2011, indicating improved frequency control, and (2) The peak of the Bell curve remains shifted slightly from 60 Hz to 60.015 Hz. These are due, in part, to the percentage of generation units that have reduced turbine governor deadband settings from 0.036 Hz to 0.017 Hz and the effect of governors on wind turbines providing primary frequency response for high frequency excursions.

The green dashed lines on the chart 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 0.036 Hz. The purple dashed lines represent governor deadband settings of 0.017 Hz. The blue dashed lines represent three times the Epsilon-1 value which is used for frequency trigger limits per draft NERC Standard BAL-001-2.

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

Figure 48 – Frequency Profile Comparison

The following figure shows the 2015 CPS1 scores by operating hour compared to 2013 and 2014.

The CPS1 score by operating hour indicates possible issues for hour-ending (HE) 06, HE07, HE23, and HE24, though the frequency control during these hours improved greatly in 2015 when compared to 2013 and 2014. These issues are related to the procedures

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.182011 2012 2013 2014 2015

ERCOT Frequency Profile

0.017 Hz deadband

0.036 Hz deadband


used by generation resource entities during unit startup and shutdown. The figure also shows that the CPS1 score by operating hour for 2015 was 7 to 10 points higher versus 2014 across all hours.

Figure 49 – CPS1 Score by Hour for 2013 through 2015

B. Time Error Correction Performance

In 2015 from January through May, there were 19 Time Error Corrections, or an average of four each month, always for slow Time Error. ERCOT made changes to tuning parameters within the Load Frequency Control (LFC) system near the end of May 2015 which resulted in zero Time Error Corrections from June through the end of the year.

100

110

120

130

140

150

160

170

180

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

2013 2014 2015


Figure 50 – Time Error Corrections 2012-2015

Month Days Fast Slow Total Hours

on Control

Average Hours per Correction

Average Corrections

per Day

% of Time on

Correction

Cumulative TEC Hours

Jan-15 31 0 5 5 15.5 3.10 0.16 2.1% 15.50

Feb-15 28 0 3 3 9.5 3.17 0.11 1.4% 25.00

Mar-15 31 0 4 4 12.25 3.06 0.13 1.6% 37.25

Apr-15 30 0 3 3 9 3.00 0.10 1.3% 46.25

May-15 31 0 4 4 13 3.25 0.13 1.7% 59.25

Jun-15 30 0 0 0 0 0.00 0.00 0.0% 59.25

Jul-15 31 0 0 0 0 0.00 0.00 0.0% 59.25

Aug-15 31 0 0 0 0 0.00 0.00 0.0% 59.25

Sep-15 30 0 0 0 0 0.00 0.00 0.0% 59.25

Oct-15 31 0 0 0 0 0.00 0.00 0.0% 59.25

Nov-15 30 0 0 0 0 0.00 0.00 0.0% 59.25

Dec-15 31 0 0 0 0 0.00 0.00 0.0% 59.25

Avg 0 1.58 1.58 4.94 1.30 0.05 0.7%

Table 14 – 2015 Time Error Correction Summary

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

0

50

100

150

200

250

300

350

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

2012 % Time on Correction 2013 % Time on Correction2014 % Time on Correction 2015 % Time on Correction2012 Cumulative Time Error Correction Hours 2013 Cumulative Time Error Correction Hours2014 Cumulative Time Error Correction Hours 2015 Cumulative Time Error Correction Hours


There is a strong negative correlation between the average time error in a given hour versus the net load change during the hour. There is also a strong positive correlation between the average time error in a given versus the percentage error in the short-term load forecast during the same interval. These correlations are shown in the graphs below.

Figure 51 – Average Net Load Change vs. Time Error by Hour of Day

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-4000

-3000

-2000

-1000

0

1000

2000

3000

4000

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

Avg Net Load Change vs Delta Time Error: by Hour of Day 2015 SecondsMW


Figure 52 – Average Short-Term Load Forecast Error vs. Time Error by Hour of Day

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-10.0%

-8.0%

-6.0%

-4.0%

-2.0%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

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

Avg STLF Error vs Delta Time Error: by Hour of Day 2015 Seconds


C. Regulation Performance

The following table and graphs show regulation exhaustion rates for both Regulation-Up and Regulation-Down. For the purpose of these charts, the regulation exhaustion rate was based on the percentage of one-minute intervals where available regulation was less than 20 MW.

Regulation Exhaustion Rates

2011 2012 2013 2014 2015

Regulation-Up 0.99% 0.61% 0.79% 1.00% 1.18%

Regulation-Down 0.52% 0.61% 0.55% 0.58% 0.76%

Table 15 – Regulation Exhaustion Rates

Figure 53 – 2015 Regulation Exhaustion Rates by Operating Hour

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

UpReg Exhaustion Rate DownReg Exhaustion Rate


D. Basepoint Deviations and Generation-To-Be-Dispatched (GTBD)

Review of regulation deployments shows a long-term bias for Regulation-Up across several operating hours as indicated in the following graph. This graph shows the net regulation deployed, calculated as Regulation-Up MW deployed minus Regulation-Down MW deployed. Except for two operating hours, the trend shows a strong bias for Regulation-Up.

Figure 54 – 2015 Net Regulation Deployments by Operating Hour

-80

-60

-40

-20

0

20

40

60

80

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

Net Regulation Deployed (Reg-Up minus Reg-Down): by Hour of Day 2015

Avg Net Reg Deployed (RegUp - RegDn)


The long-term bias for Regulation-Up appears to be caused by generation units that are not operating at their Security Constrained Economic (SCED) basepoint. The aggregated basepoint deviation across the entire ERCOT generation fleet averaged -27 MW across all operating hours for 2015, which must be compensated for by the use of Regulation-Up. The following graph shows the net regulation deployed, calculated as Regulation-Up MW deployed minus Regulation-Down MW deployed, versus the aggregated basepoint deviations for the generation fleet.

Figure 55 – 2015 Net Regulation Deployment vs. Basepoint Deviation by Operating Hour

-100

-80

-60

-40

-20

0

20

40

60

80

100

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

Avg Basepoint Deviation and Net Regulation Deployed: by Hour of Day

Avg Basepoint Deviation Avg Net Reg Deployed (RegUp - RegDn)


There is also a strong correlation between the aggregated basepoint deviation versus time error as indicated in the following graph.

Figure 56 – 2015 Time Error vs. Basepoint Deviation by Operating Hour

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-100

-80

-60

-40

-20

0

20

40

60

80

100

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

Aggregated Basepoint Deviation vs Delta Time Error: by Hour of Day 2015


The following graph shows the 7-day moving average for the aggregated basepoint deviations for the generation fleet versus the 7-day moving average for net regulation deployed. The PDCWG has sponsored a System Change Request (SCR) that will modify the current GTBD formula to add an Area Control Error (ACE) integral term that will correct the long term bias in basepoint deviation and Regulation-Up deployments. Texas RE will continue to monitor this SCR and its implementation.

Figure 57 – 2015 Basepoint Deviation vs. Net Regulation 7-day Moving Average

-100

-80

-60

-40

-20

0

20

40

60

80

100

1/1

/15

0:0

01

/5/1

5 2

0:0

01

/10

/15

16

:00

1/1

5/1

5 1

2:0

01

/20

/15

8:0

01

/25

/15

4:0

01

/30

/15

0:0

02

/3/1

5 2

0:0

02

/8/1

5 1

6:0

02

/13

/15

12

:00

2/1

8/1

5 8

:00

2/2

3/1

5 4

:00

2/2

8/1

5 0

:00

3/4

/15

20

:00

3/9

/15

17

:00

3/1

4/1

5 1

3:0

03

/19

/15

9:0

03

/24

/15

5:0

03

/29

/15

1:0

04

/2/1

5 2

1:0

04

/7/1

5 1

7:0

04

/12

/15

13

:00

4/1

7/1

5 9

:00

4/2

2/1

5 5

:00

4/2

7/1

5 1

:00

5/1

/15

21

:00

5/6

/15

17

:00

5/1

1/1

5 1

3:0

05

/16

/15

9:0

05

/21

/15

5:0

05

/26

/15

1:0

05

/30

/15

21

:00

6/4

/15

17

:00

6/9

/15

13

:00

6/1

4/1

5 9

:00

6/1

9/1

5 5

:00

6/2

4/1

5 1

:00

6/2

8/1

5 2

1:0

07

/3/1

5 1

7:0

07

/8/1

5 1

3:0

07

/13

/15

9:0

07

/18

/15

5:0

07

/23

/15

1:0

07

/27

/15

21

:00

8/1

/15

17

:00

8/6

/15

13

:00

8/1

1/1

5 9

:00

8/1

6/1

5 5

:00

8/2

1/1

5 1

:00

8/2

5/1

5 2

1:0

08

/30

/15

17

:00

9/4

/15

13

:00

9/9

/15

9:0

09

/14

/15

5:0

09

/19

/15

1:0

09

/23

/15

21

:00

9/2

8/1

5 1

7:0

01

0/3

/15

13

:00

10

/8/1

5 9

:00

10

/13

/15

5:0

01

0/1

8/1

5 1

:00

10

/22

/15

21

:00

10

/27

/15

17

:00

11

/1/1

5 1

2:0

01

1/6

/15

8:0

01

1/1

1/1

5 4

:00

11

/16

/15

0:0

01

1/2

0/1

5 2

0:0

01

1/2

5/1

5 1

6:0

01

1/3

0/1

5 1

2:0

01

2/5

/15

8:0

01

2/1

0/1

5 4

:00

12

/15

/15

0:0

01

2/1

9/1

5 2

0:0

01

2/2

4/1

5 1

6:0

01

2/2

9/1

5 1

2:0

0

Aggregate Basepoint Deviation and Net Regulation Deployed - 7 day moving average 2015


E. CPS2 Performance

ERCOT is not required to report CPS2, however it is still useful to track and trend this metric. ERCOT has maintained its CPS2 score above the NERC minimum threshold of 90 for six consecutive years, 2010-2015.

Figure 58 – CPS2 Monthly Average – June 2008 to December 2015

F. Frequency Trigger Limit Performance

The Frequency Trigger Limits (FTLs) are defined as ranges for the Balancing Authority ACE Limit (BAAL) high and low values per NERC Standard BAL-001-2 which becomes enforceable in July 2016. The BAAL-high and BAAL-low calculations will replace the current CPS2 calculation when the Standard is approved. 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.

70

75

80

85

90

95

100

CP

S2

CPS2 12-Month Rolling Avg


High/Low Frequency

2011 Total Minutes

2012 Total Minutes

2013 Total Minutes

2014 Total Minutes

2015 Total Minutes

Low (<59.91 Hz)

370 131 82 63 13

High (>60.09 Hz)

151 26 9 7 1

Table 16 – Frequency Trigger Limit Performance

G. Primary Frequency Response

The following figure shows the trend in primary frequency response for the ERCOT region. The diamond in the middle of the box represents the median value. The number in the box represents the average value. In 2015, the average frequency response was 761 MW per 0.1 Hz and the median frequency response was 720 MW per 0.1 Hz as calculated per NERC Standard BAL-003. The NERC Reliability Standards require a maximum recovery time of 15 minutes for reportable disturbances. The following graphs show the annualized primary frequency response trend per NERC Standard BAL-003 and the detailed frequency response data since 2010 for the region.

Figure 59 – Annual Primary Frequency Response Trend for ERCOT Region

0

250

500

750

1000

1250

1500

2012 2013 2014 2015

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

761

535

722 726


Figure 60 – Histogram of ERCOT Frequency Response 2012-2015

Figure 61 – Primary Frequency Response Trend for ERCOT Region

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

200 400 600 800 1000 1200 1400 1600 1800 2000

Histogram of ERCOT Frequency Response 2012-2015

0

200

400

600

800

1000

1200

1400

1600

8/1/2010 8/1/2011 7/31/2012 7/31/2013 7/31/2014 7/31/2015

Primary Frequency Response Trend


As expected, the primary frequency response for a given event is correlated to the amount of physical response capability, or spinning reserve, that was available when the event occurred.

For 2015, the average event recovery time was 4.8 minutes. The recovery time for a given event depends on the amount of regulation reserves that are deployed when the event occurs.

Figure 62 – Primary Frequency Response versus Physical Response Capability

0

200

400

600

800

1000

1200

1400

1000 2000 3000 4000 5000 6000 7000 8000

Primary Frequency Response Vs Physical Response CapabilityPFR

PRC


Figure 63 – ACE Recovery Time versus Regulation Deployed

In the ERCOT region, a 450 MW generation loss threshold and/or a frequency change of 0.09 Hz is typically used as the event threshold for review and analysis.

In 2013, there was an average of 21 unit failures per event. In 2014, the unit failure rate declined to an average of 19 unit failures per event. In 2015 after the implementation of regional Standard BAL-001-TRE, the average unit failures per event is 11.

Failures of PDCWG Metrics 2013 2014 2015 (* Note)

Events Unit Failures



Events 56 1,171 44 826 25 275

Median Frequency Response 763 882 766

Table 17 – Failures of PDCWG Metrics by Unit Type

NOTE: 2015 data is valid for period April 1, 2015 through December 31, 2015, which correlates with the implementation of regional standard BAL-001-TRE.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

0 100 200 300 400 500 600 700 800 900

Recovery Time vs Regulation Deployed in Response to Frequency DeviationMinutes

Reg Deployed


VI. Protection System Performance

Introduction

Texas RE collects Protection System Misoperation data quarterly from registered Transmission Owners, Generation Owners, and Distribution Providers that own transmission throughout the ERCOT region for transmission elements operated at 100 kV and above. The Protection System Misoperation data is separated into voltage classes, category of Misoperation, element protected, relay system type, and Misoperation cause to illustrate the types of Protection System Misoperations occurring on the BES.

Protection System Misoperations create multiple reliability issues for the BPS. If no system fault is present, a misoperation can unexpectedly remove facilities, load, and/or generation from the system creating a condition which must be mitigated by system operators. If a misoperation occurs during a system fault, the expected actions fail to occur which could lead to cascading or voltage collapse. These events, which may go beyond applicable planning criteria, may represent a tangible threat to reliability.

Additional data on misoperations analysis is presented in Appendix G.

Observations

Protection system misoperation rate: 7.2% for 2015 vs. 8.7% for 2014 Incorrect settings, logic, and design errors accounted for 42% of misoperations in 2015 Relay failures accounted for 19% of misoperations in 2015


A. Protection System Misoperation Statistics

Since January 2011, the overall transmission system Protection System Misoperation rate has a slight downward trend, from 8.8% in 2011 to 7.2% in 2015.

138 kV 2011 2012 2013 2014 2015 5-Yr Avg

Number of Misoperations

151 113 112 111 139 125

Number of Events 1734 1388 1545 1421 1712 1560

Percentage of Misoperations

8.7% 8.1% 7.2% 7.8% 8.1% 8.0%

345 kV 2011 2012 2013 2014 2015 5-Yr Avg

2015 Frequency Control in Brief

2015 345 kV misoperation rate: 4.7%

2015 138 kV misoperation rate: 8.1%

2015 345 kV misoperations: 32

2015 138 kV misoperations: 139


Number of Misoperations

33 27 36 53 32 36

Number of Events 354 316 317 456 677 424

Percentage of Misoperations

9.3% 8.5% 11.3% 11.6% 4.7% 8.5%

Table 18 – Protection System Misoperation Data

Figure 64 – Protection System Misoperation Trends

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

2011Q1

2011Q2

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

Overall 345kV 138kV% Misoperation Rate


Three main categories account for 71% of the total misoperations: incorrect settings/logic design (42%), relay failures (19%), and communications failures (10%). This is similar to the trend seen NERC-wide.

Figure 65 – Protection System Misoperations by Cause 2011-2015

Seventy-five percent of the misoperations occurred at the 138 kV voltage level.

0

10

20

30

40

50

60

70

80

90

2011 2012 2013 2014 2015


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 2012 Q4 through 2015 Q3.

Figure 66 – Protection System Misoperation Rates by Entity 2012-2015

Figure 67 – Protection System Misoperation Rates by Region 2012 Q4-2015 Q3

0%

5%

10%

15%

20%

25%

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

Overall % Misoperation Rate by Entity 2012-2015


B. Human Performance Misoperations

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

138 kV line overtrip due to 67 ground instantaneous setting too low. Recent reconductor project caused increase in fault current and existing settings were no longer appropriate.

138 kV line overtrip for a remote end bus fault due to incorrect line length setting in relay.

138 kV bus overtrip due to miswired output contacts on an SEL line relay. Carrier start contacts accidently miswired to breaker failure start.

138 kV breaker trip during trip testing on adjacent circuit breaker due to mislabeled test switch.

138 kV GSU trip due to CT wire landed on wrong terminal.

345 kV line overtrip due to incorrectly wired relay output contact which prevented carrier blocking signal from being sent.

345 kV line overtrip due to incorrect CT polarity on both primary and backup relays.

Multiple 138 kV line overtrips due to carrier ground switch closed and coax disconnected from the line tuner.

138 kV breaker failure to trip due to incorrect zero-sequence voltage applied to relay.

138 kV bus overtrip for an external fault due to CTs from new breaker not included in the differential logic.

345 kV breaker overtrip due to voltage circuits miswired (rolled) in the PT junction box.

Multiple 138 kV line overtrips during a single event due to mis-coordinated instantaneous ground overcurrent settings.

138 kV line overtrip due to logic error in the relay trip equation.

138 kV line overtrip due to incorrect CT ratio.

Generator trip due to incorrect wiring by contractor.

138 kV line overtrips due to 67 ground directional overcurrent element tripping for fault on adjacent parallel circuit.

138 kV transformer trip due to swapped circuit breaker status inputs on the relay.

Generator trip for external fault due to incorrect configuration of the GSU CT ratios.

138 kV line failure to trip. Technicians failed to apply settings sent from engineering to the relays.

138 kV line overtrip due to incorrect wiring in the PT junction box.

138 kV line overtrip due to carrier ground switch left in the closed position.

345 kV auto trip due to incorrectly wired CTs in a 138 kV circuit breaker that was recently replaced.

138 kV line overtrip due to CTs left in the shorted position after maintenance.

138 kV line overtrip by ground instantaneous due to incorrect impedance in the short circuit model.


Figure 68 – Protection System Misoperations Trend Caused by Human Performance

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

2011Q1

2011Q2

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

Protection System Misoperations - Human Error %


VII. Infrastructure Protection

Introduction

Texas RE monitors infrastructure protection issues as part of its situational awareness effort. These issues primarily consist of substation intrusions and copper theft which 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.

Observations

Critical infrastructure protection will continue to remain a priority for NERC, the Department of Homeland Security, and Texas RE. Additional data should be monitored to analyze possible locational trends in intrusions, theft, or other physical security issues.


Since September 2011, substation intrusions and copper theft have ranged from three to 28 in any one month, averaging eight per month. Four issues required reporting under the DOE-417 requirements in the EOP-004 Standard. For the purposes of this chart, physical/cyber security issues include bomb threats, sabotage, and cyber security issues.

Figure 69 – ERCOT Trend in Substation Intrusions/Copper Theft/Cyber Security Issues

0

10

20

30

40

50

60

2011Q1

2011Q2

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

Physical/Cyber Security Intrusion/Copper Theft

Infrastructure Protection


VIII. Emerging Reliability Issues

Introduction

In December 2015, the NERC Essential Reliability Services Working Group published a report detailing important directional measures to help the industry understand and prepare for the increased deployment of variable energy resources, retirement of conventional coal units, advances in demand response technologies, and other changes to the traditional characteristics of generation and load resources. The recommendations focused on the broad areas of managing frequency, load ramping, voltage control, and dispatchability. Specific recommendations included development of industry practices and measures for synchronous inertia at the Balancing Authority and Interconnection level, frequency response at the interconnection level, real time inertial models, net demand ramping variability, system reactive capability and overall reactive performance, and system short circuit strength.

Initial Data and Trends

A. Synchronous Inertia at the Balancing Authority and Interconnection level

Since ERCOT operates as the single Balancing Authority (BA) for the region, the synchronous inertia measured at the BA level and interconnection level are the same. ERCOT began calculating synchronous inertia in July 2014 in order to better understand and manage the growth in wind generation. This chart shows the calculated synchronous inertia versus the system net load. This is a fairly linear relationship between the inertia and net load.

Figure 70 – Inertia versus Net Load

y = 4.2761x + 46231R² = 0.8379

100000

150000

200000

250000

300000

350000

10000 20000 30000 40000 50000 60000 70000

Inertia vs Net Load 2015System Inertia

Net Load


This chart shows the calculated synchronous inertia versus the percentage of load served by intermittent renewable resources (IRR), i.e., wind and solar generation. This chart also indicates a fairly linear relationship between the inertia and the IRR percentage.

Figure 71 – Inertia versus Percentage of Load served by IRRs

B. Frequency Response at the Interconnection level

Frequency response for the ERCOT region is covered in Section V.G.

C. Net Demand Ramping Variability

Changes in the amount of non-dispatchable resources, system constraints, load behaviors and the generation mix can impact the ramp rates needed to keep the system in balance. The Essential Reliability Services Working Group recommended that each Balancing Authority calculate the historical and projected maximum one-hour-up, one-hour-down, three-hour-up, and three-hour-down net demand ramps.

Ramping Variability Net Demand Wind Generation

Maximum One-Hour Increase 5,621 MW 3,367 MW

Maximum One-Hour Decrease -4,963 MW -2,506 MW

Maximum Three-Hour Increase 13,077 MW 5,319 MW

Maximum Three-Hour Decrease -12,188 MW -3,912 MW

Table 19 – Maximum and Minimum One-Hour Load and Wind Ramp for 2015

0

50000

100000

150000

200000

250000

300000

0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32% 34% 36% 38% 40% 42% 44%

2015 Avg Inertia vs IRR %


The following chart shows the average variability by season and operating hour for 2015.

Figure 72 – Average Net Load Ramp by Season and Operating Hour for 2015

D. System Short Circuit Strength

ERCOT has incorporated the use of a Weighted Short Circuit Ratio (WSCR) as part of the

criteria for determining the stability limits for the Panhandle interface export limit. The

ERCOT studies were performed to provide a generic set of WSCR limits for use combined

with the voltage stability limits, in order to provide an accurate stability limit for the interface

which balances voltage stability and system strength.

E. Fault-Induced Delayed Voltage Recovery (FIDVR)

FIDVR is characterized by depressed system voltage for a prolonged period following a

system fault. When a transmission or distribution system fault occurs, it can cause a

depression in the system voltage. The voltage may remain at low levels after fault has

cleared and will slowly return to normal after several seconds. The system voltage may

overshoot during the recovery due to capacitor switching and/or load tripping. FIDVR

events are known to cause large amounts of BPS load to trip on undervoltage.

The following is a snapshot for a Phasor Measurement Unit (PMU) for an event in the

lower Rio Grande Valley in 2015. The voltage from the PMU shows the classic FIDVR

profile. 138 kV voltage is depressed to approximately 80% (0.8 per unit) of its nominal

value during the fault. It recovers to normal in approximately 7 seconds, then overshoots

to 106% (1.06 per unit).

-5000

-4000

-3000

-2000

-1000

0

1000

2000

3000

4000

5000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

2015 Average Net Load Ramp by Season and Operating Hour

Avg Net Load Ramp - Winter Avg Net Load Ramp - Spring

Avg Net Load Ramp - Summer Avg Net Load Ramp - Fall


Figure 73 – ERCOT FIDVR Event, Summer 2015

There was no transmission level, or “Consequential”, load loss due to the event. However, the end-use customer load impact was in excess of 400 MW.


Figure 74 – ERCOT FIDVR Load Impact, Summer 2015

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

Valley Load


Appendix A – References

1) NERC 2015 Long-Term Reliability Assessment 2) NERC 2014 Long-Term Reliability Assessment 3) NERC 2015 Summer Assessment 4) NERC 2014 Summer Assessment 5) NERC 2015/2015 Winter Assessment 6) NERC 2014/2015 Winter Assessment 7) NERC 2013/2014 Winter Assessment 8) NERC 2012 State of Reliability Report 9) NERC 2013 State of Reliability Report 10) NERC 2014 State of Reliability Report 11) NERC 2015 State of Reliability Report 12) NERC 2011 Risk Assessment of Reliability Performance Report 13) ERO Event Analysis Process Document

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2015LTRA%20-%20Final%20Report.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014LTRA_ERATTA.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014LTRA_ERATTA.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2015_Summer_Reliability_Assessment.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014SRA.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014SRA.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2015-16%20WRA_Report_Final.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014WRA_final.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2014WRA_final.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2013WRA_Final.pdf

http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/2013WRA_Final.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2012_SOR.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2012_SOR.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2013_SOR_May%2015.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2013_SOR_May%2015.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2014_SOR_Final.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2015%20State%20of%20Reliability.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2011_RARPR_FINAL.pdf

http://www.nerc.com/pa/RAPA/PA/Performance%20Analysis%20DL/2011_RARPR_FINAL.pdf

http://www.nerc.com/pa/rrm/ea/EA%20Program%20Document%20Library/Final_ERO_EA_Process_V2.1.pdf

http://www.nerc.com/pa/rrm/ea/EA%20Program%20Document%20Library/Final_ERO_EA_Process_V2.1.pdf


Appendix B – Disturbance Events Analysis

The 2011-2015 Event Analysis is summarized by category and cause code in the following figures:

Figure B.1 – 2011-2015 ERCOT Events by Category

Figure B.2 – 2011-2015 ERCOT Events by Cause

8354%

5837%

64%

74%

11%

2011-2015 Events

Non-Qualified

Cat 1

Cat 2

Cat 3

Cat 4 & 5

27%

3%

9%

14%

13%

1%

0%

33%

Equipment Failure

Natural Disaster/ForeignInterference

Protection SystemIssues

EMS

Sabotage/Vandalism

Cyber

Unknown

Weather

2011-2015 Event Cause


Texas RE tracks the number of Disturbance Control Standards (DCS) events and recovery time for DCS events as well as DCS events greater than the Most Severe Single Contingency (MSSC) within the region to provide any potential adverse reliability indications. Per the NERC BAL-002 Disturbance Control Standard, a Reportable Disturbance is defined as any event which causes a change in area control error greater than or equal to 80% of the MSSC, or approximately 1,100 MW for the ERCOT region. As part of the Event Analysis process, Texas RE investigates the cause and relative effect on reliability of DCS events within the region. DCS events greater than the MSSC typically do not create a 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 B.3 – DCS Events by Year

0

1

2

3

4

5

2008 2009 2010 2011 2012 2013 2014 2015

DCS Events DCS Events > MSSC

DCS EVENTS


Figure B.4 – DCS and EEA Events by Year

Texas RE also tracks the number of EEA2 events and EEA3 events within the region to provide any potential reliability indicators. EEA events occur infrequently within the region, with the exception of 2011. Two EEA2 events occurred (August 4 and August 24, 2011) due to extreme temperatures combined with generation resource unavailability, as well as one EEA3 event (Southwest Cold Weather Event of February 1-5, 2011). There were no EEA events in 2012, 2013, or 2015.

EEA Date and Level

Minimum Reserve Level During EEA (based on 2,300 MW minimum)

Duration of EEA Event

2/2/2011 – EEA2 447 MW 28.7 hours (total duration)

2/2/2011 – EEA3 447 MW 498 minutes (EEA3 only)

6/27/2011 – EEA1 2,275 MW 85 minutes

8/2/2011 – EEA1 2,123 MW 207 minutes

8/3/2011 – EEA1 1,722 MW 205 minutes

8/4/2011 – EEA2 984 MW 307 minutes

8/5/2011 – EEA1 2,122 MW 175 minutes

8/23/2011 – EEA1 2,160 MW 91 minutes

8/24/2011 – EEA2 1,192 MW 230 minutes

1/6/2014 – EEA2 1,345 MW 140 minutes

Table B.1 – EEA Event Magnitude and Duration

One key item of focus has been the identification of events where generator units trip due to faults or system conditions outside the generator’s protection system zones, or where it is not necessary

0

1

2

3

4

5

6

7

2008 2009 2010 2011 2012 2013 2014 2015

EEA 1 EEA 2 EEA 3

EEA EVENTS


to trip the unit in order to clear the fault condition. Tracking these types of events provides an indication of issues with generator protective relaying, generator excitation system issues, and control system issues. Further analysis is needed to determine common failure modes and share lessons learned to reduce these types of events. The following table shows a detailed history of multi-generator trip events since 2011.


3/14/2011 A large coal unit tripped due to externally induced voltage within the generator excitation system from a nearby 345 kV transmission fault. The induced voltage created a signal which resulted in the operation of the lockout relay.

3/23/2011 A large coal unit initiated a runback and trip due to a Master Fuel Trip following the loss of a large remote combined cycle unit.

3/25/2011 Two wind plants tripped due to a high voltage disturbance when a nearby 345 kV line was re-energized.

4/11/2011 A large coal unit tripped on turbine overspeed and a simple cycle natural gas turbine tripped due to low voltage (<30% of nominal) when two nearby 345 kV circuits on the same tower tripped due to a tower failure.

7/12/2011 Two wind plants tripped on low voltage due to a fault on a nearby 345 kV circuit. Low voltage trip parameters were found set incorrectly on the wind turbines and were subsequently modified.

3/19/2012 A combined cycle CT tripped by rotor earth fault relay due to a nearby 345kV fault.

12/25/2012 A combined cycle natural gas turbine (GT) tripped due to exhaust temperature spread simultaneously with the trip of a large coal unit.

1/8/2013 A wind plant experienced a sympathetic trip simultaneously with a nearby large unit.

1/15/2013 A 2x1 combined cycle unit experienced a runback and trip due to a defective Generator Step-Up (GSU) differential relay when a nearby coal unit tripped.

2/16/2013 Two combined cycle units at the same site tripped during gas valve testing. A defective solenoid on a gas control valve caused the loss of fuel gas to the site.

6/28/2013 Two coal units at the same site experienced run backs and were tripped manually by plant operators due to a reduction in instrument air pressure when one of the plant air compressors went off line.

9/3/2013 Two combined cycle units at the same site tripped approximately one hour apart due to a fuel gas flow issue to the site.

10/8/2013 Two combined cycle units at the same site simultaneously tripped or ran back due to a failed unit auxiliary transformer.

12/13/2013 Two coal units at the same site tripped due to an electrical fault on the plant auxiliary power bus.

3/10/2014 Two 2x1 combined cycle trains at the same site ran back and tripped due to fuel supply issues.



6/11/2014 Three coal units at the same site ran back and tripped due to loss of instrument air.


6/23/2014 Multiple natural gas turbines at the same site tripped due to an electrical fault on the isophase bus of one turbine, which resulted in the trip of the remaining turbines due to the plant configuration at the time.

7/23/2014 Two coal units at the same site tripped due to an electrical fault on the 345 kV substation bus.

8/4/2014 Two coal units at the same site ran back an tripped due to loss of auxiliary cooling water to the instrument air compressors.

1/8/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to maintenance personnel error.

1/15/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to a control system issue.

4/27/2015 Two generation facilities tripped due to a nearby 345 kV system fault combined with a protection system misoperation.

5/25/2015 Two natural gas units at the same site tripped due to a control system issue

6/26/2015 Two 2x1 combined cycle trains at the same site ran back and tripped due to generator instrument transformer issues.

7/29/2015 Two 2x1 combined cycle trains at the same site tripped due to a HV substation protection system misoperation.

12/30/2015 Two 2x1 combined cycle trains at the same site tripped due to a HV substation protection system misoperation.

Table B.2 – Historical Events with Multiple Generator Trips

E. EMS/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, control Bulk Electric System (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.

For 2013-2015, there were 15 loss of EMS/SCADA events reported in the ERCOT region. Events reported in 2015 include the following:

A Transmission Service Provider (TSP) reported that it lost its State Estimator did not run due to a software issue.

A TSP reported that its ICCP data to ERCOT ceased transmitting properly following an ICCP database update.

A TSP reported that it lost monitoring and control functionality due to a network firewall malfunction.

A TSP reported that it lost monitoring and control functionality due to a card failure.

Telemetry Availability

ERCOT telemetry performance criteria states that 92% of all telemetry provided to ERCOT must achieve a quarterly availability of 80%. The following chart shows the telemetry availability metric per the ERCOT telemetry standard. For 2015, the total number of


telemetry points failing the availability metric averaged approximately 2,890 each month, or approximately 3.1% of the total system telemetry points.

Figure B.5 – ERCOT Telemetry System Availability

Telemetry Accuracy

ERCOT telemetry accuracy criteria includes, but is not limited to, the following:

Bus summation: the sum of flows into any telemetered bus should be less than the greater of 5 MW or 5% of the largest Normal line rating at each bus.

Comparison of State Estimator (SE) versus telemetry on major transmission elements: residuals on Transmission Elements over 100 kV are <10% of emergency rating or <10 MW (whichever is greater) on 99.5% of all samples during a month period.

Comparison of State Estimator versus Telemetry on critical bus voltages: the telemetered bus voltage minus state estimator voltage shall be within the greater of 2% or the accuracy of the telemetered voltage measurement involved for at least 95% of samples measured from the kV Residuals.

Comparison of State Estimator versus telemetry on congested transmission elements: differences between the MW telemetry values and MW SE values are <3% of the largest associated Emergency Rating on at least 95% of samples measured in a one month period for predetermined congestion elements. Congested elements are those transmission elements causing 80% of congestion in the latest year for which data is available.

90.00%

91.00%

92.00%

93.00%

94.00%

95.00%

96.00%

97.00%

98.00%

99.00%

100.00%

80000

82000

84000

86000

88000

90000

92000

94000

96000

98000

100000

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

Total Telemetry Points % System Availability


The following figures show the historical trends for these accuracy requirements.

Figure B.6 – Bus Summation Telemetry Accuracy

Figure B.7 – State Estimator vs. Telemetry on Major Transmission Elements

0

50,000

100,000

150,000

200,000

250,000

300,000

0

50

100

150

200

250

300

Ma

y-1

3

Jun

-13

Jul-

13

Au

g-1

3

Sep

-13

Oct

-13

No

v-1

3

De

c-1

3

Jan

-14

Feb

-14

Ma

r-1

4

Ap

r-1

4

Ma

y-1

4

Jun

-14

Jul-

14

Au

g-1

4

Sep

-14

Oct

-14

No

v-1

4

De

c-1

4

Jan

-15

Feb

-15

Ma

r-1

5

Ap

r-1

5

Ma

y-1

5

Jun

-15

Jul-

15

Au

g-1

5

Sep

-15

Oct

-15

No

v-1

5

De

c-1

5

SE Bus Sum Telemetry Accuracy

# Buses w/ Values out of rangeTotal # Buses Outside of defined range for all SE runs3 per. Mov. Avg. (Total # Buses Outside of defined range for all SE runs)

# Buses w/ values out of range

Total # Buses out of range for all SE runs

0

20,000

40,000

60,000

80,000

100,000

120,000

0

200

400

600

800

1,000

1,200

SE Transmission Telemetry Accuracy

# Lines/Xf w/ Values out of range Total # Lines Outside of defined range for all SE runs3 per. Mov. Avg. (Total # Lines Outside of defined range for all SE runs)

# Lines w/ values out of range

Total # Lines out of range for all SE runs


Figure B.8 – State Estimator vs. Telemetry on Congested Transmission Elements

Figure B.9 – State Estimator vs. Telemetry on Top 20 Critical Buses

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

0

10

20

30

40

50

60

70

80

SE MW Residuals on Congested Elements

# Lines/Xf w/ Values out of range Total # Lines Outside of defined range for all SE runs

3 per. Mov. Avg. (Total # Lines Outside of defined range for all SE runs)

# Lines w/ values out of range

Total # Lines out of range for all SE runs

0

400

800

1,200

1,600

2,000

2,400

2,800

3,200

3,600

4,000

0

2

4

6

8

10

12

14

16

18

20

SE Important Bus Voltage Telemetry Accuracy

# Buses w/ Values out of range Total # Buses Outside of defined range for all SE runs

# Buses w/ values out of range

Total # Buses out of range for all SE runs


Appendix C – 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)

2008 258 8,917.8

2009 274 9,312.5

2010 290 9,601.0

2011 310 9,845.6

2012 316 10,049.4

2013 371 13,285.6

2014 397 14,193.2

2015 413 14,832.0

Table C.1 – 2008-2015 End of Year Circuit Data

Automatic Non-Automatic

Outage Information

Count Duration (hours)

Count Duration (hours)

2010 195 1,090.0 548 31,299.8

2011 279 1,908.6 787 46,712.4

2012 230 682.6 516 36,295.1

2013 198 1,936.5 723 64,945.8

2014 276 2,917.3 728 71,093.2

2015 477 10,806.9 404 53,147.6

5-Yr Average 236 1,707.0 660 50,069.3

Table C.2 – 2010-2015 345 kV Automatic and Non-Automatic Outage Data

Automatic Outage Data

For 2010-2015 for the 345 kV system, Failed AC Substation Equipment represented 16% of sustained outage cause and 25% of sustained outage duration. For 2010-2015 for the 345 kV system, Failed AC Circuit Equipment represented 8% of sustained outage cause and 51% of sustained outage duration. Ninety-two percent of Failed AC Circuit Equipment outages lasted two hours or more, with an average duration of 201 hours.

2015 had a significant increase in lightning-related momentary and sustained outages on the 345 kV system.


Figure C.1 – 345 kV Automatic Outages by Month

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

0

10

20

30

40

50

60

70

80

90

100


2010 2011 2012 2013 2014 2015Automatic Outages by Month

0

500

1000

1500

2000

2500


2010 2011 2012 2013 2014 2016Automatic Outage Duration by Month (Hours)


Figure C.3 – 345 kV Momentary Outages by Cause

Figure C.4 – 345 kV Sustained Outages by Cause

0

20

40

60

80

100

120

1402010 2011 2012 2013 2014 2015

Momentary Outages by Cause

0

10

20

30

40

50

602010 2011 2012 2013 2014 2015

Sustained Outages by Cause


Figure C.5 – 345 kV Sustained Outage Duration (hours) by Cause

Common and Dependent Mode Outage Data

For 2010-2015 combined, Dependent Mode outages and Common Mode outages on the 345 kV represented 10.4% of all momentary outages, 28.7% of all sustained outages and 48.7% of sustained outage duration.

For 2010-2015 combined, Failed AC Circuit equipment represented 8% of the Common Mode and Dependent Mode outages, but resulted in over 78% of the Common Mode and Dependent Mode outage duration.

0

1000

2000

3000

4000

5000

60002010 2011 2012 2013 2014 2015

Sustained Outage Duration (Hours) by Cause


Figure C.6 – 345 kV Average AC Circuit Momentary Automatic Outage Mode

Figure C.7 – 345 kV Average AC Circuit Sustained Automatic Outage Mode

90%

1%

3%

6%1%

2010-2015 Avg AC Circuit Momentary Automatic Outage Mode

Single Mode


Dependent Mode

Common Mode


71%

2%

17%

9%1%

2010-2015 Avg AC Circuit Sustained Automatic Outage Mode

Single Mode


Dependent Mode

Common Mode



Figure C.8 – 345 kV Momentary Outage Modes Comparison

Figure C.9 – 345 kV Sustained Outage Modes Comparison

The following charts show the 2010-2015 Dependent Mode outages and Common Mode outage data broken down by cause, duration, and event type.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%


Dependent Mode Common Mode Common ModeInitiating

2010 2011 2012 2013 2014 2015

Momentary Outage Modes Comparison

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%


Dependent Mode Common Mode Common ModeInitiating

2010 2011 2012 2013 2014 2015

Sustained Outage Modes Comparison


Figure C.10 – 2010-2015 345 kV Common Mode/Dependent Mode Outages by Cause

Figure C.11 – 2010-2015 345 kV Common Mode/Dependent Mode Outages by Duration

8%

6%

0%

6%

2%

3%

0%

30%0%

14%

8%

0%

0%

1%

15%

4%4%

Weather, excluding lightning

Lightning

Environmental

Contamination


Fire

Vandalism, Terrorism, or Malicious Acts

Failed AC Substation Equipment

Failed AC/DC Terminal Equipment

Failed Protection System Equipment

Failed AC Circuit Equipment

Failed DC Circuit Equipment

Vegetation


Human Error

Unknown

Other

2010-2015 Common Mode and Dependent Mode Outages by Cause

0%0%

0%

1%3%

0%0% 10%

0%1%

78%

0%0%

3% 2%0%

0%

Weather, excluding lightning

Lightning

Environmental

Contamination


Fire

Vandalism, Terrorism, or Malicious Acts

Failed AC Substation Equipment

Failed AC/DC Terminal Equipment

Failed Protection System Equipment

Failed AC Circuit Equipment

Failed DC Circuit Equipment

Vegetation


Human Error

Unknown

Other

2010-2015 Common Mode and Dependent Mode Outages by Duration


Figure C.12 – 2010-2015 AC Circuit Outage Mode (> 200 kV)

TADS data from 2010-2015 shows a total of 39 events (3.6%) affecting two of more elements on common structures.

2%

22%

20%

24%

16%

0%

0%

13%

3%5- Single Bus fault

11 - Single Elem outage

13-Two or more Elem outage

31-Two or more Elem on Common Str

49-Outages not covered by 5-31

60-Breaker failure

61-Protection Sys failure

62-Protection Sys misoperation

90-Outages not covered by 60-62

2010-2015 Common/Dependent Mode Outages by Event Type


Appendix D – Generation Availability Analysis

GADS provides and also permits comparison of unit performance by fuel type. A summary of key performance metrics for the entire ERCOT generation fleet is provided in the following figures.

Figure D.1 – 2015 Net and Gross Capacity Factors for the ERCOT Generation Fleet

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


GG - Gas Nuclear GasTurbine/Jet

Engine(Simple Cycle

Operation)

CombinedCycle Block

CC GT units CC steamunits

Net Capacity Factor (NCF) Gross Capacity Factor (GCF)


Figure D.2 – 2015 Scheduled/Forced Outage Factors for the ERCOT Generators

For 2012 through 2015, there was an average of 217.7 immediate de-rate events each month, with an average duration of 10,513 hours each month and an average capacity of 91.7 MW per derate event. For 2012 through 2015, there was an average of 157.4 immediate forced outage events each month, with an average duration of 7,739 hours each month and an average capacity of 220 MW per outage event.

For 2012 through 2015, immediate de-rates, immediate forced outages, and startup failures increased during peak periods and during ramp periods prior to the winter and summer peak periods.

There was an increase is the unavailable MWH from forced outages and startup failures in the winter of 2015 compared to previous years, however, it does not appear to be related to any significant cold weather issues.

0%

2%

4%

6%

8%

10%

12%

14%

16%


GG - Gas Nuclear Gas Turbine/JetEngine (Simple

CycleOperation)

CombinedCycle Block

CC GT units CC steam units

Scheduled Outage Factor (SOF) Forced Outage Factor (FOF)


Figure D.3 – 2013-2015 Lost MWH from Forced Outages

Figure D.4 – 2012-2015 Cumulative Events by Operating Hour - Summer

0

2,000,000

4,000,000

6,000,000

8,000,000

10,000,000

12,000,000

Summer Winter Shoulder

2013-2015 Forced Outage & Startup Failure MWH2013 2014 2015

0

50

100

150

200

250

300

350

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Cumulative Events By Hour of the Day - Summer

Derates (D1) Outages (U1) Startup Failures


Figure D.5 – 2012-2015 Cumulative Events by Operating Hour - Winter

0

20

40

60

80

100

120

140

160

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Cumulative Events By Hour of the Day - Winter

Derates (D1) Outages (U1) Startup Failures


Appendix E – Demand Response Historical Data

The following table provides a list of load resource deployments greater than 100 MW since

2010.


5/15/2010 1,112 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple generators. Deployment time: 36 minutes

6/23/2010 246 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple generators. An additional 571 MW were deployed by System Operators to assist with the frequency recovery. Deployment time: 40 minutes

8/20/2010 1,150 MW of Load Resources deployed manually by System Operators to assist with the frequency recovery due to the trip of multiple generators. Deployment time: 13 minutes

11/3/2010 680 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple generators. An additional 603 MW were deployed by System Operators to assist with the frequency recovery. Deployment time: 47 minutes

2/2/2011 887 MW of Load Resources and 468 MW of EILS were deployed manually by System Operators due to the loss of multiple generators. System Operators issued directives to shed 4,000 MW of firm load due to EEA3 conditions. Deployment time: 33.5 hours for LR and EILS, 7.4 hours for firm load shed

3/23/2011 393 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple generators. Deployment time: 37 minutes

5/19/2011 113 MW of Load Resources deployed automatically by underfrequency relay due to the trip of a large generator. Deployment time: 9 minutes

8/4/2011 881 MW of Load Resources and 514 MW of EILS were deployed by System Operators due to EEA2 conditions. Deployment time: 3 hours 4 minutes

8/24/2011 634 MW of Load Resources were deployed by System Operators due to EEA2 conditions. Deployment time: 2 hours 13 minutes


7/10/2012 195 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple large generators. Deployment time: 14 minutes

7/30/2012 317 MW of Load Resources deployed automatically by underfrequency relay due to the trip of multiple large generators. Deployment time: 14 minutes





1/6/2014 1,085 MW of Load Resources and 607 MW of ERS deployed manually by System Operators due to EEA2 conditions. Deployment time: 56 minutes

Table E.1 – Demand Response Deployments since 1/1/2010


Appendix F – Frequency Control Performance Analysis

The following graphs show historical regulation exhaustion rates for both Regulation-Up and Regulation-Down. For the purpose of these charts, the regulation exhaustion rate was based on the percentage of one-minute intervals where available regulation was less than 20 MW.

Figure F.1 – Comparison of 2011-2015 Regulation-Up Exhaustion Rates by Month

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%


2011 Up-Regulation Exhaustion Rate 2012 Up-Regulation Exhaustion Rate

2013 Up-Regulation Exhaustion Rate 2014 Up-Regulation Exhaustion Rate

2015 Up-Regulation Exhaustion Rate


Figure F.2 – Comparison of 2011-2015 Regulation-Down Exhaustion Rates by Month

Figure F.3 – Comparison of 2012-2015 Reg-Up Exhaustion Rates by Operating Hour

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%


2011 Down-Regulation Exhaustion Rate 2012 Down-Regulation Exhaustion Rate2013 Down-Regulation Exhaustion Rate 2014 Down-Regulation Exhaustin Rate2015 Down-Regulation Exhaustion Rate

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

2012 UpReg Exhaustion Rate 2013 UpReg Exhaustion Rate 2014 UpReg Exhaustion Rate 2015 UpReg Exhaustion Rate


Figure F.4 – Comparison of 2012-2015 Reg-Down Exhaustion Rates by Operating Hour

Figure F.5 – Comparison of 2012-2015 Average Regulation Deployed by Operating Hour

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

2012 DownReg Exhaustion Rate 2013 DownReg Exhaustion Rate2014 DownReg Exhaustion Rate 2015 DownReg Exhaustion Rate

-200

-150

-100

-50

0

50

100

150

200

250

300

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

2012 Avg Reg Deployed 2013 Avg Reg Deployed 2014 Avg Reg Deployed 2015 Avg Reg Deployed


Non-Spin Reserve Service (NSRS): ERCOT Protocols require the QSE to update the telemetered ancillary service schedule for NSRS to reflect the deployment amount within 20 minutes, per NP 8.1.1.4.3(3)(a). Within 25 minutes following a deployment instruction, an Off-Line Generation Resource must be On-Line with an Energy Offer Curve and the telemetered net generation must be greater than or equal to the Resource’s telemetered low stability limit (LSL) multiplied by P1 where P1 is defined in the “ERCOT and QSE Operations Business Practices During the Operating Hour” per NP 8.1.1.4.3(3)(b). The following figure shows the trend in NSRS deployments, total hours deployed, and Non-Spin failures.

Figure F.6 – Non-Spin Reserve Service Deployment History

Hourly Ancillary Supply Responsibility: ERCOT Protocols require the QSEs to have no more than three hours during an Operating Day and no more than 74 hours during a month that contains Current Operating Plan (COP) Ancillary Service Resource Responsibility validation failures per NP 8.1.2(2). The following figure shows the trend in hourly ancillary service supply failures. For 2015, QSEs averaged 98 hourly ancillary service supply failures each month, compared to an average of 139 per month in 2014. The following figure shows this notable downward trend in ancillary service supply failures.

0

10

20

30

40

50

60

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

# Non-Spin Gen Deployments Total Hours Deployed # Non-Spin Failures


Figure F.6 – Hourly Ancillary Service Supply Responsibility Failures

Responsive Reserve Service (RRS) Deployments: ERCOT Protocols require the QSE to update the telemetered Ancillary Service Schedule(s) for Responsive Reserve to reflect the deployment amount within one minute per NP 8.1.1.4.2(1)(a). The following figure shows the trend in RRS total hours deployed and resource schedule update failures.

Figure F.7 – Responsive Reserve Service Deployment History

13.6

13.0

17.117.7

14.9

11.9

13.7

11.2

5.76.3

6.8

12.7

5.2

3.9 4.25.0 5.2

3.9

0.0

5.0

10.0

15.0

20.0

25.0

30.0

0

200

400

600

800

1000

1200

2011Q3

2011Q4

2012Q1

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

Hourly Ancillary Service Supply Responsibility Failures Avg Hourly Failures/QSE

11

13

16

2

1

3

1

2

1 1

0

8

5

0 0.00

2.50

5.00

7.50

10.00

12.50

15.00

17.50

20.00

0

5

10

15

20

25

30

35

40

2012Q2

2012Q3

2012Q4

2013Q1

2013Q2

2013Q3

2013Q4

2014Q1

2014Q2

2014Q3

2014Q4

2015Q1

2015Q2

2015Q3

2015Q4

RRS - Total Hours Deployed # Resource Schedule Update Failures


Appendix G – Protection System Misoperations Analysis

The following graphs show historical protection system misoperation data for 2011-2015, broken down by voltage, misoperation category, relay system type, equipment protected, and cause.

Figure G.1 – Protection System Misoperation Data for 2011-2015 by Voltage

Figure G.2 – Protection System Misoperation Data for 2011-2015 by Category

73%

22%

5%

138

345

<100

By Voltage

4% 0% 2%

0%

52%

42%

Failure to Trip - DuringFault

Failure to Trip - Otherthan Fault

Slow trip - During Fault

Slow trip -Other thanFault

Unnecessary Trip -During Fault

Unnecessary Trip -Other than Fault

By Category


Figure G.3 – Protection System Misoperation Data for 2011-2015 by Relay System Type

Figure G.4 – Protection System Misoperation Data for 2011-2015 by Equipment Protected

20%

3%

72%

5%

Electromechanical

Solid State

Micro Processor

N/A

By Relay System Type

59%

11%

9%

1%

5%3%

0%11% 0%

1%

Line

Transformer

Generator

Shunt Capacitor

Bus

Shunt Reactor/Inductor

Dynamic Var Systems

Breaker

Series Reactor/Inductor

Series Capacitor

By Equipment Protected


Figure G.5 – Protection System Misoperation Data for 2011-2015 by Cause

Figure G.6 – Protection System Misoperations 2011-2015 Relay Failures by System Type

8%

10%

10%

4%

33%

3%

5%

19%

5%3%

AC system

As-left personnel error

Communication failures

DC system

Incorrect settings

Logic errors

Design errors

Relayfailures/malfunctionsUnknown/unexplainable

Other/explainable

By Cause

39%

8%

53% Electromechanical

Solid State

Micro Processor

Relay Failures by System Type


Figure G.7 – Protection System Misoperations by Year by Cause 2011-2015

Figure G.8 – Protection System Misoperations by Misoperation Type 2011-2015

0

10

20

30

40

50

60

70

80

90

2011 2012 2013 2014 2015

0

20

40

60

80

100

120

Failure to Trip - DuringFault

Slow trip - During Fault Unnecessary Trip -During Fault

Unnecessary Trip - Otherthan Fault

2011 2012 2013 2014 2015


Unnecessary trips during a fault on transmission lines remain the main type of misoperation, accounting for 35% of the total number of misoperations.

Figure G.9 – Protection System Misoperation Data by Cause and Element

These figures indicate that misoperations due to incorrect settings, logic errors, design errors, and communication errors caused an unnecessary trip during a fault condition over 70% of the time. This is in contrast to the percentage of relay failures (71%) that caused misoperations when no fault was present. Also, 83% of the misoperations associated with generators were unnecessary trips when no fault was present.

3049

95

11

204

20

13 9614 29

9

17

0

2

28

10

423

56

15

6

0

5

35

1

0 20

14

36

0

8

11

5

1 13

2

15

16

3

827

84 24

1615

2 0 0 2 30 0 0 1

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Shunt Reactor/Inductor Breaker Bus Shunt Capacitor Generator Transformer Line


Figure G.10 – Protection System Misoperation Data by Cause 2011-2015

Figure G.11 – Protection System Misoperations by Category 2011-2015

2 2 3

5

5 2 0 8 1 01 3 4

1

41

02 0 2

31

5368

3

192 27 16

40 611

39

2723

19

72 134

123 17 25

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Unnecessary Trip - Other than Fault Unnecessary Trip - During Fault

Slow trip - During Fault Failure to Trip - During Fault

15 4 2 04

2 015 1

01

0

10

329

40

15

315

35

5

142

56

69

528

31

18

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Unnecessary Trip - Other than Fault Unnecessary Trip - During Fault

Slow trip - During Fault Failure to Trip - During Fault

2013 Assessment of Reliability Performance Services/2015 Texas … · The goals of the 2015...

Documents

Transcript of 2013 Assessment of Reliability Performance Services/2015 Texas … · The goals of the 2015...