industry experience following the 2011 pacific … measurement... · the nerc synchronized...

NERC | Report Title | Report Date I

Phase Angle Monitoring: Industry Experience Following the 2011 Pacific Southwest Outage Recommendation 27 Technical Reference Document

June 2016

NERC | Phase Angle Monitoring | June 2016 ii

Table of Contents

Acknowledgments ..................................................................................................................................................... iv

Preface ........................................................................................................................................................................ v

Executive Summary ................................................................................................................................................... vi

Recommendations .................................................................................................................................................... vii

Introduction ................................................................................................................................................................1

The Meaning of a Synchrophasor Phase Angle ......................................................................................................1

Fundamental Drivers for Phase Angle Differences .................................................................................................2

Fundamental Need for Synchrocheck Relay Thresholds ........................................................................................2

Southwest Outage Recommendation 27 ...................................................................................................................4

Overview of Outage and Angle Implications ..........................................................................................................4

2011 Southwest Outage Report Finding 27 ........................................................................................................5

2011 Southwest Outage Report Recommendation 27 .......................................................................................5

Phase Angle Difference Monitoring – Synchro-Check Awareness .............................................................................6

Integration with EMS Network Applications ..........................................................................................................6

Synchrophasor-Based Tools & Real-Time Awareness ......................................................................................... 11

Benchmarking PMU and SCADA Data .................................................................................................................. 12

Time Alignment ................................................................................................................................................ 13

Synchrophasor Data Validation Options .......................................................................................................... 13

Signal Mapping Guidelines ............................................................................................................................... 13

PMU Data Quality ............................................................................................................................................. 14

EMS and PMU Data Quality Flags .................................................................................................................... 15

Benchmarking Calculations .............................................................................................................................. 15

Mitigation and Operating Procedures for Line Restoration ................................................................................ 19

Actions for Excessive Phase Angle Differences ................................................................................................ 19

Correlating Phase Angle with System Conditions ................................................................................................... 20

Phase Angle and Real Power Correlation ............................................................................................................ 20

Identification of Key (Optimal) Angle Differences ................................................................................................... 25

Major Transmission Interfaces or Transfer Paths ................................................................................................ 25

WECC Intertie Paths ......................................................................................................................................... 25

Oscillatory Stability Analysis ................................................................................................................................ 26

Phase Angle Visualization of Operating Boundaries ............................................................................................ 30

Voltage Stability and Phase Angle ....................................................................................................................... 30

Linking Phase Angles with System Studies .............................................................................................................. 32

Table of Contents

NERC | Phase Angle Monitoring | June 2016 iii

Definition of Safe & Alert Operating States ......................................................................................................... 32

Defining Inter-Area Stability Limits Based on Phase Angle ................................................................................. 32

References ............................................................................................................................................................... 35

Appendix A – Utility Practices ................................................................................................................................. 36

Peak Reliability Coordinator (Peak Reliability) .................................................................................................... 36

California ISO (CAISO) .......................................................................................................................................... 36

Arizona Public Service (APS) ................................................................................................................................ 36

Salt River Project (SRP) ........................................................................................................................................ 37

San Diego Gas & Electric (SDG&E) ....................................................................................................................... 37

NERC | Phase Angle Monitoring | June 2016 iv

Acknowledgments The NERC Synchronized Measurement Subcommittee (SMS) gratefully acknowledges the invaluable assistance of the following industry experts in the preparation of this technical reference paper:

• Aftab Alam (California ISO)

• Tony Faris (Bonneville Power Administration)

• Hassan Ghoudjehbaklou (San Diego Gas & Electric)

• Dmitry Kosterev (Bonneville Power Administration)

• Naim Logic (Salt River Project)

• Ken Martin (Electric Power Group)

• Tariq Rahman (San Diego Gas & Electric)

• Alison Silverstein (North American Synchrophasor Initiative)

• Jeff Sundin (Arizona Public Service)

• Dan Trudnowski (Montana Tech)

• Marianna Vaiman (V&R Energy)

• Hongming Zhang (Peak Reliability)

NERC | Phase Angle Monitoring | June 2016 v

Preface The North American Electric Reliability Corporation (NERC) is a not-for-profit international regulatory authority whose mission is to assure the reliability of the bulk power system (BPS) in North America. NERC develops and enforces Reliability Standards; annually assesses seasonal and long-term reliability; monitors the BPS through system awareness; and educates, trains, and certifies industry personnel. NERC’s area of responsibility spans the continental United States, Canada, and the northern portion of Baja California, Mexico. NERC is the electric reliability organization (ERO) for North America, subject to oversight by the Federal Energy Regulatory Commission (FERC) and governmental authorities in Canada. NERC’s jurisdiction includes users, owners, and operators of the BPS, which serves more than 334 million people. The North American BPS is divided into eight Regional Entity (RE) boundaries, as shown in the map and corresponding table below.

The Regional boundaries in this map are approximate. The highlighted area between SPP and SERC denotes overlap as some load-serving entities participate in one Region while associated transmission owners/operators participate in another.

FRCC Florida Reliability Coordinating Council

MRO Midwest Reliability Organization

NPCC Northeast Power Coordinating Council RF ReliabilityFirst

SERC SERC Reliability Corporation

SPP-RE Southwest Power Pool Regional Entity TRE Texas Reliability Entity

WECC Western Electricity Coordinating Council

NERC | Phase Angle Monitoring | June 2016 vi

Executive Summary The purpose of this report is to provide an update on industry practices for phase angle difference monitoring and limit determination, phase angle-related applications, and operating experience using innovative software tools. Under direction of the FERC/NERC 2011 Pacific Southwest Outage Reporting Finding and Recommendation 27, the paper reports on the current state of how system operators and operations engineers have expanded angle monitoring capability. Findings, recommendations, and examples provided in this report are preliminary yet specific to guide phase angle monitoring implementation strategies, work scopes, and technical solutions. The report reaches the following conclusions and findings:

1. Transmission Operators and Reliability Coordinators are monitoring phase angle differences and comparing those real-time angle differences to synchrocheck relay settings. This comparison is taking place in the State Estimator (SE) and Real-Time Contingency Analysis (RTCA), as well as in advanced supervisory applications.

2. PMU phase angle signals are being baselined and benchmarked against SCADA measurements and SE solutions.

3. Phase angle is strongly correlated to active power transfer and system topology, and is analyzed across WECC Paths in OSIsoft PI system. Continued efforts will focus on using phase angle difference as an additional system stress indicator in conjunction with Path MW flow.

4. PMU phase angle differences are a useful quantity to monitoring immediately following an unplanned and/or unstudied forced outage event. The real-time angle difference provides the system operator with an immediate awareness of system strength and stress.

5. Static or dynamic Path Limits are conventionally based on MW flow. Using phase angle difference may be an additional indicator that provides system operators with better awareness of system stress, particularly for stressed outage conditions where MW flow changes minimally but phase angle increases due to Path separation or other outages.

6. Wide area oscillation modes and angle separation correlation are presented. The oscillatory modes of the Western Interconnection are well understood and correlating modal characteristics to phase angle differences can provide additional situational awareness and engineering insights to stressed operating conditions.

7. Baselining phase angle differences to determine ‘Normal’ or ‘Abnormal’ operating states is being explored using data mining techniques. “Big data” analytics should be applied with engineering judgment and integrated with historical data, operational experience, study results, and event validation analysis.

8. The phase angle monitoring practices of utilities in the Pacific Southwest are presented to provide a status update approximately 5 years after the 2011 Pacific Southwest Outage.

NERC | Phase Angle Monitoring | June 2016 vii

Recommendations The recommendations listed below are based on the assessment of phase angle monitoring, limit calculation, and alarming techniques. While the report focuses on practices in the Western Interconnection following the Pacific Southwest Outage, it was determined that these recommendations apply to all interconnections and entities performing the respective functions. The concepts of phase angle monitoring and alarming using both SCADA-based and time synchronized measurements for improved situational awareness apply to any Bulk-Power System.

1. The contingency risk of interest is the outage of a transmission circuit and the phase angle difference across the out-of-service terminals of that line exceeding synchrocheck relay limits. Post-contingency angle differences should be monitored in real-time. The Planning Coordinator and/or Reliability Coordinator should identify key transmission circuits for which this monitoring is required. It is recommended that awareness of synchrocheck relay limit exceedances be provided to system operators for EHV transmission circuits, where applicable, with nominal voltage greater than or equal to 345 kV.

2. Phase angle differences for potential contingency conditions should be monitored in real-time and compared against synchrocheck relay settings, if applicable, for all EHV transmission circuits using Real-Time Contingency Analysis (RTCA) tools. Any N-1 or credible N-2 or N-1-1 exceedances of these limits should be provided to the system operator for advanced notice of potential line restoration issues.

3. Wide-area phase angle difference monitoring provides an additional layer of situational awareness for system operators, and wide-area limits based on known risks such as transient stability, voltage stability, small signal stability, or overloads can effectively be developed based on operations studies or advanced online applications. Utilities should consider extracting phase angle difference values during system studies for stability risks, in conjunction with conventional MW flow limits.

4. Line-based phase angle difference monitoring and comparison against known synchrocheck limits is not presently a universally adopted operating practice. It is recommended that the NERC Synchronized Measurement Subcommittee (SMS), in coordination with the NERC Operating Committee (OC), explore how this practice could be used or more widely adopted by the industry.

5. In the Western Interconnection, phase angle difference is correlated to oscillatory stability issues particularly during high transfer conditions. Tools such as Mode Meter, Oscillation Detection, and Phase Angle Difference (PAD) tools provide advanced analytical capabilities to detect any oscillatory issues linked with phase angle stress. It is recommended that utilities continue pursuing advancements in these tools for further situational awareness.

NERC | Phase Angle Monitoring | June 2016 1

Introduction The importance of monitoring phase angle differences has been highlighted in the report for the causes and recommendations of the August 2003 Northeast Blackout and the September 2011 Pacific Southwest Outage. Specifically, the August 2003 Northeast Blackout report states that operators in several of the events leading to the blackout were unaware of the vulnerability of the system to the next contingency, in part because they had insufficient situational awareness and no operator monitoring of stability measures like power transfer angle. The report on the 2003 blackout recommended that the industry “review phase-angle restrictions that can prevent reclosing of major interconnections during system emergencies” [1]. The 2011 Pacific Southwest Outage report highlighted the lack of tools at the time to determine the phase angle difference between the two terminals of a line after the line tripped, and recommended that grid operators have “1) the tools necessary to determine phase angle differences following the loss of lines, and 2) mitigation and operating plans for reclosing lines with large phase angle differences.” It also recommended that operators should be trained to effectively respond to phase angle differences [2]. Synchrophasor technology improves the capability of grid operators and operations engineers to visualize and monitor a wide-area view of the bulk power system. Phasor Measurements Units (PMUs) provide direct measurement of system voltage and current phasors (magnitude and phase angle) and frequency. This time-synchronized data can be used for early detection of system disturbances, assessing and maintaining stability following a major event, and alarming system operators to view precise real-time data within seconds of a system event. The phase angle difference between buses’ voltage phasors, as measured by PMUs on the bulk power system, is an indication of system stress and stability. An angle difference within a predetermined limit is acceptable but needs to be monitored closely for early warnings. An increasing phase angle difference can be a serious problem when the deviation gets large enough to cause instability either pre- or post-contingency. This report identifies industry practices for utilizing synchronized phase angles and phase angle differences for monitoring system stress and providing situational awareness to grid operators. The Meaning of a Synchrophasor Phase Angle A Phasor Measurement Unit (PMU) measures voltage and current signals and estimates a time-synchronized phasor representation (magnitude and phase angle) of these electrical quantities. These voltage and current phasors are referred to as synchrophasors. The synchrophasor phase angle is defined as:

A PMU estimates synchrophasor phase angle based on the nominal system frequency synchronized to UTC (Global Positioning System (GPS)). The PMU estimates the sinusoidal component of the AC waveform from a voltage or current input. Using a time input, usually from a GPS source, it constructs a synchronized reference cosine waveform at the nominal system frequency (60 Hz) such that positive peak is at a UTC second rollover. The synchrophasor phase angle is the phase difference between these signals at the given reporting time.

PMUs also report analog quantities such as derived active and reactive power and digital quantities such as breaker status. Frequency and rate-of-change-of-frequency (ROCOF) are derived from a phasor signal (generally voltage phasor) estimated by the PMU. Figure 1 depicts a phasor representation of a sinusoidal waveform. Synchronization to a common time reference for time = 0 is performed using a reference waveform as described above.

Introduction


Figure 1: Sinusoidal Waveform and Phasor Representation

[Source: Electric Power Group] Fundamental Drivers for Phase Angle Differences Phase angle is fundamentally linked to power transfer and system topology. Consider the equation for active power flow, 𝑃𝑃𝑠𝑠𝑠𝑠, across a short high voltage transmission line1

𝑃𝑃𝑠𝑠𝑠𝑠 =𝑉𝑉𝑠𝑠𝑉𝑉𝑠𝑠𝑋𝑋𝑙𝑙

sin(𝜃𝜃𝑠𝑠𝑠𝑠).

𝑉𝑉𝑠𝑠 and 𝑉𝑉𝑠𝑠 are the sending and receiving end voltage magnitudes, respectively, 𝑋𝑋𝑙𝑙 is the line impedance, and 𝜃𝜃𝑠𝑠𝑠𝑠 is the phase angle difference between bus voltage phasors at each terminal of the line. This equation can be rewritten to show the relationship with respect to phase angle difference.

𝜃𝜃𝑠𝑠𝑠𝑠 = sin−1 �𝑃𝑃𝑠𝑠𝑠𝑠𝑋𝑋𝑙𝑙𝑉𝑉𝑠𝑠𝑉𝑉𝑠𝑠

�

This clearly shows that phase angle difference is directly related to power flow, impedance, and voltage magnitudes. However, voltage magnitude is held relatively constant within reasonable operating schedules near 1.0 pu (or higher for EHV transmission). Therefore, phase angle difference is primary driven by power flow and electrical impedance. Phase angle can change drastically for major topology changes; hence, phase angle differences being a strong indicator of system topology conditions and switching events. Power flows can vary over a wide range and also have a relatively significant impact on phase angle differences. Power flows from a higher voltage phase angle to a lower voltage phase angle. A large phase angle difference between the source and the sink or a pair of buses indicates greater power flow between those points. This implies higher static stress across that interface and closer proximity to instability. The relationships described here and uses of phase angle differences in power system applications are the focuses of this paper. Fundamental Need for Synchrocheck Relay Thresholds When a transmission line is removed from service (forced or planned), the phase angle difference between its terminals generally increases because the electrical impedance between these two points increases. Substantially large phase angle differences can lead to system instability and loss of synchronism for generating resources. Power swings from reclosing or restoring lines with a large phase angle, and subsequent oscillations, could lead to system instability or collapse.

1 This equation is only valid when XL >> RL, where XL is the line series reactance and RL is the line series resistance.

Introduction


Reclosing a transmission circuit with large phase angle difference across its terminals near generators can result in a large transient torque on the shaft of the generator. Large phase angle difference is directly related to the generator rotor being out of phase with the bulk power system; therefore, the transient torque is generated to move the rotor shaft position back into phase with the system. Significantly large transient torques can cause instantaneous damage or cumulative fatigue to the generator shaft, and deteriorate the life of the machine [3]. To prevent the harmful effects of closing transmission lines on the transmission system with a high phase angle difference, many utilities use synchrocheck relay schemes on their bulk power transmission lines. Such schemes measure the voltage magnitude difference, frequency slip and phase angle difference between the voltages and supervise against a pre-determined setting prior to restoring a transmission line.


Southwest Outage Recommendation 27 Overview of Outage and Angle Implications On the afternoon of September 8, 2011, an 11-minute system disturbance occurred in the Pacific Southwest, leading to cascading outages and approximately 2.7 million customers without power. The outages affected parts of Arizona, Southern California, and Baja California, Mexico. All of the San Diego area lost power, with nearly 1.5 million customers losing power, some for up to 12 hours. During the event, system stress steadily increased in seven distinct steps during the sequence of events. This is illustrated in Figure 2, showing the line MVA/current loading for South of SONGS (San Onofre Nuclear Generating Station) interface [2]. The initiating event was the forced outage of the Hassayampa-North Gila 500kV line. This led to tripping of sub-transmission transformers and generation, operation of a response-based Remedial Action Scheme (RAS), and subsequently the separation of the South of SONGS interface when it exceeded its phase current limit.

Figure 2: Seven Phases of the 2011 Pacific Southwest Outage

The public report did not provide a detailed analysis of phase angle values during this event; however, monitoring phase angle was a key finding and recommendation from this outage analysis. Finding and Recommendation 27 from the outage report are described below.

Southwest Outage Recommendation 27


2011 Southwest Outage Report Finding 27 Report Finding 27 focused on the phase angle separation following loss of major transmission lines, specifically the Hassayampa-North Gila 500kV line. In particular, when the line was tripped out of service, system stress drove the phase angle difference between the two terminals of the line to larger than the synchrocheck relay setting. Therefore, the line was incapable of returning to service if switched in due to this large phase angle. The report highlighted the need for monitoring phase angle differences for the purposes of returning to service transmission elements in a coordinated, efficient manner. Finding 27 states:

“Phase Angle Difference Following Loss of Transmission Line: “A TOP did not have tools in place to determine the phase angle difference between the two terminals of its 500 kV line after the line tripped. Yet, it informed the RC and another TOP that the line would be restored quickly, when, in fact, this could not have been accomplished.”

2011 Southwest Outage Report Recommendation 27 Based on this finding, the FERC-NERC report highlighted two major areas of focus for Transmission Operators (TOPs) related to restoration of transmission lines: situational awareness tools, and mitigation and operating plans. Recommendation 27 states:

“TOPs should have: (1) the tools necessary to determine phase angle differences following the loss of lines; and (2) mitigation and operating plans for reclosing lines with large phase angle differences. TOPs should also train operators to effectively respond to phase angle differences. These plans should be developed based on the seasonal and next-day contingency analyses that address the angular differences across opened system elements.”


Phase Angle Difference Monitoring – Synchro-Check Awareness Integration with EMS Network Applications Commercial EMS software applications provide Transmission Operators (TOP), Balancing Authorities (BA), and Reliability Coordinators (RC) with the capability of defining node or bus angle pairs and comparing calculated phase angle differences against defined settings such as synchrocheck relay settings. Monitoring these calculations in the network model and alarming operators on angular separation exceedance in real-time (i.e., SCADA and State Estimator (SE) base cases) and potential post-contingency states (i.e., Real-Time Contingency Analysis (RTCA)) provides system operators with near-real time awareness of phase angle differences. Comparing these limits to the synchrocheck relay limits identifies any lines that could be unable to be restored after tripping. According to the FERC-NERC investigation report and NERC Reliability Standard requirements, system operators shall be positioned to proactively operate the system in a secure N-1 state during normal system conditions and to restore the system to a secure N-1 state as soon as possible, but no longer than 30 minutes. By using RTCA tools that usually run every 5 minutes or less, system operators gain near real-time awareness of phase angle difference exceedances under pre-defined single contingency (N-1) or credible N-2 or N-1-1 contingency operating conditions. There are multiple ways for system operators to obtain awareness of system angular separation conditions using State Estimator (SE) tools. At each state estimate solution, bus voltage phase angles are estimated and the phase angle difference across defined transmission circuits can be reported as a branch record as shown in Figure 3.

Figure 3: Reported Phase Angle Differences over Threshold in SE

[Source: Peak Reliability] The derived phase angle separation across these branch elements can then be compared against a user-defined limit. The SE tools allow users to define a group of node pairs (NP) with pre-assigned Normal and Emergency Limits. These NP limits can be directly linked to the limits defined in the synchrocheck relays for those respective lines2. Figure 4 shows a screenshot of the definition of NP limits for the North Gila-Imperial Valley 500kV

2 Synchrocheck relay settings change relatively infrequently; therefore, static limits are generally acceptable for these types of alarms.

Phase Angle Difference Monitoring – Synchro-Check Awareness


transmission circuit. Both Normal and Emergency Limits are set to 50 degrees for the phase angle difference as per defined operating procedures.

Figure 4: Phase Angle Difference Limits Based on Synchrocheck Relay Settings

[Source: Peak Reliability] Once the SE-calculated phase angle difference for a given NP record is approaching or exceeding its limit, the advanced application reports an NP exceedance violation of Normal Limit for base case conditions and gives the time of the occurrence of the violation. This is shown in Figure 5.

Figure 5: Alarming of Current Operating Condition Large Phase Angle Differences

In addition to steady-state operating condition awareness, Real-Time Contingency Analysis (RTCA) is also running on a timeframe of approximately 5 minutes or faster. The RTCA will further check for NP angle limit violations during post-contingency conditions compared against the Emergency Limit (Figure 6). Unlike the SE application, RTCA is designed to detect a potential angular separation exceedance ahead of the actual incidence, so that system operators can develop timely mitigation plans to prevent a large angular separation from causing an operational issue. The mitigation plan can be derived from operations engineering guidance as well as operating procedures.



Figure 6: Post-Contingency Angle Monitoring through RTCA [Source: Peak Reliability]

Sensitivity analysis tools are being explored to provide automated mitigation analysis within the EMS platform. When an angle separation exceedance is detected by the angle limits in the SE and RTCA, the software initiates network sensitivity analysis against the angle separation constraint to identify which control actions are available to remove or mitigate the angle separation violation. These actions include:

1. Generation re-dispatch to shift path or line MW flow;

2. Coordinated phase shifting transformer (PST) tap movements;

3. Curtailment of bilateral point to point (PTP) power transfer transaction;

4. Transmission switching including lines and series compensation elements3;

3 In the WECC footprint, there are over 200 series capacitors installed on transmission circuits to allow compensating long distance transmission lines by up to 80% of the line reactance.



Visualizing emergent N-1 contingencies is a tremendous help to the real-time system operators to have a physical representation of possible problem areas rather than trying to digest conventional contingency lists. Figure 7 shows the Geo-Spatial Visualization System (GVS) application displayed on the control center wall map to increase operational situational awareness4. Each operator console has high resolution monitors that allow the operator to customize their own view using laying capabilities. Violations, including phase angle differences exceeding defined limits, are shown in tabular form as well as geo-spatially on the map.

Figure 7: Visualization of RTCA Results [Source: APS]

4 GVS application is currently deployed by Arizona Public Service in real-time operations.



SE and RTCA provide a near real-time means of monitoring system angular separation. These tools use SCADA data and employ steady-state analysis tools for monitoring angle separation. PMU data can complement this by monitoring phase angle difference dynamics and with much higher resolution. In Figure 8, the trend clearly demonstrates the advantage of using synchronized PMU voltage angles to complement EMS network applications. PMU phase angles are down-sampled to 1 sample/second, as compared with the SE solved bus angle difference calculations for that same transmission circuit cycled at 1 minute intervals. As the grid dynamically changes, the SE results become stale and do not reflect the changing nature of the grid.

Figure 8: SE Solved Bus Angle vs Down-Sampled PMU Phasor Angle

[Source: Peak Reliability]



Synchrophasor-Based Tools & Real-Time Awareness Emerging synchrophasor technology enables utilities to monitor angular separation conditions with high resolution synchronized PMU phase angle measurements at rates of 30-60 samples per second. PMU measurements are being integrated with conventional EMS network applications discussed in the previous section. Synchrophasor-based applications are also developing innovative ways to monitor, visualize, and alarm on large angle exceedances to complement existing tools. Figure 9 simply shows the high resolution voltage phase angle difference calculation between two specific points on the grid over multiple minutes using PMU data. Overlaid on the plot is a hypothetical phase angle measurement from a SE solution provided event 1-minute. Note that the phase angle can fluctuate significantly during these 1-minute intervals and being able to visualize and understand this is crucial.

Figure 9: Voltage Phase Angle from PMU Data & SE Solution [Source: Peak Reliability]

Synchrophasor-based applications use the time-synchronized phase angle measurements from PMUs to calculate phase angle differences at high resolution. These applications can also issue alarms if monitored phase angle differences exceed limits immediately following a major system event. The benefit of using PMU data to supplement the EMS applications is that it gives the operators immediate alarming of angle exceedances, rather than waiting for the subsequent state estimator solution and RTCA alarms (as shown in Figure 9 above). That period of time immediately following a major grid disturbance is when system operators are trying to understand the event and discern if operating conditions are acceptable and any actions are necessary to mitigate potential problems. Angle monitoring and alarming provides the operator with advanced notice of system stress. Figure 10 shows a synchrophasor-based display of angle differences compared against their respective limits.

Figure 10: Synchrophasor-Based Angle Alarming Tool [Source: Peak Reliability]



Arizona Public Service (APS) has extended its Geo-Spatial Visualization System (GVS) to include synchrophasor applications. Figure 11 shows a wide-area visualization of PMU phase angle differences across the APS system. Each defined phase angle difference (PAD) is shown in the geo-spatial map, as well as defined limits for each phase angle. Again, a map feature allows the operator to visualize system stress and angle differences across the network. PMU application results can be transferred in real-time to the EMS system, facilitating the ability for system operators to monitor angular separation conditions with both EMS-based and PMU-based applications in a complementary way.

Figure 11: Visualization of Phase Angle Differences [Source: APS]

Benchmarking PMU and SCADA Data Ensuring the quality of synchrophasor data is a critical business need, particularly when analyzing angle differences across large interconnected systems and across different operating entities (multiple Transmission Owners). The primary purpose of benchmarking PMU (down-sampled) data, SCADA data, and solved state estimator data is to determine the relative accuracy of each data source and identify any bad data issues within any of the data sources. While SCADA and SE data are widely accepted and relatively trusted, PMU data benchmarking can be used bi-directionally. Any one data source can be an outlier when compared with the others; PMU data can be flagged as poor quality or can serve as a benchmark against the SE solution or SCADA data. One key goal of benchmarking PMU data is to ensure they are within an acceptable range for the device they are measuring and compare relatively close with other data sources, including:

• Voltage Magnitude: PMU bus voltage magnitude is compared against SCADA telemetry bus voltage.

• Voltage Phase Angle: PMU bus voltage phase angle is compared against state estimator solved bus angle. One of the PMU reference angle points needs to be enabled in the SE in order to baseline the SE solved bus angles against the PMU measurement such that relative angle differences can be compared5.

• Frequency: PMU frequency is compared against SCADA telemetry frequency.

5 Only enable validated PMU voltage angles in the state estimate solution and reasonable limits need be assigned to filter out bad PMU angle values.



• Current Magnitude (and Angle): PMU current magnitudes are compared against SCADA telemetry current signals if ICCP is available. Otherwise, PMU current and voltage phasors are used to calculate line active (P) and reactive (Q) power flow to compare with SCADA P & Q values.

Benchmarking results are meaningful and valid only if the input data is representative of the physical equipment being monitored. This assumes that the SE model is accurate, the PMU and SCADA data points are accurately mapped, naming conventions are addressed in the PMU and Phasor Data Concentrator (PDC) as well as EMS, and time alignment and down-sampling is handled appropriately. These can all be sources of error in benchmarking that may or may not represent actual bad electrical data measurements.

Time Alignment Raw synchrophasor data is captured at a rate of 30 samples per second, which is then down-sampled into EMS at a rate of 1 sample per second. SCADA data is received from entities at a rate of 1 sample per 10 seconds, and the EMS state estimator solves at 1 sample per 60 seconds. In general, there will be more PMU values than SCADA or SE. The tool must carefully filter and choose values which are most time aligned.

Synchrophasor Data Validation Options There are numerous methods being explored within the industry for determining the accuracy of synchrophasor data. These are actively being developed within NASPI community and other forums. A formal definition or process for data quality tracking is not yet well understood among industry professionals; however, industry is making strides improving data quality prior to using the data in advanced applications. Examples of industry efforts focused on data quality include:

• A Linear State Estimator (LSE) application using C37.118 streamed data to estimate the system state using direct non-iterative solutions at the PMU reporting rate. The linear state estimate can be used to identify bad data quality using PMU data only.

• PDQ Tracker6 provides real-time data validation functions by interfacing with the LSE. This real-time data validation methodology is intended to check PMU stream availability other than data quality.

Signal Mapping Guidelines PMU measurements must be accurately mapped to matching SCADA analog data points to perform benchmarking. Raw PMU measurements are integrated into EMS/SCADA with down-sampled PMU data, which supplement SCADA measurements. The down-sampled PMU angle data points in SCADA need be mapped to the EMS network model in order to be enabled for the SE solution as well as used for benchmarking. An example of this mapping is illustrated in Table 1. The measurements must have matching units or an appropriate scaling factor, and ideally should be monitoring the same network Elements as closely as possible. This may include using SCADA and PMU measurements from the same current or voltage transformer. The mapping table can be incorporated into the model update process, which includes checks for valid records, facilitating continual updating of this table. A standardized naming convention for PMUs should be established for effective mapping.

6 Grid Protection Alliance (GPA). “PDQ Tracker”. [Online]. Available: https://pdqtracker.codeplex.com

https://pdqtracker.codeplex.com/



Table 1: PMU Signal and EMS Naming Conventions [Source: Peak Reliability]

EMS Fields PDC Fields

Area Substation Device Device ID PMU Name Signal Name Signal

BPA ALLSTON BUS 500_NORTH_BUS FRQ W001ALLSTON___01 A500FREQ_____1F_ F

APS PINPKAPS BUS 900A KVA W066PINPKAPS__02 B500BUSA_____1VP A

BC-HYD NICOLA ZBR MICA_NIC__19Z2 KVM W030NICOLA____01 L500MICA_____1VP M

IPCO KINPORT LN KINP_POPU_1345 AA W034KINPORT___01 L345POPULUS__1IP A

LADWP ADLNTO LN ADLN_VICT_1500 AM W068ADLNTO____01 L500VICTVL___1IP M

PMU Data Quality Poor PMU data quality, such as loss of synchronization, can have a significant impact on phase angle difference calculations. Typically an PMU that is not time synchronized will report rapidly changing phase angle values that can cause erroneous values in difference calculations. However, if appropriate data quality flags are utilized and configured, these bad calculations can be omitted from applications using the data. Figure 12 highlights a common PMU data quality issue that could affect an angular separation monitoring application. In this example, a transmission line part of a major transmission interface is out-of-service, resulting in loss of PMU data integrity and accuracy because the PMU is installed on the line side. After line tripping, PMU voltage angle fluctuates between 180 and -180 degrees while the line is out of service (Figure 13). Network applications need data quality intelligence to quickly and accurately flag this data as bad quality such that the calculated information does not get displayed or portrayed to the system operator.

Figure 12: PMU Raw Data after Loss of Line Equipment [Source: Peak Reliability]



Figure 13: PMU Data Quality Issue after Loss of Line Equipment [Source: Peak Reliability]

EMS and PMU Data Quality Flags Benchmarking must account for data quality flags from all data sources as well as the state estimator. Quality flags should be pulled for all SCADA and PMU values prior to any comparison calculations. Only points with good quality should be included in the validation calculations. Each EMS vendor has defined their own EMS/SCADA quality flags. The State Estimator solution must also be converged network solution with reasonable mismatches. It is assumed that the SE solution is of good quality if these conditions are met. The C37.118 protocol includes data quality flags in the Status Word bits, and these can be manipulated throughout the PMU data stream if issues are detected. The percentage of time a PMU cannot be validated due to communications network quality problems should be calculated and included in a log file for awareness. Benchmarking Calculations Benchmarking results can be tracked as a percentage of time (number of points) the data is within the deviation limit compared against the total amount of time (points) archived in the historian during the validation start and end times, referred to as “Good %”. Figure 15 shows a process flowchart of PMU validation against SCADA and SE values7. For benchmarking PMU and SE solution data, the PMU measurements and SE solved quantities should be sufficiently close enough. For example, Peak RC uses a deviation within ±2 degrees over 98% of time for a given time period for validating phase angle measurements. When this is the case, both the network model and PMU measurements can be validated as “good”.

7 The method proposed here has been vetted and presented in WECC Joint Synchronized Information Subcommittee (JSIS).



Figure 14: PMU (Down-Sampled in SCADA) and SE Estimated Angle Comparison




Figure 15: EMS-PMU Validation Logic Diagram [Source: Peak Reliability]



Without this benchmarking and accuracy review, notable gaps can exist between the PMU phase angle differences and the state estimator solved bus voltage angles. This highlights an issue with, and need to review, either the PMU measurements or the network model. While many utilities rely and fall back on their network model as more accurate, there needs to be objective analysis to determine where the differences are originating. Also to consider are the potential inaccuracies in the instrument transformers that can lead to differences between these values. Peak RC currently uses a threshold of ± 2 degrees to filter out PMU voltage angles inconsistent with SE solved bus voltage angles. The 2 degree threshold is based on time lag in SE values, inconsistencies with the instrument transformers, and noise in the high-resolution PMU measurements. Figure 16 shows an example of the PMU measurements compared with the network model angles, highlighting nearly an 8 degree difference between these two. This is a prime example of a discrepancy between SE and PMU data, highlighting a need to determine the source of error.

Figure 16: Line-Based Angle Separation Curves – PMU (Down-Sampled in SCADA) vs. SE

Values [Source: Peak Reliability]



Mitigation and Operating Procedures for Line Restoration As the 2011 Southwest Outage Finding and Recommendation 27 highlighted, monitoring excessive phase angle differences between adjacent transmission circuit terminals to identify lines that cannot be restored following outage is a viable and relatively straightforward use of PMU data. The thresholds for these angle differences are based on the synchrocheck relay settings. Criteria to identify these phase angle difference exceedances should be based on the following:

• Different threshold settings and design philosophies among Transmission Owners within a Reliability Coordinator (RC) footprint;

• Mitigation strategies based on actual synchrocheck settings at the location where large phase angle difference is occurring;

• Assessments including high transfer analysis to stress the power system among key interties and transfer paths to identify likely phase angle difference exceedances; and

• Known operational restrictions or limitations; and

• Planned outage conditions or historical outage events of interest. Actions for Excessive Phase Angle Differences The utility industry has worked collectively to develop effective mitigation strategies for excessive phase angle separation. Here are practical control actions commonly employed:

• Reconfiguration of in-series capacitors/reactors for compensation of transmission circuits

• Generation redispatch

Reducing generation on the sending end of the angle difference path

Increasing generation on the receiving end of the path

• Use of phase-shifting transformers to reduce power flow (if available)

• Reconfiguration of system topology to reduce power flow (if possible)

• Curtailment of interruptible load, if necessary

• Firm load shedding, if necessary

• Point-to-point transmission service curtailment


Correlating Phase Angle with System Conditions Correlation analysis between bus angle differences and MW flows provides a deeper understanding of the dynamic properties of the interconnected bulk power system. This better understanding will allow certain thresholds for abnormal system operating regions to be determined. Identified thresholds can then be used to configure alarms and operating procedures. Correlation analysis requires access to historical data and real-time advanced network application results including:

• PMU measurements

• SCADA measurements

• EMS Network Applications – state estimator solutions, RTCA results, voltage stability and transient stability analysis results, oscillation event detection or mode meter results, etc.

Phase Angle and Real Power Correlation Correlation between phase angle and real power flow is self-evident from the DC load flow (DCLF) equation, which estimates active power line flow on AC power systems, neglecting reactive power. This non-iterative solution will converge but is less accurate than a full AC power flow solution. DCLF is used wherever repetitive and fast load flow estimations are required, as well as for sensitivity analysis around a defined operating state. In a DCLF, the nonlinear model of the AC system is simplified to a linear form through these assumptions:

• Line resistances (active power losses) are negligible (i.e. R<< X)

• Voltage angle differences are assumed to be small (i.e. sin(θ) = θ, cos(θ) = 1)

• Magnitudes of bus voltages are set to 1.0 per unit (flat voltage profile)

• Tap settings are ignored The bulk power system, at voltage generally 100kV and above, consists of transmission lines that exhibit an X/R ratio that is usually small; therefore, the assumptions of DCLF are relevant for certain types of analysis. Based on the above assumptions, voltage angles are the variables of a DCLF and active power injections are known in advance. As a result, active power flow through transmission line i with reactance XLi between buses s and r can be calculated by

𝑃𝑃𝐿𝐿𝐿𝐿 =1𝑋𝑋𝐿𝐿𝐿𝐿

(𝜃𝜃𝑠𝑠 − 𝜃𝜃𝑠𝑠)

It is clear that, as an estimate, MW flow across an in-service transmission line is proportional to the angle difference between the sending and receiving bus phase angles of the line. The phase angle difference resulting from the outage of a given transmission circuit is non-linear and based on stress pattern and the Thevenin equivalent impedance between the two terminals. Figure 17 shows a 20-day PI trend of transmission interface MW flow (e.g. Path flow) vs. phase angle difference between two buses of the substations associated with the interface. The plot shows correlation between MW flow magnitude and phase angle separation. Figure 18 shows a 2-Hour X-Y plot of transmission interface MW flow (e.g. Path flow) vs. phase angle difference between two buses of the substations associated with the interface. The plot shows correlation between MW flow magnitude and phase angle separation. The correlation coefficient between the path flow and the angle pair is 0.96336 which indicates strong correlation. The dashed straight line represents the linear correlation line.

Correlating Phase Angle with System Conditions


Figure 17: Interface MW Flow and Phase Angle Difference Correlation – 20 Days


Figure 18: Interface MW Flow and Phase Angle Difference Correlation – 2 Hours




This correlation can be observed more clearly when the plot is zoomed in on a 2-day window (Figure 19). Some interesting observations from this specific example correlation analysis include:

• Phase angle separation can increase prior to interface MW flows increase (circled in red);

• Angular separation rate-of-change can be sharper than the rate-of-change of interface MW flow; and

• Phase Angle separation changes can lag behind changes in interface MW flows (circled in blue). This indicates that system operators may gain earlier awareness of significant changes in operating conditions from appropriate angular separation monitoring, in conjunction with and complementing interface MW flow monitoring. A transmission interface consists of multiple lines and/or transformers that are connected to or located at different substations. The correlation is sensitive to selection of a specific angle pair, and the correlation analysis should consider these angle selection differences. This is particularly useful when defining backup or alternative measurements in the event of loss of primary signal. For a single transmission line, the correlation of angular separation and MW flow becomes highly observable. Figure 20 indicates that:

• MW flow pattern matches the phase angle difference pattern before line tripping; and

• Phase angle difference may indicate system stress (higher angle difference) while the MW flow has dropped to zero after line tripping.

Figure 19: Correlation between Interface MW Flow and Phase Angle Difference

[Source: Peak Reliability] Patterns of phase angle difference relative to MW flows can vary as a function of system conditions, topology changes, angle pairs chosen, and other factors. It is necessary to study a long series of phase angle and MW flow data for a variety of angle pairs and grid conditions to identify meaningful angle difference thresholds and consistent indicative angle pairs.



Figure 20: Angle Difference between Line Terminals after Line Tripping (Zero MW)

[Source: Peak Reliability] Large interconnected power system operating limits are often defined by establishing major transmission interfaces or cut planes. These are defined as System Operating Limits (SOLs) and Interconnection Reliability Operating Limits (IROLs) [4,5]. For example, the Western Interconnection has nearly one hundred established transmission paths (“WECC Paths”) to monitor system operations [6]. Operations engineers run offline studies to determine the Path operating limit (in MW) upon “assumed” system topology and forecasted area load levels. Due to the strong correlation between MW and angle, in conjunction with the proliferation of PMU coverage and redundancy, phase angles may be a suitable operating limit criteria for the future. Figure 21 shows 1) time series of a large system event in WECC measured by PMUs, and 2) the relative comparison of angle vs. active power transfer in the pre-contingency, post-contingency and during-contingency operating conditions.

Shown in Figure 22 is a hypothetical MW Limit, which could be associated with a SOL or IROL. Assuming this MW limit is determined to be 1325 MW, a corresponding phase angle limit can also be defined based on either offline or online methods. In this example, two hypothetical phase angle limits are determined:

1. Limit = 6.75 deg: This limit, based on the correlation analysis, is likely not to be hit prior to the MW limit defined. In this case, the angle can really serve as a backup limit for unexpected conditions.

2. Limit = 6.60 deg: This limit, based on the correlation analysis, is likely to be hit prior to the MW limit defined. In this case, the angle is a better indicator of system security compared to the MW flow lilmit and could be used a primary or supplemental limit.

ANGULAR SEPARATION vs LINE MW FLOW



Figure 21: Path MW and Angle Measurement for Contingency Event

[Source: Bonneville Power Administration]

Figure 22: Path MW and Angle Compared with Interface Limits

[Source: Bonneville Power Administration] The examples shown here provide some insights into how correlation analysis can be useful for phase angle monitoring and limit monitoring.


Identification of Key (Optimal) Angle Differences Major Transmission Interfaces or Transfer Paths Wide-area phase angle monitoring of major transmission interfaces is a useful application of PMU technology. These interfaces are used to operate the bulk power system within known operating limits, and having additional visibility into system conditions including angular separation provides operators with situational awareness. The WECC Paths are used as an example to illustrate angle monitoring of major transmission interfaces. WECC Intertie Paths WECC coordinates a number of high voltage transmission interfaces in the Western Interconnection that consist of one or more transmission circuits, called the WECC Paths. These interfaces include transmission lines and transformers separating the system into cohesive areas or systems between various operating entities and geographic regions. These areas can be quite distant, such as Path 65 (the Pacific DC Intertie (PDCI) between the The Dalles, Oregon and Los Angeles, California) or relatively short such as Path 5 (West of Cascades – North), primarily within the BPA footprint in Washington State. These Paths are currently numbered 1 to 83, with a few numbers intentionally omitted. Since the interties may consist of multiple circuits, the operating voltages for these circuits ranges from 55 kV to 500 kV. Let us consider Path 5, West of Cascades North, which interconnects the Seattle, WA load center to Upper Columbia generation. This Path consists of a handful of 500 kV, 230 kV, and 345 kV circuits. To illustrate the complementary benefit of phase angle monitoring, consider an operational scenario where one of the Path 5 500 kV lines is removed from service. In this scenario, the Path 5 transfer stress is analyzed using bus phase angle difference rather than MW flow and limit. Figure 23 shows pre- and post-contingency conditions visualized by (1) Path MW flow, (2) Path MW limit, (3) Line MW flow; and (4) PMU phase angle difference (PAD) across the terminals of the line removed from service [7]. The example reveals that:

1. There is a visible correlation between Path MW flow and PAD before the line is removed;

2. The Path operating limit remains constant before and after the line is lost (the Path operating limit is not dynamically calculated in real-time for a forced outage);

3. The Path MW flow is reduced by about 200 MW as a result of line outage; and

4. The PAD has a sharp increase in response to loss of the line, indicating system stress. Traditionally, system operators gain awareness of operating stress for a given Path by monitoring Path loading as a percentage of the Path Limit in MW.

𝑃𝑃𝑃𝑃𝑃𝑃ℎ 𝑆𝑆𝑃𝑃𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 =𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝑀𝑀𝑀𝑀𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑃𝑃𝑀𝑀𝑀𝑀

∗ 100

In this scenario, the Path loading % actually decreases following the outage, so system operators don’t perceive an apparent increase in system stress for the Path after the line outage. On the other hand, the angle separation experiences a step increase, illustrating the system operator should also be aware of these values as well because it reflects the topological change of impedance.

https://en.wikipedia.org/wiki/Path_65

https://en.wikipedia.org/wiki/Path_61

Identification of Key (Optimal) Angle Differences


Figure 23: WECC Path 5 Flow & Limit, Line Flow and Phasor Angle Difference

[Source: Peak Reliability] Oscillatory Stability Analysis Inter-area oscillations are a wide-area electromechanical phenomenon involving coherent generators oscillating against other generators. Local oscillations involve a single or small group of generators oscillating against the rest of the system. Both these types of oscillations can affect the stability of the entire interconnected bulk power system and impact transmission interfaces between these oscillating generators. For example, the North-South modes in the Western Interconnection involve generators in Canada and the Pacific Northwest swinging against generators in the Desert Southwest and Southern California. These oscillations are usually well damped but do manifest themselves as power swings on major transmission circuits and interfaces such as the California–Oregon Intertie, British Columbia–Northwest, and WECC Path 26. One major difference between synchrophasor data and scan-rate SCADA data is that PMU data can detect inter-area modes excited by a major grid disturbance while SCADA measurements cannot. The August 10, 1996 outage in the Western Interconnection is a prime example of a sequence of events that led to undamped inter-area oscillations and eventual system separation (Figure 24) [8]. While this plot shows voltage, wide-area phase angles were also oscillating due to large power swings from electromechanical modes on the system. PMU data captures these types of oscillations, and phase angle differences can be used to measure event oscillation ringdown for major events. Upon detection of an event, automated tools can capture the modal characteristics such as oscillation frequency and damping ratio. These tools could have detected a sustained undamped oscillation had they been in place in 1996.

LINE MW FLOW PATH RATING in MW PATH MW FLOW BUS ANGLE SEPARATION



Figure 24: August 10, 1996 Oscillation – Malin 500kV Voltage [Source: BPA]

Another useful example in the Western Interconnection is the August 4, 2000 oscillation event that caused large voltage, power, and angular swings on the bulk power system following a contingency. As Figure 25 shows, the oscillation damping ratio was very low, requiring over 60 seconds for the oscillation to settle [9]. Operator awareness that the system is marginally stable for contingency events provides awareness of the state of the bulk power system. Identifying these types of events quickly and effectively helps the system operator decide whether and how to respond. In this case, cutting scheduled transfers and generation redispatch would reduce the oscillation risk being driven by wide-area power transfers.

Figure 25: August 4, 2000 Oscillation – Malin 500kV Voltage [Source: BPA]

Various oscillation monitoring applications and software tools using PMU data have been deployed at Bonneville Power Administration, Southern California Edison, other WECC utilities, and Peak Reliability. Table 2 shows the predominant (known) oscillation modes in the Western Interconnection and the PMU phase angle pairs used as



input to calculate oscillation frequency and damping ratio in some of the tools. Currently, the North-South Mode A and Mode B are being monitored in real-time at Peak RC and BPA control centers using PMUs across the Western Interconnection.

Table 2: Oscillation Modes in the Western Interconnection

Mode Oscillation Frequency PMU Locations Monitored

North-South Mode A 0.25 Hz Custer 500 – Malin 500

Custer 500 – Captain Jack 500* (different PMU than BPA)

North-South Mode B 0.34 Hz Malin 500 – John Day 500

Malin 500 – Big Eddy 500

East-West Mode 0.45 Hz Not Monitored

Alberta Mode 0.60 Hz Not Monitored

Montana Mode 0.80 Hz Not Monitored

Maps of mode shape are used to illustrate how the inter-area electromechanical modes are manifesting on the bulk power system. Figure 26 shows N-S Mode A and N-S Mode B mode shape maps. N-S Mode A involves the Northern generation fleet oscillating against Southern California generation, while N-S Mode B has Alberta generation oscillating in phase with the Southern California units. It is clear that these modes are heavily influenced by the status of Alberta connection with the remaining Western Interconnection [10].

Figure 26: N-S Mode A (left) and N-S Mode B (right) [10] [Source: WECC] Input signals of PMU angle differences for N-S Mode A (Canada-U.S. border) and N-S Mode B (Lower Columbia Hydro-California to Oregon border) were selected for the Mode Meter engine by both BPA and Peak RC because of the correlation with North-South flows predominantly across the California-Oregon Intertie (COI) (Path 66) [12]. Monitoring a particular mode requires PMU placement at the substations where the electromechanical mode is



most observable. Observability of a mode can be determined using offline system studies and oscillation analysis. Monitoring the locations with highest observability provides a more accurate estimation of frequency and damping ratio of the mode. Monitoring the shape of the mode requires a higher level of placement with PMUs located at as many possible observability points within the interconnection. In general, is has been found that mode shapes rarely change unless major grid topology or generation shifts occur. N-S mode B results calculated from both BPA and Peak’s Mode Meter applications are shown in Figure 27. These applications using phase angle differences in their modal estimation algorithms. The software platforms and installations were benchmarked against each other to ensure consistent results. As Figure 27 shows, mode frequency and damping ratio results are very close. For the system conditions tested, the N-S Mode B damping ratio was impacted due to the events on the WECC system.

Figure 27: Brake Test – N-S Mode B Results – Peak vs. BPA [Source: Peak, BPA]



Phase Angle Visualization of Operating Boundaries Utilities are working with transient stability analysis software vendors to develop a new phase angle visualization feature. One desired feature allows users to visualize angle separation between the two regions where power is being transferred. It uses the angle separations calculated from base case powerflow and transient stability based transfer limits. This visualization feature shall contain two graphics:

• Bar charts for interface and source MW limits

• Dial gauges for angle spread limits

Figure 28: DSA Manager/TSAT Angle Separation Visualization [Source: Peak Reliability]

The user will have the ability to select different nodal pairs from the dial gauges. The software shall provide a display of angle-spreads base case and limit values in the “History” window. This new visualization feature can facilitate correlation baselining between power transfer level and bus angle separation degree for transient stability limited WECC Paths and cut planes. The identified bus angle pairs and correlation baselining results will be valuable input to angle separation exceedance alarms. Voltage Stability and Phase Angle Utilities are working with voltage stability analysis software vendors to develop a new approach to compute phase angle limit. The intended software is capable of computing the angle limits from State Estimator data only, PMU data only, or both for three types of different scenarios of stressing:

1. User-defined scenario (e.g. “sources” and “sinks”) for stressing Conventional, but sub-optimal, stressing technique. The maximum transfer capability may be reached due to exhausting resources before reaching voltage/thermal/steady state stability violation. The tool computes both the most critical (sensitive) phase angle differences and user-defined angle pair differences based on SE base case for each stressing. The function is applicable in both real time and off-line. The approach is being explored by ISO-NE in more detail.

2. Optimal scenario based stressing Planning tool to determine the maximum interface flow and phase angle limits. In principle, the software maximizes the interface/path flow by optimal source and sink grouping so that maximum phase angle limits may be calculated for the given interface/path.

3. “Natural” direction of system stressing PMU or SCADA data are used to determine the “natural” direction of stressing based on historical information. Phase angle limits are computed based on SE and PMU data. PMU data is used to determine the change in system conditions, and thus the direction in which the system is stressed, while SE data is used to determine the limit for the “natural” stressing direction. No traditional stressing needs to be performed in the approach.



Figure 29: ISO-NE Angle Separation Stressing Analysis [Source: V&R Energy]

Figure 30: Computing Interface Limits-Angle vs MW [Source: V&R Energy]


Linking Phase Angles with System Studies Definition of Safe & Alert Operating States Angle separation limits should be defined in the context of system operation reliability and security criteria, and are contextually dependent on a number of factors such as system dispatch and topology. If violated, it signals that the system is being operated under unsafe/adverse conditions such as:

• Synchrocheck relay reclose angle difference exceedance

• Excessive thermal limit violation that potentially leads to cascading outages

• SOL or IROL exceedance due to voltage stability and transient stability concerns

• Subsynchronous resonance (SSR) driven by bulk wind generation and weak system connection issues

• Triggering RAS/SPS unexpectedly

• Transaction schedules or economic transfer constraints

• Combination of multiple factors listed above The angle separation limit can be examined and confirmed by various reproducible system studies for a given base case and post-contingency states (steady state and transient state) and other assumptions. Such angle separation limit is justified with a “Safe” operating state. Otherwise, the angle separation limit simply indicates historically defined “Normal” or “Off-Normal” security region. This type of angle separation limit basically gives “Alert” or “Warning” of unusual system operating state. These two types of limits are complementary in that one identifies violations and the root cause while the other defines action plans when adverse conditions occur. Defining Inter-Area Stability Limits Based on Phase Angle Transient stability studies are performed to ensure that oscillation damping ratios do not drop to unacceptable levels after credible contingencies occur in order to maintain some margin of stability in the system. Ringdown analysis can be performed on the transient stability results to identify the oscillation frequency and damping ratio of the predominant electromechanical modes. Example analysis tools include Prony analysis and VARPRO developed for these purposes8. Pre- and post-contingency damping ratios are strongly correlated based on a number of factors. Figure 31 shows this correlation for a number of stability studies performed at various operating conditions, stress patterns, dispatches, topologies, etc. Notice the relationship between pre- and post-contingency damping ratio values for the critical contingency studied in this example. In this case, operating conditions are identified that result in < 0% damping ratio (unstable) for pre-contingency damping ratio values around 6%. This provides information to the real-time tools for setting limits on oscillation detection and mode meter applications, in coordination with SCADA or PMU data relating the damping ratio calculation to system condition information.

8 D. Trudnowski, J. Johnson, J. Hauer, “Making Prony analysis more accurate using multiple signals,” IEEE Trans. on Power Systems, vol. 14, no. 1, pgs. 226-231, 1999. A. Borden, B. Lesieutre, “Variable Projection Method for Power System Modal Identification,” IEEE Trans. on Power Systems, vol. 29, no. 6, 2014.

Linking Phase Angles with System Studies


Figure 31: Pre- and Post-Contingency Damping Ratios Based on Studies [Source: BPA]

Linking Phase Angles with System Studies


In transient stability simulations, pre-contingency operating conditions are extracted from the steady-state base case (Figure 32) along with generation dispatch and major intertie real power flows. This data can be used for comparison and correlation analysis; there is an inverse correlation between phase angle differences and oscillation damping ratios. For each angle pair identified, one can determine a pre-contingency phase angle difference threshold to ensure sufficient damping ratio is maintained post-contingency for the critical contingency studied. For example, if a 2% post-contingency damping ratio is defined as the limit, then each angle difference pair, or more likely selected critical phase angle pairs, can be assigned a phase angle limit as well.

Figure 32: Post-Contingency Damping Ratio and Phase Angle Difference [Source: BPA]


References

[1] Federal Energy Regulatory Commission, “Final Report on the August 14, 2003 Blackout in the United States and Canada: Causes and Recommendations,” U.S.-Canada Power System Outage Task Force, Apr 2004. [Online]. Available: http://energy.gov/sites/prod/files/oeprod/DocumentsandMedia/BlackoutFinal-Web.pdf.

[2] Federal Energy Regulatory Commission, “Arizona-Southern California Outages on September 8, 2011: Causes and Recommendations,” FERC and NERC Staff, Apr 2012. [Online]. Available: https://www.ferc.gov/legal/staff-reports/04-27-2012-ferc-nerc-report.pdf.

[3] M. J. Thompson, “Fundamentals and Advancements in Generator Synchronizing Systems,” Schweitzer Engineering Laboratories, Inc., 2012 Texas A&M Conference for Protective Relay Engineers, 2012.

[4] NERC Reliability Standard. FAC-010-2.1: System Operating Limits Methodology for the Planning Horizon. Atlanta, GA. [Online]. Available: http://www.nerc.com/pa/stand/Pages/ReliabilityStandardsUnitedStates.aspx?jurisdiction=United States

[5] NERC Reliability Standard. FAC-011-2: System Operating Limits Methodology for the Operations Horizon. Atlanta, GA. [Online]. Available: http://www.nerc.com/pa/stand/Pages/ReliabilityStandardsUnitedStates.aspx?jurisdiction=United States

[6] Western Electricity Coordinating Council, “Path Rating Process,” Salt Lake City, 2015. [Online]. Available: https://www.wecc.biz/PlanningServices/Pages/PathRatingProcess.aspx

[7] OSIsoft PI System. [Online]. Available: https://www.osisoft.com/Default.aspx

[8] D. Kosterev, C. Taylor, W. Mittelstadt, “Model Validation for the August 10, 1996 WSCC System Outage,” IEEE Trans. on Power Systems, vol. 14, no. 3, pgs. 967-979, 1999.

[9] D. Kosterev, “Composite Load Model Development and Implementation,” NERC-DOE FIDVR Conference, Alexandria, VA, September 2015.

[10] WECC, “Modes of Inter-Area Power Oscillations in Western Interconnection,” Salt Lake City, UT, Nov. 30 2013. [Online]. Available: https://www.wecc.biz/Reliability/WECC%20JSIS%20Modes%20of%20Inter-Area%20Oscillations-2013-12-REV1.1.pdf.

[11] D. Brancaccio, “Peak Reliability Synchrophasor Project (PRSP),” WECC JSIS Meeting, March 2015. [Online]. Available: https://www.wecc.biz/Administrative/Dan%20Brancaccio%20%20JSIS%2003-03-15%20PRSP.PDF.

[12] D. Trudnowski, J. Pierre, N. Zhou, J. Hauer, M. Parashar, “Performance of Three Mode-Meter Block-Processing Algorithms for Automated Dynamic Stability Assessment,” IEEE Trans. on Power Systems, vol. 23, no. 2, pgs. 680-690, 2008.

http://energy.gov/sites/prod/files/oeprod/DocumentsandMedia/BlackoutFinal-Web.pdf

https://www.ferc.gov/legal/staff-reports/04-27-2012-ferc-nerc-report.pdf

https://www.wecc.biz/Reliability/WECC%20JSIS%20Modes%20of%20Inter-Area%20Oscillations-2013-12-REV1.1.pdf

https://www.wecc.biz/Reliability/WECC%20JSIS%20Modes%20of%20Inter-Area%20Oscillations-2013-12-REV1.1.pdf

https://www.wecc.biz/Administrative/Dan%20Brancaccio%20%20JSIS%2003-03-15%20PRSP.PDF

https://www.wecc.biz/Administrative/Dan%20Brancaccio%20%20JSIS%2003-03-15%20PRSP.PDF


Appendix A – Utility Practices Peak Reliability Coordinator (Peak Reliability) Peak RC is planning three phases to implement the angular separation correlation study and baselining project.

1. Identify a list of sensible phase angle pairs that provide strong indication of system power transfer stress on WECC Paths, IROL-related, and other cut planes for reliable operation monitoring. By leveraging use of both PMU and EMS data in PI Historian, Peak will develop preliminary thresholds for alarming angular separation exceedance.

2. Perform in-depth offline studies using current EMS applications’ SE/RTCA and Sensitivity Calculator, plus voltage stability analysis, transient stability, small signal stability analysis and oscillation detection, etc. to gain a better understanding of angular separation limits, implications of angle exceedances, and actionable measures to mitigate the large angle separation.

3. Employ “Big Data” techniques to leverage historical PMU and EMS data, operation study results, real-time transfer analysis (e.g. ATC and different real-time system stability analysis results, etc.) to calculate and verify the angle separation limits.

California ISO (CAISO) At CAISO, system operators are currently provided the pre-contingency and post-contingency angle differences computed using State Estimator solutions and using PMU data. These are available to the operators in the EMS display and allow operators to:

1. Know what is the existing SE and PMU based phase angle difference across the terminals of a line and what is the synch-check relay setting for the corresponding line. This information is displayed to the operators on substation onelines where the synch-check relay resides allowing operators to make an informed decision on when it is safe to close in the breakers at the terminals of a line that is currently out. In addition, operators are also provided approved mitigation steps in the same oneline next to the phase angle differences allowing operators to take quick actions in the event an outaged line is to be brought back into service.

2. Know what would be the SE based phase angle difference following a credible contingency and how the post-contingency phase angle difference would compare to the synch-check relay setting for the corresponding line. This would provide operators with an indication of system stress following the loss of a line and if any pre-contingency actions are necessary to mitigate the effects of the monitored contingencies.

Arizona Public Service (APS) APS has recently upgraded the operators’ real-time tools (State Estimator and Real Time Contingency Analysis application) to incorporate phase angle differences in the contingency results. APS trains all system operators to reliably identify and address large phase angle differences, including any coordination with neighbors or the RC. APS system operators are provided the pre-contingency and post-contingency angle differences using State Estimator solutions and PMU data to improve operational situational awareness. Currently APS has 23 pairs of PMU phase angle differences within its EMS and 103 open-ended BES angles programmed in its Contingency Analysis. These angle differences are available in EMS and allow the operator to:

1. Know what is the existing SE and PMU-based phase angle difference across the terminals of a line and compare it to the synch-check relay setting for the corresponding line terminal breakers.

Appendix A – Utility Practices


2. Know what would be the SE-based phase angle difference following a credible contingency and how the post-contingency phase angle difference would compare to the synch-check relay setting for the corresponding line.

APS has implemented a Geo-spatial Visualization System (GVS) which allows for visualizing PMU data and SE/RTCA contingency results. A key attribute of this application is that it will geo-spatially display angle differences on the Control Center Wall display and/or operator console. If phase angle difference across the terminals of an open line is exceeding synch-check relay settings then a visual alarm is generated. Salt River Project (SRP) SRP has expanded their contingency set with thirty-six (36) new open-ended 500kV line contingencies. Therefore, system operators are currently provided with pre- and post-contingency angle differences computed using the State Estimator solutions. These are available to the operators in the EMS display and allow operators to:

1. Know what the existing SE phase angle difference across the terminals of a line are and what the synch-check relay setting for the corresponding line is.

2. Make informed decisions on when it is safe and allowable to close in the breakers at the terminals of a line that is currently out of service.

3. Be warned/alarmed if phase angle difference across the terminals of an open line is approaching or exceeding synchro-check relay settings.

Currently, engineers are preparing mitigation steps allowing operators to take quick actions in the event an outaged line is to be brought back into service. SRP is also working towards providing appropriate PMU data to operators, enabling monitoring of phase angle differences explained in (1) in near real-time fashion. This would also help operators to be more familiar with the fact that phase angle difference is an excellent indicator of power system stress. San Diego Gas & Electric (SDG&E) SDG&E transmission operators have situational awareness when closing transmission lines with closing angle limits. The operators know where closing angle limits are entered/displayed in the real-time tool, and are aware that these limits should always be enforced for monitoring in the base case (SE solution) and contingency case (power flow). RTCA displays the voltage angle difference when the closing angle limit is exceeded. SDG&E sends certain PMU measurements to Peak RC and CAISO. CAISO calculates the phase angle differences and sends them to SDG&E via ICCP and these angle differences are displayed on an EMS screen. For example, the voltage angle for the 500kV North Gila bus is provided by APS to the CAISO, and SDG&E provides the Imperial Valley 500kV bus voltage angle. The CAISO calculates the difference and provides this data to APS and SDG&E for monitoring and when closing the Imperial Valley – North Gila 500kV line. There is no need to wait for this line to be energized to become aware of the actual angle difference as measured on the breakers. Currently SDG&E is measuring voltage phase angles at every 230 kV and 500 kV bus and values are archived in SEL SynchroWAVe Central and OSIsoft PI database at 30 samples per second. Phase angle differences are also provided in SynchroWAVe application to operating engineers. Engineers can look at the angles using graphical displays. SDG&E also records unwrapped phase angles at specific buses along with the wrapped phase angles. Unwrapped phase angles are shown to be valuable in calculating frequency and identifying frequency events.