6853 PhaseShifting KG 20180906 Web
6853 PhaseShifting KG 20180906 Web
6853 PhaseShifting KG 20180906 Web
Bill Cook
San Diego Gas & Electric
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained
for all other uses, in any current or future media, including reprinting/republishing this material
for advertising or promotional purposes, creating new collective works, for resale or
redistribution to servers or lists, or reuse of any copyrighted component of this work in other
works.
This paper was presented at the 71st Annual Conference for Protective Relay Engineers and can
be accessed at: https://doi.org/10.1109/CPRE.2018.8349815.
For the complete history of this paper, refer to the next page.
Presented at the
45th Annual Western Protective Relay Conference
Spokane, Washington
October 16–18, 2018
Abstract—San Diego Gas & Electric® (SDG&E®) initiated a Fig. 1 shows a simplified schematic drawing for the project,
project to add two parallel 400 MVA (+31.3° to −80.1°) phase- including the source, load, and bypass circuit breakers.
shifting transformers (PSTs) at a 230 kV interconnection
substation. California ISO (CAISO) proposed the PST project to
provide flow control between SDG&E and Comisión Federal de
Electricidad (CFE) 230 kV systems during critical N-1 or N-1-1
500 kV line contingencies. With the need to integrate renewable
generation, many utilities are using PSTs to manage the grid (e.g.,
American Electric Power [AEP] has eight PSTs in their system).
SDG&E is presently reviewing the need for additional PST
projects. The authors collaborated to address the unique
challenges of parallel PST protection and control for this
wide-ranging PST application, including implementation of
CAISO control and automatic contingency-based tap-changer
runback.
This paper discusses SDG&E’s process to execute the project,
including settings development, simulation, lab and field testing,
and in-service testing. The authors discuss oscillography analysis
used during lab testing, energization, and loading to verify the Fig. 1. Two-Core PST – Simplified One-Line Diagram
overall design and programming.
The utility formed a project team in 2014, including
I. INTRODUCTION personnel in engineering, project management, environmental
management, and supply management. Contract entities
The Imperial Valley 500 kV/230 kV substation provides the included a substation engineering design consultant who
San Diego Gas & Electric (SDG&E) utility with a 500 kV provided the substation design package and a protection and
interconnection to Arizona Public Service (APS) and 230 kV control (P&C) engineering services consultant who provided
interconnections to Comisión Federal de Electricidad (CFE) the P&C solutions. The engineering services consultant had
and Imperial Irrigation District (IID). The total interconnected provided P&C solutions for more than 20 PST projects
generation at Imperial Valley is 1,880 MW, including throughout North America with a proven design.
1,100 MW of combined cycle and 780 MW of solar generation. CAISO runs a market-based software application to
In addition, 570 MW of wind generation is connected at the determine generating-unit set points in real time. CAISO
utility’s adjacent 500 kV substations. There are two 500 kV decided in late 2015 to set the PST load tap changer (LTC) tap
lines from Imperial Valley into the San Diego-area load and an set points as generated by the market-based software, with the
underlying 230 kV east-to-west path through the CFE 230 kV possibility of adjusting set points on a 15-minute schedule. It
system. When a 500 kV line is out of service, the 230 kV path became clear that the PST controllers would need to enable
can be loaded to high levels. efficient parallel operation in an expeditious manner while
A. Analysis and Requirements interfacing with a signal from CAISO.
California ISO (CAISO) proposed a project in 2014 to install The utility and CAISO discussed in early 2016 the proposed
a flow controller on the 230 kV CFE interconnection line at operation for when a PST trips. Rather than cross-tripping the
Imperial Valley. The project goal was to provide a means of remaining unit if overloaded, the utility proposed an automatic
controlling the flow through the CFE system during N-1 and scheme to operate the remaining PST LTC in the retard
N-1-1 operating conditions, enabling the 230 kV path to remain direction, reducing the flow to a value below the 400 MVA
in service during stressed conditions. While a back-to-back dc continuous rating. This logic was implemented in the controller
flow controller was considered, two parallel 400 MVA design.
phase-shifting transformers (PSTs) were proposed for B. PST Project Features
installation at Imperial Valley. Based upon power flow studies, In summary, the unique features of this PST project were:
these PSTs were specified to have a range of +31.3° to −80.1° • Parallel high-capacity PSTs.
over 65 tap positions, with a range from +16 advance tap to −48 • A very wide phase angle regulating range.
retard tap. CAISO targeted an in-service date of May 1, 2017.
• Nonlinear impedance throughout the operating range.
2
Fig. 4. PST Impedance Variation With Tap Position Fig. 6. Main 2 Overall Differential Protection
primary windings of the excitation transformer. In addition, the If the bypass is closed, most of the current flows through the
scheme is not responsive to turn-to-turn faults on the secondary bypass breaker. Circulating current can be calculated using (2)
windings of the excitation transformer. through (4).
3) Sequence Component Differential Protection IA ADD =IA BYP + ( IA PST1 + IA PST2 ) (2)
Positive-sequence and negative-sequence differentials are
implemented in both the 87S relay and the overall differential IASUB = IA BYP − ( IA PST1 + IA PST2 ) (3)
relay. This includes the ability to compensate for a PST angle IASUB
using electrical measurements instead of mechanical indication IA CIRC = (4)
IA ADD
of tap-changer position. The sequence component differentials
require voltage transformer (VT) inputs for angle The ratios for (2), (3), and (4) are as follows:
compensation. The VTs allow directional elements to provide • For even distribution of load current between the
external fault detectors to put the sequence component parallel branches, the ratio is 0.
differentials in a high-security mode. The directional elements • For the extreme of no-load current in one of the
are then used in a permissive overreaching transfer trip (POTT) branches, the ratio is 1.
scheme. • The only way the ratio can be greater than 1 is if
4) Overall Differential Protection circulating current is present.
The 87-PSTnO relay, shown in Fig. 6, is the overall relay B. Front-Panel Indications
with protection functions similar to the 87-PSTnS relay. The
Fig. 7 shows the 87P relay front panel with all protection
only exception is that different source and load CTs are selected
elements and external trips. Pushbuttons (PBs) and PB light-
for the 87-PSTnO relay. The use of the breaker CTs instead of
emitting diodes (LEDs) are not used for this relay. Fig. 8 shows
the PST bushing CTs allows the 87-PSTnO relay to provide
the front panel for the 87S and 87O relays. There are no
complete fault coverage, because the sequence component
transformer device trips for the 87S and 87O relays.
differentials only use the two CTs at the boundary of the zone
and are therefore responsive to all faults between them. In this
way, the 87-PSTnO relay provides coverage for faults between
the source and load breakers and the PST.
5) Ground Fault Protection
The 87-PSTnS and 87-PSTnO relay systems provide
primary winding (REF) and secondary winding (64T)
equipment ground protection. Ground backup protection of the
primary winding is disabled for this application because this is
a five-legged core transformer. No ground fault contribution is
expected for the external faults.
6) Bypass Off Neutral Protection
The 87-PSTnS and 87-PSTnO relays provide circulating
current protection (32CC), also known as bypass off neutral
logic. The element detects circulating current generated in the
loop comprising the PST and the bypass path if the PST is
accidentally bypassed when the LTC is off neutral. The
differential elements generally do not respond to these
potentially damaging currents, so a dedicated protection
scheme is provided. If the PST is bypassed off the neutral, the
protection system trips the load-side circuit breaker, breaking
the circulating path while still leaving the line in service
through the PST bypass breaker.
The phase currents are measured on the load side of both
PSTs and the bypass circuit breaker. An additive and a
subtractive current are calculated from these measurements.
The additive current represents the load current down the line.
The subtractive current is a measurement of the circulating
current in the bypass loop. The ratio of subtractive current to Fig. 7. 87P Relay PST Protection Front Panel
additive current indicates whether circulating current is present.
5
follower controller. In the event the PST with the master master or switches to Independent mode, respectively. The
controller trips, the follower controller mode is changed to controllers also switch to Independent mode if the
Independent mode. The operating state in Independent mode communications channel (serial cable) fails or either PST trips.
(Off, Automatic, or Manual) will be identical to the master prior
C. Tap Position Control Logic
to the trip.
The Master/Follower or Independent modes can be chosen When Automatic mode is enabled, the control operates to
locally or via SCADA. When one of the controllers is in Master maintain the tap position as specified by CAISO. SCADA sends
mode, the other controller automatically becomes the follower. the set point via DNP3 to the controller. Fig. 10 shows the
Similarly, if the Follower or Independent command is asserted custom logic that reads and processes the CAISO tap for this
on either of the controllers, the other automatically becomes the PST. The control includes features to mitigate overloading. The
2. Close bypass breaker. types of faults and relay protection operation. Faults were
3. Close source-side PST breaker. analyzed for each location (i.e., single-phase, phase-to-phase,
4. Verify PST in neutral; no trip is asserted. and three-phase faults). The faults at fault location (FLOC) 1
5. Close PST load breaker. through FLOC 7 are internal faults. The faults at FLOC 8 and
FLOC 9 are external faults. It can be concluded from Table I
V. FACTORY ACCEPTANCE TEST AND REAL-TIME DIGITAL that multiple protection elements and relays provide coverage
SIMULATOR RESULTS for the internal faults. As discussed in Section III, the 87P relay
The factory acceptance test (FAT), including hardware-in- is based on KCL, and the 87S/87O relay is programmed for
the-loop testing (HIL) with a real-time digital simulator, was POTT, ATB, and custom sequence protection (SEQ) elements
held at the P&C engineering services consultant’s lab in July [7].
2016 [9] [10]. Utility attendees included engineers from system
protection, substation operations, and grid operations. The
purpose of the tests was to observe PST P&C system
performance and determine whether any relay setting or control
logic changes were required.
A. Lab Test Setup and Results
PST and tap-changer motors were modeled in the real-time
digital simulator. Protection and control relays were connected
in a closed loop with the real-time digital simulator model to
verify the PST operation. A detailed FAT plan was developed
to test and verify all protection elements, the front-panel Fig. 14. System One-Line Diagram and Tap Position Control Logic
display, and other features for internal and external faults. The
detailed test plan also included various operation modes of the Fig. 15 and Fig. 16 show the three-phase and single-phase
PST and controller. For each location and PST tap position, faults at FLOC 1 for the neutral tap position. These results
multiple scenarios using scripts were run to verify the operation match the results indicated in Table I. Fig. 17 shows a
for various faults and incident angles. Fig. 14 shows the three-phase fault at FLOC 2 with tap −10 retard. POTT and
simplified one-line diagram and various fault locations where ATB schemes on the 87S and 87O relays will clear this fault.
analysis was performed. Table I summarizes results for various
controller. This testing was performed in both Master/Follower 2. Advance tap on PST 1 to +1, and retard PST 2 one tap
and Independent modes. position to −1. Complete in-service testing, including
a circulating apparent power measurement. Check the
A. Field Results – Tap-Change Process
circulating current against the calculated value.
Data from both PST controllers were recorded for analysis 3. Open PST 2, placing all load on PST 1. Complete in-
and future reference. For each tap-position operation, relay data service testing on PST 1.
analysis indicated that BCD inputs used for tap-position 4. Move PST 1 to neutral tap, close bypass, open PST 1,
calculations disappeared 0.4 seconds after the motor started and move PST 2 to neutral tap, and open bypass. Then,
reappeared 2.9 seconds after the motor started. The motor move PST 2 tap to match PST 1 tap from Step 3.
completed the run in 3.5 seconds. It was observed during testing Complete in-service testing on PST 2.
that the motor ran for ~3.5 seconds for each tap change. See 5. Move PST 2 to neutral tap, close bypass, close PST 1,
Fig. 18 for the details of this operation. When passing though and open bypass. Advance tap on PST 1 to +1, and
the neutral, the motor runs three times with 0.9 seconds
retard PST 2 one tap position to −1. Complete in-
between each start and stop. The motor runs ~14−15 seconds. service testing, including a circulating current
In addition, Fig. 19 shows that the PST is moving from measurement. As done in Step 2, check the circulating
−1 retard to the neutral to +1 advance. For simplicity, only the current against the calculated value.
main Sequential Events Recorder (SER) variables are shown. 6. Verify the compensation using inrush current. During
the field testing, validate compensation and
magnitudes during first energization and loading
conditions.
The team developed calculations for commissioning to
determine the 87P and 87S/87O relay currents for operation at
+6 advance tap with 160 MW flow. These calculations were
based on the loading allowed by CAISO when the PST was
loaded for the first time. The ISOLO primary current was
calculated for the 87P relay at 84 A; the excitation winding
primary current was 896 A. Fig. 20 shows the details and
Fig. 18. PST Typical Tap-Change Operation Timing Diagram calculations for 160 MW loading and flows in the 87P, 87S, and
87O relays.
Fig. 19. SER Data for PST Typical Tap-Change Operation From R1 to A1
Matrix 3 (i.e., CCW 30 • 3 = 90 degrees for ABC phase rotation) The met dif command is shown in Fig. 23. Hence, there is
[7]. no operate current and high-restraint current. This verification
Fig. 21 shows the current magnitudes and angle before was done during commissioning for the 87P, 87S, and 87O
compensation for the 87S and 87O relays. Once the relays during no-load and load conditions.
compensation matrix is applied for both source and load, the
results indicate that AT balance is maintained. In addition, the
current magnitudes for 160 MW match the calculations
performed as shown in Fig. 20. Fig. 22 shows the current after
compensation for source and load currents (exciter transformer
primary) and excitation transformer secondary current on the
U-winding. The operate current is very low with high-restraint Fig. 23. met dif Command for PST
current, hence the compensation is correct. It is clear from Fig. 24 that the Phase A and Phase B CT
currents are rolled (not wired correctly). The transformer
differential operated and declared this to be an internal fault.
Once the CT connections are corrected for Phase A and
Phase B, the in-service test determined that the operate current
is low with high-restraint current. Because all the logic was
already tested and carefully validated during the FAT, the team
was confident in the settings and logic. Hence, the
commissioning team was able to troubleshoot and determine
the root cause of this misoperation very quickly by using the
oscillography and event records from the relays.
Fig. 21. 160 MW, 87S and 87O Relays Phasors Uncompensated
Fig. 22. PST2 87S and 87O Relays Phasors and Compensation
Fig. 24. PST1 87P Relay Incorrect CT Wiring
14
D. PST Operation and Lessons Learned 12. Develop a startup plan to incorporate in-service testing
Several months after placing the PSTs in service, a PST LTC needs during energizing and loading of the PSTs.
tap change failed due to a tripped LTC motor breaker on one of 13. Document and save controller and protective relay
the PSTs. The PST controllers operated properly to block SER files for future reference.
additional tap changes and initiated a circulating current alarm
because the PST tap positions were one step apart. The SER IX. ACKNOWLEDGMENTS
data were valuable in allowing analysis of the failed tap change The authors gratefully acknowledge the contributions of
compared to normal tap changes from the original data. The Karl Iliev and Ahsan Mirza of SDG&E. The authors are also
team was able to see the motor breaker trip indication, and they thankful for the support of Jordan Bell and Chris Knox of
knew that the motor breaker did not trip due to controller action. Schweitzer Engineering Laboratories.
This same event occurred several more times over three The work described in this paper was funded by California
months. utility customers under the auspices of the California Public
Ultimately, the LTC manufacturer made a site visit and Utilities Commission.
provided control circuit additions to provide additional
monitoring of the LTC control circuit. X. REFERENCES
[1] D. A. Tziouvaras and D. Hou, “Out-of-Step Protection Fundamentals
VIII. CONCLUSION and Advancements,” proceedings of the 57th Annual Conference for
Protective Relay Engineers, College Station, TX, March 2004.
PST control and protection schemes require a thorough
[2] “Protection of Phase Angle Regulating Transformers: A Report to the
understanding of PST design for successful implementation. Substation Subcommittee of the IEEE Power System Relaying
Correct design practices and coordination among multiple Committee Prepared by Working Group K1,” IEEE Special Publication,
teams over the design period were critical. During the course of October 1999.
this project, design changes were made to the utility operating [3] D. A. Tziouvaras and R. Jimenez, “Protecting a 138 kV Phase Shifting
Transformer: EMTP Modeling and Model Power System Testing,”
standards and requirements of the controller design. Real-time
proceedings of the 29th Annual Western Protective Relay Conference,
digital simulator testing and a FAT procedure helped utility Spokane, WA, October 2002.
engineers witness and verify the design and provide feedback. [4] M. J. Thompson, H. Miller, and J. Burger, “AEP Experience With
Hence, the design was verified and finalized in a lab Protection of Three Delta/Hex Phase Angle Regulating Transformers,”
environment, which helped in the commissioning. The proceedings of the 60th Conference for Protective Relay Engineers,
College Station, TX, March 2007.
commissioning was short and limited to only critical tests as
[5] M. J. Thompson, “Percentage Restrained Differential, Percentage of
needed for the field verification. Any misoperation was What?” proceedings of the 64th Annual Conference for Protective Relay
troubleshot very quickly, as the team had 100 percent Engineers, College Station, TX, April 2011.
confidence in logic and settings after the detailed FAT. [6] M. J. Thompson, “Apparatus and Method for Providing Differential
In summary, the team feels the following actions are key to Protection for a Phase Angle Regulating Transformer in a Power
the verification of PST control and protection systems: System,” U.S. Patent 7,319,576 B2, January 2008.
1. Identify control and protection solutions early in [7] M. Thompson, “Protection System for Phase-Shifting Transformers
Improves Simplicity, Dependability, and Security,” proceedings of the
project implementation. 39th Annual Western Protective Relay Conference, Spokane, WA,
2. Assemble a design and verification team to vet and October 2012.
review project design requirements. [8] B. Cook, M. Thompson, and K. Garg, “New Advancements in Power
3. Review the design from the early stages to ensure the Flow Regulating Tap-Changer Control Systems for Phase-Shifting
electrical design flows properly. Transformers,” proceedings of the 85th International Conference of
Doble Clients, Boston, MA, April 2018.
4. Identify special project design needs to customize the
[9] K. Zimmerman, “SEL Recommendations on Periodic Maintenance
standard design based upon utility operating Testing of Protective Relays,” December 2010. Available:
requirements. https://selinc.com.
5. Develop detailed control and protection logic. [10] IEEE C57.135-2011, IEEE Guide for the Application, Specification, and
6. Model PSTs and the transmission system for simulator Testing of Phase-Shifting Transformers, 2011.
model development.
7. Incorporate PST factory test data in the simulator XI. BIOGRAPHIES
model. Bill Cook is the Project Manager – Grid Operations at San Diego Gas &
Electric (SDG&E). Bill started his career at SDG&E in 1976 as an engineer in
8. Conduct a FAT to prove control and protection logic,
the SDG&E Control Center. He moved to the field in 1982, working in the
sharing results with utility engineers. Substation and System Protection groups. He moved to the manager position in
9. Conduct training for technicians, electricians, System Protection and Control Engineering in 1997. In 2014, he moved to his
engineers, and trainers using a simulator lab setup. present position in Grid Operations. Bill earned his B.S.E.E. from California
Polytechnic State University in San Luis Obispo. He is a registered professional
10. Fine-tune the logic and relay settings based on engineer in California and a member of IEEE. He has been a member of the
simulator results. WECC Remedial Action Scheme Reliability Subcommittee (RASRS) since
11. Develop detailed functional test plans to verify logic, 1999.
settings, and field wiring.
15
Michael J. Thompson received his B.S., magna cum laude, from Bradley
University in 1981 and an M.B.A. from Eastern Illinois University in 1991.
Upon graduating, he served nearly 15 years at Central Illinois Public Service
(now AMEREN). Prior to joining Schweitzer Engineering Laboratories, Inc.
(SEL) in 2001, he was involved in the development of several numerical
protective relays while working at Basler Electric. He is presently a Fellow
Engineer with SEL Engineering Services, Inc. He is a senior member of the
IEEE, member of the IEEE PES Power System Relaying and Control
Committee, past chairman of the Substation Protection Subcommittee of the
PSRCC, and received the Standards Medallion from the IEEE Standards
Association in 2016. Michael is a registered professional engineer in six
jurisdictions, was a contributor to the reference book, Modern Solutions for the
Protection Control and Monitoring of Electric Power Systems, has published
numerous technical papers and magazine articles, and holds three patents
associated with power system protection and control.
Kamal Garg received his M.S.E.E. from Florida International University and
India Institute of Technology, Roorkee, India, and his B.S.E.E. from Kamla
Nehru Institute of Technology, Avadh University, India. Kamal worked for
POWERGRID India and Black & Veatch for several years at various positions
before joining Schweitzer Engineering Laboratories, Inc. (SEL) in 2006.
Presently, he is a senior protection engineer at SEL Engineering Services, Inc.
Kamal has experience in protection system design, system planning, substation
design, operation, remedial action schemes, synchrophasors, testing, and
maintenance. Kamal is a licensed professional engineer in the U.S. and Canada
and a member of IEEE.