Академический Документы
Профессиональный Документы
Культура Документы
This document, and all information contained herein, is proprietary to SEL. Any unauthorized use, distribution, or reproduction of this document (in whole or in part)
or of any information contained herein is specifically prohibited. This legend must appear on any authorized reproduction (in whole or in part)
Table Of Revisions
Rev Datecode Notes Author
0 20150604 First issue of this document SEL Engineering Services, Inc.
1 20151023 Based on internal review SEL Engineering Services, Inc.
2 20170113 Pictures updated SEL Engineering Services, Inc.
All brand or product names appearing in this document are the trademark or registered trademark of their respective holders. No SEL, Inc. trademarks may be used
without written permission. SEL, Inc. products appearing in this document may be covered by U.S. and Foreign patents.
Contents
1 Executive Summary 4
2 System Architecture 5
11 About SEL 33
11.1 Making Electric Power Safer, More Reliable, and More Economical . . . . . . . . . . . . . . . . . . . . . 33
11.2 Innovative Research and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
11.3 Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
11.4 Customer Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
11.5 SEL Worldwide 10-Year Warranty, Over 25-Year Service Life . . . . . . . . . . . . . . . . . . . . . . . . 35
By properly collecting, manipulating, and presenting power system data as usable information, the SEL PowerMAX
system enables operators, maintenance personnel, and engineering staff to diagnose system events, predict equipment
failures, and minimize unnecessary maintenance.
SEL PowerMAX is built on a fully redundant dual Ethernet network with SEL-2730M Ethernet Switches and fiber
optic communication links. This architecture is extremely robust because the loss of a single Ethernet switch or of
a communication link will not result in any loss or data, or in degraded performances. The SEL-2730M Ethernet
Switches will be connected in a dual redundant ring configuration. Rapid Spanning Tree protocol(RSTP) implemen-
tation convergence times are quicker than the competition, ensuring that failure scenarios do not result in missing or
delayed packets.
SEL strongly suggests using network architectures that are optimized for high reliability, such as a simple “ladder”
network design that minimizes network downtime due to switch or link disruptions. The picture below describes the
proposed Ethernet network architecture and the expected recovery times in the event of a fault.
In System Engineering, Dual Modular Redundancy (DMR) indicates a type of system configuration based on two
fully redundant components. In the case one of the components experiences a hardware failure, it detects it internally
and immediately disable itself. Traditional DMR systems detected system or software problems with output compari-
son, but did not offer any resilience due to the incapacity to determine which component had the correct output.
SEL DMR technique provides a solution to this problem with a supervising circuit that compares and replaces the
outputs of one of the components using a system quality algorithm. SEL LSS utilizes two schemes which can be used
independently or in a primary and backup configuration. The two schemes are:
• Contingency based Load Shedding Scheme (CLSS) - typically used as primary scheme because of its speed
• Underfrequency based Load Shedding Scheme (UFLSS) - typically used as backup scheme for its resilience to
system faults
Both the CLSS and UFLSS schemes dynamically select from a list the number of loads that are necessary to shed. In
both schemes, the individual plant loads are automatically selected for shedding based on operator-selectable priorities
and power system conditions.
The preparation of the list of loads to shed happens continuously on the SEL LSS Controllers. When a contingency
event occurs, or when one of the frequency based thresholds is reached, the LSS algorithm has already prepared the
list of loads to shed, so that the remedial actions can take place immediately.
• Pre-armed load-shedding events. These arming signals are commonly loaded into a construct called a cross-point
switch matrix for ease of indexing and operator display.
• Operator selection of sheddable-load priorities.
• Operator selection of Incremental Reserve Margin (IRM) for each generator.
• Event reports that capture detailed analog and digital information from each event that occurs, with up to 1ms
accuracy and time durations of up to 30 seconds.
• Event logs that capture detailed information of the event in a printable format for quick analysis.
• SOE logs that capture all changes of state to digital signals with 1-millisecond accuracy.
• System topology tracking: This includes complete tracking of all breakers and disconnect statuses carrying power
between sources and loads. Load-shedding algorithms must know the routes of power flow between sheddable
loads and sources.
Option B Perform multiple, fast contingency-based load-shedding without any lockout. This method is safe only when us-
ing data-freeze type logic. This system operates without any operator intervention because there are no lockouts.
When a power system event occurs, the power system requires time to return to steady-state conditions.
During these conditions, it can be dangerous to rely on power system analog data (MW, MVAR flows) for sub-
sequent firings of contingencies. Therefore, in these systems the SEL contingency-based logic processors freeze
system conditions (MW, MVAR flows) for an adjustable period after a primary contingency trigger.
Generally, this time is set for between 2 and 10 seconds, depending on the system inertia and the primary modes
(eigenvalues) of oscillation in the Customer’s power system. During the adjustable data-freeze period, the load-
shedding system continues to track system topology and adjusts for loads that were previously triggered to shed
and for subsequent generators that go offline.
• Generator trips
• Bus couplers trips
• Utility ties trips
• Asset overloads
The calculation of the amount of power to shed is based on the power loss due to the contingency event. This
amount might be reduced by the amount of active spinning reserve or capacity available at the other power sources.
When a contingency occurs, this scheme will shed loads to make the total system load equal to the remaining available
generation capacity, thereby stabilizing system frequency.
• Df/dt thresholds
SEL frequency based scheme utilizes the above frequency thresholds to initiate the load shedding remedial action,
but then it deviates from typical protective relay based under-frequency schemes. Such schemes typically relay on the
protective devices that detected the under-frequency to shed at a feeder level. This usually results in over shedding,
since many protective relays sense the under-frequency at the same time and proceed with tripping their own feeder.
SEL frequency based load shedding scheme dynamically selects from a list of user-configurable loads that are necessary
to shed and equalize the generation to load.
The time delays associated with under-frequency load shedding are set in protective relays (usually set between 0.1
and 1.5 seconds). After this initial time delay, the round trip time calculation is basically identical to the one dis-
played for the contingency based load shedding. The truly remarkable advances over traditional under-frequency load
shedding that this system provides include the following items:
• Loads are dynamically selected (only active loads are selected to shed)
The figure below displays a typical SEL frequency based load shedding HMI screen.
The integration starts when the power produced by a generator is above a user-settable percentage of the individ-
ual generator capacity, referred to as “excess load threshold.” It will integrate proportionally to the difference between
this user-settable threshold (percent of rated capacity) through the HMI and the present output of the unit. The
integrator will use the same proportionality to integrate backwards so that a brief power output lower than the user-
defined set point will not cause the integrator to reset immediately.
For example, for a single generator rated at 36.4 MW at 27.3°C from the factory but slightly derated because of
ambient temperature, its actual capacity might be 34.609 MW after 5 percent derating. If the user enters a value of 95
percent, then the integrator will begin integrating when this single generator exceeds 33.16 MW of load. The lowest
progressive overload setting will be coordinated with the PMS upper limit setting.
The amount of load to shed will be a set point that the user will set to trip off enough loads to return the genera-
tion to well below the integration threshold and begin the integrator counting down. The figure below depicts the
logic used within the controller.
The user-settable integrator pickup “in percent of actual MW capacity” and the user-settable threshold “in Power
(pu) • Seconds” are determined from a thermal-loading model of the GTGs. These parameters must ensure that the
system sheds load before the thermal limits of the governor cause the turbine to throttle back and suddenly drop load.
The LSS will calculate the MW to shed quantity to bring loading down to less than the user-entered integrator pickup
value. If all the units on an island are evenly loaded, the amount of load required to shed to reduce the loading to
below the integrator threshold is calculated using (2):
For a system with four identical generators, if a user sets 95 percent as the threshold of a 32 MW generator, then
the worst scenario is that there is an excess of (0.05 • 32) = 1.6 MW of load on this generator. In order to reduce
this generator to less than its pickup threshold, the system will have to lose [1.6 MW • 4 generators (worst case)]
= 6.4 MW. Rounding up, in order to be conservative, yields 7 MW of load to shed. Lastly, the Maximum Reset time
should be set longer than the system response and maximum lag times of the power measurements returning to the
LSS controller. The figure below displays a typical SEL progressive overload load shedding HMI interface.
SEL GCS is a set of applications to control multiple generators across several electrical islands. It is capable of per-
forming the following functions:
• Active load sharing and frequency control with the SEL AGC module
• Reactive load sharing and voltage control with the SEL VCS module
• Start/stop commands
The GCS runs on two SEL GCS Controllers operating in hot standby. The SEL GCS Controllers send control actions to
the generator governor and exciter through I/O modules which will be installed for each generator.
The Automatic Generation Control (AGC) module dispatches turbine governor set points for equal load sharing while
simultaneously controlling either the utility tie flow or an islanded system frequency. During grid-connected condi-
tions, the AGC controls the power flow from the generators across the buses using the utility interconnected system.
During islanded conditions, the generators are controlled to keep the frequency of all islands at nominal value.
The voltage Control System (VCS) module dispatches generator exciters for equal reactive load sharing, controls
generator terminal voltages to stay within acceptable limits, manages the utility tie line power factor, and maintains
key busbar voltages at the substations. The VCS can also control the automatic/manual and tap-up/tap-down func-
tionality of the On-Load Tap Changers (OLTCs).
The Island Control System (ICS) module controls the modes of the governors and exciters during islanded and non-
islanded modes. The ICS can also control the automatic/manual status of OLTCs. The ICS also advises the governor
and exciter controllers of the Island mode conditions. All island-mode conditions and all utility ties connections are
automatically detected by the ICS.
The PMS shall contain a universal autosynchronizing system for all bus couplers and all tie lines between plants. This
autosynchronization system measures voltage and frequency on both sides of the breaker, sends raise/lower correction
pulses to adjust the governor and exciter as necessary, and automatically closes the breaker on synchronization. This
process enables safe, secure, unattended resynchronization of independent islands. The relay shall have a configurable
angle window, breaker-close mechanism for time compensation, voltage-magnitude difference, frequency-magnitude
difference, and frequency rate-of-change and voltage rate-of-change acceptance criteria to allow automatic closing.
This relay shall send its voltage magnitude difference and frequency magnitude difference to the master generation
dispatch SEL PMS Controller; this SEL PMS Controller shall send raise and lower signals automatically to the correct
generators to reduce the slip and voltage difference across the breaker being synchronized.
Disabled Mode
The Disabled mode removes the generator from all control routines within the POWERMAX and prevents noncommis-
sioned generators from affecting the portion of the system that has already been commissioned. The AGC system does
not send any up or down controls to the generator governor. The generator will be treated by the AGC as though the
it is still in Local mode. This mode will only be available at the Engineer Access level. Only users with Engineer Access
can take a unit in or out of Disabled mode.
Maintained MW Mode
Unit output is controlled to the load set point displayed on the generator dialog box only; the AGC maintains the
generator at a constant power output. This mode can be used to maintain the units at a constant power output but
with the governors in Droop mode only. This mode is stable because in the event of a sudden load increase or decrease
on the system, this generator will still throttle up or down quickly in response to the step load change, the same as all
the other generators, based on its droop characteristic. This mode will be available at the Operator Access level. Users
with Operator Access can change a unit back and forth between Maintained MW and Maintained MW With Regulation
modes only. Maintenance MW mode will not be operational if the generator is in isochronous mode.
The AGC system shall remain operational and dynamically recalculate control set points under all system bus con-
figurations and all combinations of generators in and out of service. This module runs on an SEL GCS Controllers.
SEL solution will properly handle all contingencies (such as a loss of generation) with its adaptive algorithm. For
example, load sharing will automatically re-converge a solution within a few milliseconds after a loss of generation.
If this loss of generation is followed by a frequency excursion, the SEL algorithm will deviate from its control of the
tie-line flow to help support the frequency of the combined system. This is accomplished with the non-linear control
algorithms in the SEL AGC.
Disabled Mode The VCS system does not send any up or down controls to the generator exciter. The GTG will
be treated by the VCS as though the generator is still not in Remote mode. This mode will only be available at the
Engineer Access level. Only engineers can take a unit in or out of Disabled mode.
Maintained MVAR Mode Unit output is controlled to the MVAR set point entered on the GTG dialog box menu
only. The VCS maintains the generator at a constant MVAR output. This mode should be used primarily for tuning the
VCS during commissioning. However, this mode could be used to maintain the units at a constant MVAR output. It
will be available at the Operator Access level. Users with Operator Access can change a unit back and forth between
Maintained and Maintained MVAR With Regulation modes only.
Maintained MVAR With Regulation SEL recommends this mode be the normal operating mode for all generators.
Unit output is biased towards the MVAR bias point but is not constrained to it. Rather than attempting to maintain
each generator at a specific set point, the MVAR bias point is used to weigh the relative load sharing of each generator
as a percentage of generator capacity. If all generators are set with the same percentage of rated MVAR, then all gener-
ators will be biased equally and will share load equally. This mode will be available at the Operator Access level. Users
with Operator Access can change a unit back and forth between Maintained and Maintained MVAR With Regulation
modes only.
The VCS shall remain operational and dynamically recalculate control set points under all system bus configurations
and all combinations of generators in and out of service. This module runs on an SEL GCS Controllers.
When the utility interties are open, the VCS module automatically changes its operational configuration to regu-
late the reactive power output of generators, UPSs, capacitors, and OLTCs (with regulators) so as to maintain system
voltages of the islanded system. The VCS controls the reactive power flow through the utility transformer by adjusting
the generator exciters up and down. The main bus voltage shall simultaneously be controlled by raising and lowering
tap positions on the On-Load Tap Changer (OLTC) at the utility interface. When the system is islanded, the control
shall automatically revert to bus voltage regulation while simultaneously dispatching the unit exciters to share the
MVAR load of the plant.
System voltage stability and robustness is best ensured if the exciters are run in Voltage Control mode (where the
voltage feedback signal is the generator terminals, not the main bus). SEL strongly advises against any form of trans-
former impedance (Xc) compensation; this has a poor track record in the industry because of causing voltage stability
problems between parallel running generators.
For most systems, the ICS sends one generator on every island into isochronous mode, with all other units in droop
mode. The frequency is controlled by the isochronous unit, but the AGC system will control the droop units to keep
the isochronous unit at the desired bias point.
SEL auto-synchronization module replaces manual breaker closing: it is more precise than manual systems, it in-
cludes advanced reporting and communications capabilities, and it does not require external switching of VT signals.
SEL auto-synchronization module features protection-class equipment and high-speed (sub-cycle) communications
over long distances. It can synchronize multiple machines across multiple locations.
Slip-Compensated Advance Angle Logic As the incoming and running buses slip past each other, a logic variable
asserts at an advanced angle based upon the slip rate and the breaker closing mechanism delay setting to cause clo-
sure to occur at zero degrees (12 o’clock on a synchroscope). The advance angle is calculated using the formula in
Equation below.
The picture below displays a typical HMI Auto-synchronization screen. The HMI screen duplicates the front panel
relay used for auto-synchronization. These are LEDs indications that are usually configured on the relay:
• Close Fail: Illuminates when the relay fails to perform a successful close during autosynchronization.
• Close Fail Lockout: Illuminates when the relay fails to perform a successful close during autosynchronization.
It will remain lit until reset by issuing a Target Reset.
• Comm Alarm: Illuminates when a communications alarm is detected by the relay on any upstream device.
The HMI workstation is an SEL-3355 Computers. The tough SEL-3355 Computer meets stringent IEEE C37.90 and
IEC 60255 requirements, has no moving parts or fans, operates over the range of –40° to +75°C, and has an innovative
watchdog processor to detect and alarm system problems.
The following is a summary of the SEL PMS HMI loaded on the Operator and Engineering workstations:
• Zenon HMI software and HMI applications for configuring and testing the SEL control systems. This includes
graphical user interfaces (GUIs) to administer set points and monitor the status of the AGC, system frequency
control system, automatic system islanding detection system, and the high-speed automatic load-shedding sys-
tem. This system includes a user logon system and the ability to limit rights associated with different users.
• This HMI system will visualize electrical data included in the provided I/O table. It will also process real-time
and historical alarms, real-time and historical trends, operator logs, and communications statuses.
The effectiveness of the overall scheme should be functional tested, validated in a controlled test environment us-
ing a digital transient simulator such as the RTDS®.
SEL IDSS solution can support different schemes that can be programmed in protective relays. For some of these
schemes, the relays need to be capable of exchanging peer to peer synchrophasor data for high speed decision making.
(i) Direct Transfer Trip (DTT) Scheme
Fast Rate-of-change-of-frequency (81RF) The fast rate-of-change-of-frequency protection, 81RF, provides a faster
response compared to the frequency (81) and rate-of-change-of-frequency (81R) elements. The fast operating speed
makes the 81RF element suitable for detecting islanding conditions. The element uses a characteristic (see Figure 14)
based on frequency deviation from nominal frequency (DF = FREQ – FNOM) and rate-of-change-of-frequency (DF3C)
to detect islanding conditions. The element shall use a time window of at least three cycles to calculate the value of
DF3C. Under steady-state conditions, the operating point is close to the origin. During islanding conditions, depending
on the accelerating or decelerating of the islanded system, the operating point enters Trip Region 1 or Trip Region 2 of
the characteristic.
One State Rate-of-change-of-frequency This is the traditional rate of change of frequency element used to detect
frequency changes in power systems. The rate of change of frequency is typically used to identify the rate at which the
unbalance between load and generation is happening. This rate above a certain value can be used to indicate system
disturbances and can be used to detect and initiate a decoupling scheme.
Traditional Under/Over Frequency and Under/Over Voltage Alongside of 81RF and rate-of-change-of-frequency
protection, the IDDS schemes in microprocessor relays can be programmed with the traditional over and under fre-
quency elements and voltage elements with qualifying time delays. This way, the element that has the best chance of
picking up the disturbance is allowed to initiate the decoupling process.
Angle Difference based. This scheme shall utilize synchrophasor angles between two locations (local and remote)
to calculate the standing angle difference and shall be able to trip based on predefined thresholds. The local location
should be at the plant end of a transmission network in reference to a reliable point on the utility side.
Slip and Acceleration based. The slip and acceleration method should use Out of Step (OOST) characteristic. When
compared to the 81RF element, slip is frequency difference between the two power systems and the acceleration is the
rate of change of slip.
In each of our four world class manufacturing centers, we maintain a close relationship between Manufacturing
and R&D, which enables us to rigorously adhere to high-quality standards, beginning with initial designs and extend-
ing through manufacturing and field support. We design and manufacture fault indicators and sensors in Lake Zurich,
Illinois, and panels and control houses in San Luis Potosí, Mexico. Our other product lines are manufactured at our
corporate headquarters in Pullman, Washington, and at our facility in Lewiston, Idaho.
In addition to manufacturing, SEL provides services, education, and support for a variety of industries. Our Gov-
ernment Services Division is dedicated to the security, safety, and reliability of electric power at facilities ranging from
SEL University offers training that helps our customers meet the technical challenges of integrating digitally based
technologies into the expanding power system infrastructure. We work daily to succeed at our mission—to make
electric power safer, more reliable, and more economical—by focusing on innovation, quality, and customer service.
11.3 Quality
State-of-the-art manufacturing facilities reflect SEL’s commitment to offering customers unmatched value in products
and services. We lead in price, quality, features, innovation, delivery, and service. We practice lean manufacturing
techniques to keep costs low and to provide a better value for our customers. In the spirit of continuous improvement,
SEL designs new ways to reduce waste and make our products more reliable. To ensure superior, reliable functionality,
we test our products thoroughly and verify that they will perform for decades under demanding and harsh conditions.
SEL complies with the highest quality standards—such as IPC-A-610 Class 3 and ISO 9001:2008—and constantly
exceeds those requirements, and customer expectations, in specific ways:
• Monitoring and controlling processes to exceed the ISO 9001:2008 Quality Management Systems Standard
• Developing robust, repeatable, and scalable manufacturing processes to address design and process errors
• Ensuring that our test and calibration laboratories use the latest equipment and follow National Institute of
Standards and Technology (NIST) traceable standards for accuracy and maintenance
• Partnering with our suppliers for the highest possible quality and value
SEL products are backed by our exceptional worldwide, ten-year, no-questions-asked product warranty, with excep-
tion for fault indicators, which are warrantied for five years. This warranty is the best in the industry and is proof of the
confidence we have in our product quality. We design our products for a wide operating temperature range of –40° to
+85°C (–40° to +185°F) and a service life of over 25 years. We test our products to IEEE and IEC standards with addi-
tional margins to verify that they will perform under demanding conditions for decades to come. SEL uses best-quality
components and conservative designs that apply components well within their ratings. We implement product margin
testing, exacting assembly standards, and manufacturing cleanliness. This strong commitment to quality enables us
to offer our ten-year product warranty. SEL control houses, panels, and custom enclosures are also included in this
ten-year warranty, with exception for customer-specified non-SEL products, which carry the product manufacturer’s
warranty.