U.S. patent application number 13/282032 was filed with the patent office on 2012-02-16 for emergency frequency load shedding scheme.
This patent application is currently assigned to ELECTRIC POWER RESEARCH INSTITUTE, INC.. Invention is credited to Nicholas C. Abi-Samra.
Application Number | 20120041612 13/282032 |
Document ID | / |
Family ID | 41380775 |
Filed Date | 2012-02-16 |
United States Patent
Application |
20120041612 |
Kind Code |
A1 |
Abi-Samra; Nicholas C. |
February 16, 2012 |
EMERGENCY FREQUENCY LOAD SHEDDING SCHEME
Abstract
A load shedding system and method for controlling imbalances in
a power system includes a signal generator adapted to generate a
signal representative of a local system, and a modifying circuit
adapted to adjust set frequencies. The system also includes a
processing circuit adapted to process data received from the signal
generator and modifying circuit and set load shedding priorities
for the local system based on the processed data.
Inventors: |
Abi-Samra; Nicholas C.;
(Carlsbad, CA) |
Assignee: |
ELECTRIC POWER RESEARCH INSTITUTE,
INC.
Charlotte
NC
|
Family ID: |
41380775 |
Appl. No.: |
13/282032 |
Filed: |
October 26, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12131986 |
Jun 3, 2008 |
|
|
|
13282032 |
|
|
|
|
Current U.S.
Class: |
700/292 |
Current CPC
Class: |
Y04S 20/222 20130101;
H02J 3/14 20130101; Y02B 70/3225 20130101 |
Class at
Publication: |
700/292 |
International
Class: |
G06F 1/28 20060101
G06F001/28 |
Claims
1. A method of pre-empting a frequency decline, comprising the
steps of: (a) continuously monitoring voltage of a local system
over a set time interval; (b) estimating a rate of change of
frequency with time from voltage measurements taken during the time
interval; and (c) using the rate of change of frequency with time
to determine a frequency load shedding point.
2. The method according to claim 1, further including the step of
deriving a frequency signal indicative of the local system.
3. The method according to claim 1, further including the step of
determining a magnitude of power carried by the local system.
4. The method according to claim 1, further including the step of
determining a change of power with frequency.
5. The method according to claim 1, further including the step of
setting load shedding priorities for the local system.
6. A method of pre-empting a frequency decline, comprising the
steps of: (a) continuously monitoring voltage and current of a
feeder; (b) estimating a rate of change of frequency with time from
voltage and current measurements taken during a set time interval;
(c) estimating a rate of change of power with frequency; (d) using
the rate of change of frequency to determine a frequency load
shedding point; and (e) using the rate of change of power with
frequency to set load shedding priorities.
7. The method according to claim 6, further including the step of
deriving a frequency signal indicative of the local system.
8. The method according to claim 6, further including the step of
determining a magnitude of power carried by the local system.
9. The method according to claim 6, further including the step of
comparing the rate of change of frequency with time to a load
shedding constant.
10. The method according to claim 9, wherein if the rate of change
of frequency with time is more negative than the load shedding
constant, load shedding occurs at higher frequencies.
11. The method according to claim 9, wherein if the rate of change
of frequency with time is less negative than the load shedding
constant, load shedding occurs at predetermined frequencies.
12. The method according to claim 6, further including the step of
communicating load shedding priorities to relays and breakers.
13. The method according to claim 6, wherein the more negative the
rate of change of power with frequency is the higher the load
shedding priority is for a feeder.
Description
[0001] This application is a divisional application of U.S. patent
application Ser. No. 12/131,986, filed Jun. 3, 2008, which is
currently pending.
BACKGROUND OF THE INVENTION
[0002] The present invention relates generally to a system and
method for controlling imbalances between load and supply in power
systems, and more particularly to a load shedding system and method
for controlling imbalances in a power system.
[0003] In power systems of today, it is essential that a balance
between the amount of power being used ("load") and the amount of
power being supplied ("supply") be maintained to provide a secure
and reliable power system, thereby preventing blackouts and power
interruptions. Every imbalance between the supply and load causes a
deviation of the frequency. More importantly, if the load exceeds
the supply and the deficit is not counteracted, the deficit in
supply can lead to system interruptions and even blackouts.
[0004] One way to control these imbalances is to use load shedding.
Load shedding (i.e., opening a circuit breaker to reduce the load)
has been used to mitigate imbalances between supply and load since
the infancy of power systems, and has basically remained unchanged
over the last 100 years. During the same period, the logic used to
shed the load has also only changed slightly. Unfortunately, the
power systems, expectations of the users, and their loads have
changed drastically, thereby necessitating better ways to handle
these imbalances.
[0005] Currently, there are basically two kinds of under frequency
load shedding schemes: the first is based on pre-determined
frequencies encountered during a frequency decline, and the second
is based on the rate of decline of the frequency (i.e., first
derivative of frequency with respect to time). Both schemes use
different time delays (from zero delay to several seconds, in some
cases).
[0006] The first scheme drops the load based on predefined
frequencies, f1, f2, f3, . . . fn in the hope that the dropped load
would be enough to reverse the frequency decline by narrowing the
gap between the supply and the connected load. The predefined set
points are based on previous studies, and they do not consider the
actual state of the system when the disturbance occurs or any
dynamic system conditions ensuing from the disturbance. Though this
scheme is relatively good when there is a smaller disturbance
coupled with a long time to respond, it is unable to differentiate
between some "normal" system oscillations and actual system
disturbances. Hence, it may not shed enough load to arrest a
frequency decline associated with real disturbances. This could
cause the disturbance to cascade and result in the system actually
collapsing because the appropriate action (shed enough load) was
not taken at the right time.
[0007] The second scheme for shedding load is based on the rate of
change (first derivative) of the frequency decline (combined often
with predefined frequencies). This scheme solves some of the
concerns with the first scheme above by providing a means for a
faster response to disturbances and a better method for
differentiating disturbances from "normal" system oscillations. The
rate of change of the frequency (df/dt) is used at some or all of
the predefined frequencies.
[0008] Unfortunately, the last blackouts, including the 2003 North
East Blackout, have shown that the above existing load shedding
schemes are not sufficient to arrest frequency declines and can not
save power systems from large disturbances. Accordingly, there is a
need for a load shedding scheme that can sufficiently arrest
frequency declines that result from large disturbances.
BRIEF SUMMARY OF THE INVENTION
[0009] These and other shortcomings of the prior art are addressed
by the present invention, which provides an emergency frequency
load shedding sheme that preempts a frequency decline by shedding
load based on the severity of a disturbance.
[0010] According to one aspect of the present invention, a load
shedding system includes a signal generator adapted to generate a
signal representative of a local system; a modifying circuit
adapted to adjust set frequencies; and a processing circuit adapted
to process data received from the signal generator and modifying
circuit and set load shedding priorities for the local system based
on the processed data.
[0011] According to one aspect of the present invention, a method
of pre-empting a frequency decline includes the steps of
continuously monitoring voltage or current of a local system over a
set time interval; estimating a rate of change of frequency with
time from voltage and current measurements taken during the time
interval; and using the rate of change of frequency with time to
determine a frequency load shedding point.
[0012] According to one aspect of the present invention, a method
of pre-empting a frequency decline includes the steps of
continuously monitoring voltage and current of a feeder; estimating
a rate of change of frequency with time from voltage and current
measurements taken during a set time interval; estimating a rate of
change of power with frequency; using the rate of change of
frequency to determine a frequency load shedding point; and using
the rate of change of power with frequency to set load shedding
priorities.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The subject matter that is regarded as the invention may be
best understood by reference to the following description taken in
conjunction with the accompanying drawing figures in which:
[0014] FIG. 1 shows a load shedding system according to an
embodiment of the invention;
[0015] FIG. 2 shows input circuitry of the system of FIG. 1;
[0016] FIG. 3 shows a modifying circuit of the system of FIG.
1;
[0017] FIG. 4 shows a logic circuit of the system of FIG. 1;
[0018] FIG. 5 shows an individual relay of the system of FIG.
1;
[0019] FIG. 6 shows how a sample of load frequency dependency for a
feeder would react using the system of FIG. 1; and
[0020] FIG. 7 shows a top level process for preempting a frequency
decline using the system of FIG. 1.
DETAILED DESCRIPTION OF THE INVENTION
[0021] Referring to the drawings, an exemplary emergency frequency
load shedding scheme or system according to the present invention
is illustrated in FIG. 1 and shown generally at reference numeral
10. The system 10 is an adaptive emergency load shedding system
which can differentiate between disturbances and normal frequency
oscillations. Unlike the prior art, the system may preempt a
frequency decline by shedding load based on the severity of a
disturbance. This allows the system to shed a load earlier than
what the predetermined frequencies dictate.
[0022] Referring to FIG. 2, the system 10 uses a signal generator
11 to provide a localized reading of a local power system frequency
and power (P) carried by different feeders. Further, the signal
generator 11 generates a signal equal to the frequency of the local
system frequency and the power carried by the different feeders to
which the system 10 acts on.
[0023] As shown in FIG. 3, a modifying circuit 12 is used to adjust
predefined set frequencies (f1, f2, . . . fn) based on the severity
of a disturbance. As illustrated, the circuit 12 uses the
instantaneous rate of frequency declines ("df/dt") to change the
next trip frequency. For example, if df/dt is negative and the
magnitude is larger than a user definable constant ("k"), which can
be either determined or applied on a local basis or system wide,
then the next pre-defined tripping frequency will be increased by a
load shedding constant ("K.sub.LS"), i.e., the load shedding will
take place at a slightly higher frequency than the original
corresponding set point. K.sub.LS may be based on engineering
judgment or on the results of studies or simulations of the system
to which the system 10 is applied, but may be in the range of 0.1
to 0.3 Hz. It is expected that different power systems will have
slightly different k's. For example, for large systems which are
spread out over large geographic areas or where generating plants
are concentrated in a certain area, applying slightly different K's
at different locations could further enhance the system due to
further differentiation of needed responses, and the ability to
tailor the response of the grid more closely to what is needed to
resolve the mismatch between generation and load.
[0024] As shown, if the rate of frequency, df/dt is greater than or
equal to -k, then K.sub.LS is set to zero and, hence, the load
shedding will take place at the predetermined scheduled frequencies
without any correction applied to them. However, if df/dt is
smaller than -k (i.e., more negative), then K.sub.LS, will be non
zero and the load shedding will take place at an earlier value of
frequency than the predetermined scheduled frequency.
[0025] Referring to FIGS. 4 and 5, a processing circuit 13
processes data received from the signal generator 11 and the
modifying circuit 12 and compares the data to predetermined
frequencies (f.sub.1, t.sub.1 . . . f.sub.n, t.sub.n). It also
calculates the rate of change of power with frequency
("dP.sub.i/d.sub.f") generated by the signal generator 11. Once the
data is analyzed by a load shedding processor 14, priorities are
set for feeder tripping based on their frequency characteristics.
Thus, processor 14 prioritizes the order of shedding between
feeders in real time based on the type of load they carry. Dropping
loads with less frequency dependency is more beneficial than
dropping loads with more frequency dependence.
[0026] As shown in FIG. 6, feeders having different load
characteristics are tripped in different orders. As shown, the
priority of tripping will be from the less frequency dependent
(flatter characteristics like feeder 1) to the most frequency
dependent (as feeder 4). Tripping the less frequency dependent
first is more advantageous than tripping the more frequency
dependent load because the power carried is independent of the
frequency drop and it establishes a balance between load and supply
quicker.
[0027] In use, FIG. 7, voltage ("V") and current ("I")
measurements, and time, ("t") are continuously monitored by a
control system, computer, processor, or other suitable system or
device. From the raw values of V and I, a frequency ("f") signal is
derived. Time is established with a GPS device, while f can be
derived based on zero crossing of the voltage signal, or other
suitable method. From the measured V and I, magnitudes of the real
power (P) carried by the different feeders are calculated.
[0028] Monitoring is done on a continuous basis; hence, the change
in frequencies forms two measurements separated by a sampling
interval; thus, yielding an estimate of the rate of change of
frequency with time. This is also known as the first derivative of
frequency with time (df/dt). Similarly, the rate of change of P
with respect to frequency, or the first derivative of power with
frequency (dP/df) may be derived from consecutive P and f
measurements.
[0029] The value of df/dt provides an indication of the severity of
the disturbance. The more negative in magnitude df/dt is, the more
severe the disturbances. A positive df/dt indicates that the
frequency decay is recovering. In the case of dP/df, a zero (0)
value implies that the load is independent of frequency. A less
negative value implies less dependence of frequency. A positive
dP/df would indicate an increase of the load with the frequency;
however, it is unlikely that this would occur.
[0030] From the df/dt values, the system 10 determines a new
frequency load shedding point. This is based on the next
predetermined (set point) load shedding value. The current df/dt
alters the value at which the next load shedding event will take
place, with more severe disturbances represented when df/dt is more
negative than K.sub.LS and resulting in shedding load at higher
frequencies. When df/dt is less negative than K.sub.LS, then load
shedding will take place at the predetermined load shedding
frequency.
[0031] Similarly, dP/df is used to set the priority of the shedding
of the different feeders based on the frequency dependency of the
load they carry. This is determined by how negative the rate of
change of power with frequency (dP/df) is for each feeder, with
smaller values of dP/df having higher priority. This priority is
then communicated as trip signals to different actual relays, and
then to associated breakers.
[0032] The foregoing has described an emergency frequency load
shedding scheme. While specific embodiments of the present
invention have been described, it will be apparent to those skilled
in the art that various modifications thereto can be made without
departing from the spirit and scope of the invention. Accordingly,
the foregoing description of the preferred embodiment of the
invention and the best mode for practicing the invention are
provided for the purpose of illustration only and not for the
purpose of limitation.
* * * * *