CN103457268A - Optimal load curtailment control method based on parallel mode searching - Google Patents

Optimal load curtailment control method based on parallel mode searching Download PDF

Info

Publication number
CN103457268A
CN103457268A CN2013104234649A CN201310423464A CN103457268A CN 103457268 A CN103457268 A CN 103457268A CN 2013104234649 A CN2013104234649 A CN 2013104234649A CN 201310423464 A CN201310423464 A CN 201310423464A CN 103457268 A CN103457268 A CN 103457268A
Authority
CN
China
Prior art keywords
search
load
target function
cutting load
searching
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN2013104234649A
Other languages
Chinese (zh)
Inventor
张文磊
宋军英
陈跃辉
江全元
姚国强
李志浩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
State Grid Corp of China SGCC
Original Assignee
Zhejiang University ZJU
State Grid Corp of China SGCC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU, State Grid Corp of China SGCC filed Critical Zhejiang University ZJU
Priority to CN2013104234649A priority Critical patent/CN103457268A/en
Publication of CN103457268A publication Critical patent/CN103457268A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B70/00—Technologies for an efficient end-user side electric power management and consumption
    • Y02B70/30—Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
    • Y02B70/3225—Demand response systems, e.g. load shedding, peak shaving
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
    • Y04S20/20—End-user application control systems
    • Y04S20/222—Demand response systems, e.g. load shedding, peak shaving

Landscapes

  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses an optimal load curtailment control method based on parallel mode searching. The method includes the steps of firstly, reading in data of an electricity system through a computer; secondly, conducting initial simulation on the electricity system to obtain the power angle of each electric generator at each moment, calculating an objective function value, taking the objective function value as a primary iteration point; thirdly, selecting Nlc loads which can be curtailed from all loads; fourthly, initializing parameters of parallel mode searching, and determining the set of searching directions and the like according to the Nlc loads which can be curtailed and are determined in the third step, wherein the data of the electricity system include line parameters, generator parameters and load parameters, and the parameters of the parallel mode searching include the number of iterations, the searching precision and the initial search range. The method is a non-gradient optimization method, the calculation workloads are greatly reduced, the optimal control strategy can be reliably obtained, and the control strategy can be adjusted at any time according to the actual operation condition. Meanwhile, the reasonable distribution of transient stability calculation tasks is achieved through the frame of parallel searching, the searching efficiency is improved, and a solid foundation is laid when the method is applied to the actual electricity system.

Description

Optimum load shedding control method based on the parallel schema search
Technical field
The invention belongs to analysis and the control field of electric power system, be specifically related to a kind of optimum load shedding control method based on the parallel schema search.
Background technology
Along with the fast development of China's economic construction, required power consumption is growing per capita, and electric power has become the direct driving force of social production.In this process, the electric power system scale constantly increases, and electric network composition is day by day complicated, and single-machine capacity further improves, and interregional interconnection and long-distance and large-capacity power transmission system constantly occur simultaneously, and the contradiction between its fail safe and economy is aggravated increasingly.
When the measure of needs employing safety and stability emergency control is broken down in electric power system, load shedding control method is a kind of measure of commonly using.In electrical network generation catastrophe failure or after losing a large amount of power supplys, excise a part of load and can reduce the unbalanced power in network, maintain the stable operation of system.
At present, the actual method of formulating emergency control policy is mainly trial and error procedure.Trial and error procedure is calculated constantly examination by numerical integration and is gathered out control strategy, until find, a kind ofly can make system stability and controls the control strategy that cost is less.Therefore this algorithm needs very large amount of calculation, and is difficult to obtain best control strategy.And the network configuration of real system, operational mode and forecast accident are all in constantly variation, so current method can only, to certain limit mode of system operation, be difficult to adjust at any time emergency control policy for practical operation situation for reducing workload.Therefore, be badly in need of a kind of optimum load shedding control method that can consider actual electric network ruuning situation, obtain fast control strategy.
Summary of the invention
The purpose of this invention is to provide a kind of optimum load shedding control method based on the parallel schema search, to consider improving computational efficiency and result reliability by the electric power system model of actual complex in optimum load shedding control method.
Optimum load shedding control method based on the parallel schema search provided by the invention, comprise the steps:
Step 1, utilize computer to read in electric power system data, comprise line parameter circuit value, generator parameter, load parameter;
Step 2, electric power system is carried out to initial emulation, obtain each merit angle of every generator constantly, the calculating target function value ,
Figure 943148DEST_PATH_IMAGE002
for the primary iteration point;
Step 3, in all loads, choose
Figure 2013104234649100002DEST_PATH_IMAGE003
but individual cutting load;
The parameter of step 4, the search of initialization parallel schema, comprise iterations , search precision
Figure 2013104234649100002DEST_PATH_IMAGE005
with the initial ranging step-length
Figure 307581DEST_PATH_IMAGE006
, iteration point
Figure 591932DEST_PATH_IMAGE002
but in comprise each cutting load (
Figure 2013104234649100002DEST_PATH_IMAGE007
) cutting load controlled quentity controlled variable information, but according to the cutting load that step 3 is determined, determine the set of the direction of search
Figure 597934DEST_PATH_IMAGE008
:
Figure 2013104234649100002DEST_PATH_IMAGE009
In formula:
but be cutting load
Figure 2013104234649100002DEST_PATH_IMAGE011
corresponding unit vector, dimension with
Figure 352056DEST_PATH_IMAGE002
identical;
If step 5 current search step-length
Figure 276150DEST_PATH_IMAGE012
be less than search precision
Figure 175973DEST_PATH_IMAGE005
, finish the parallel schema search, obtain optimum cutting load controlled quentity controlled variable, forward step 9 to; Otherwise forward step 6 to;
Step 6, according to current iteration point
Figure 2013104234649100002DEST_PATH_IMAGE013
and step-size in search
Figure 335558DEST_PATH_IMAGE012
, distribute to each direction of search and form corresponding search terms
Figure 184697DEST_PATH_IMAGE014
(
Figure 2013104234649100002DEST_PATH_IMAGE015
), determine search listing:
Figure 279692DEST_PATH_IMAGE016
In formula:
Figure 2013104234649100002DEST_PATH_IMAGE017
it is direction of search set
Figure 260286DEST_PATH_IMAGE008
in element;
Step 7, each search terms is carried out to transient stability calculating, according to transient stability result of calculation calculating target function value
Figure 833350DEST_PATH_IMAGE018
, determine target function value minimum in the current iteration process
Figure 2013104234649100002DEST_PATH_IMAGE019
:
Figure 989525DEST_PATH_IMAGE020
If step 8 target function value be less than the target function value of iteration point
Figure 2013104234649100002DEST_PATH_IMAGE021
, upgrade iteration point
Figure 208465DEST_PATH_IMAGE022
,
Figure 2013104234649100002DEST_PATH_IMAGE023
,
Figure 975432DEST_PATH_IMAGE024
, increase step-size in search simultaneously
Figure 986114DEST_PATH_IMAGE012
, forward step 5 to; Otherwise reduce step-size in search
Figure 422911DEST_PATH_IMAGE012
, forward step 5 to;
Step 9, the cutting load controlled quentity controlled variable that obtains according to step 5 are as the scheme of electrical network cutting load control measure, in order to control the excision amount of each load, to improve entire system economy and fail safe.
Further technical scheme is:
In described step 2, be to utilize the constraint method for transformation, Transient Stability Constraints is transformed, add in target function the calculating target function value as penalty term ,
Figure 778117DEST_PATH_IMAGE002
for the primary iteration point;
Former Transient Stability Constraints:
Figure 502360DEST_PATH_IMAGE026
Transform and obtain penalty term by constraint
Figure 2013104234649100002DEST_PATH_IMAGE027
and middle entry
Figure 110059DEST_PATH_IMAGE028
:
Figure 2013104234649100002DEST_PATH_IMAGE029
；
Figure 34765DEST_PATH_IMAGE030
；
Obtain target function
Figure 2013104234649100002DEST_PATH_IMAGE031
:
In formula:
Figure 2013104234649100002DEST_PATH_IMAGE033
mean each load excision power sum;
Figure 332071DEST_PATH_IMAGE034
be the penalty function factor, get very large number;
Figure 2013104234649100002DEST_PATH_IMAGE035
it is the number of loading in system;
( ) be the former meritorious power of each load;
Figure 525603DEST_PATH_IMAGE038
(
Figure 579010DEST_PATH_IMAGE037
) for each ratio of loading and excising, as the controlled quentity controlled variable of each load;
Figure 2013104234649100002DEST_PATH_IMAGE039
it is each controlled quentity controlled variable
Figure 949948DEST_PATH_IMAGE038
the dominant vector formed;
Figure 774816DEST_PATH_IMAGE040
it is the moment of taking the cutting load control measure;
Figure DEST_PATH_IMAGE041
it is the emulation terminal moment;
Figure 67257DEST_PATH_IMAGE042
it is the relative merit of the generator angle upper limit;
Figure DEST_PATH_IMAGE043
it is generator number of units in system;
Figure 393196DEST_PATH_IMAGE044
it is the center inertia size of whole system generator;
Figure 680958DEST_PATH_IMAGE046
it is the merit angle of each generator;
Figure DEST_PATH_IMAGE047
Figure 801361DEST_PATH_IMAGE046
it is the inertia time constant of each generator.
Described step 3 is the sensitivity to target function of cutting load controlled quentity controlled variable by calculating each load, gets the sensitivity maximum
Figure 190885DEST_PATH_IMAGE003
individual load (
Figure 586094DEST_PATH_IMAGE007
but) as cutting load;
Sensitivity calculations:
In formula:
Figure DEST_PATH_IMAGE049
(
Figure 285246DEST_PATH_IMAGE037
) be the sensitivity of each load;
Figure 552279DEST_PATH_IMAGE050
that load is according to controlled quentity controlled variable
Figure 954442DEST_PATH_IMAGE038
after excision, the target function value that method is calculated according to claim 2;
it is Perturbation.
In described step 6, should judge each search terms
Figure 623320DEST_PATH_IMAGE014
whether in actual cutting load controlled quentity controlled variable scope, if, exceed the controlled quentity controlled variable scope, this search terms does not add search listing, otherwise this search terms is added to search listing.
In described step 7, described search terms is that a plurality of processes are carried out Parallel Transient Stability calculating simultaneously, realizes the parallel schema search, to accelerate to solve speed.
The invention has the beneficial effects as follows, the inventive method has adopted the parallel schema searching method, and utilize the constraint method for transformation, the optimum cutting load control problem of considering Complex System Models is converted into to unconstrained optimization problem, simultaneously according to searching for the characteristics of mutual decoupling zero between point in search procedure, increase substantially by parallel computing the speed of solving, the application for it in practical power systems is taken a firm foundation.With existing optimum load shedding control method, compare, the method that the present invention proposes mainly contains following improvement:
1, the method for initial value and parameter require lowly, and substantially there is no convergence problem, can effectively provide Stable Control Strategy.
2, the parallel schema search can reduce the dynamic simulation number of times of individual process effectively, thereby improves overall calculation speed.
3, can provide feasible cutting load control strategy in iterative process, and obtain the required dynamic simulation number of times of feasible solution seldom.
The accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the inventive method.
Fig. 2 is the validation verification figure that the inventive method obtains the cutting load control measure.
Fig. 3 is the efficiency comparison diagram of the inventive method under different process numbers.
Embodiment
Optimum load shedding control method based on the parallel schema search, comprise the steps:
1, utilize computer to read in electric power system data, comprise line parameter circuit value, generator parameter, load parameter etc.;
2, electric power system is carried out to initial emulation, obtain each merit angle of every generator constantly, utilize the constraint transformation technology, Transient Stability Constraints is transformed, add in target function the calculating target function value as penalty term
Figure 724010DEST_PATH_IMAGE001
,
Figure 195443DEST_PATH_IMAGE002
for the primary iteration point, making optimum cutting load control becomes a unconstrained optimization problem;
Former Transient Stability Constraints:
Figure 476569DEST_PATH_IMAGE026
Transform and obtain penalty term by constraint
Figure 718194DEST_PATH_IMAGE027
and middle entry
Figure 727738DEST_PATH_IMAGE028
:
Figure 105630DEST_PATH_IMAGE029
；
Figure 50583DEST_PATH_IMAGE030
；
Obtain target function
Figure 513926DEST_PATH_IMAGE031
:
Figure 858320DEST_PATH_IMAGE032
In formula:
Figure 825139DEST_PATH_IMAGE033
mean each load excision power sum;
Figure 393523DEST_PATH_IMAGE034
be the penalty function factor, get very large number;
Figure 953949DEST_PATH_IMAGE035
it is the number of loading in system;
Figure 836454DEST_PATH_IMAGE036
( ) be the former meritorious power of each load;
(
Figure 366158DEST_PATH_IMAGE037
) for each ratio of loading and excising, as the controlled quentity controlled variable of each load;
Figure 52355DEST_PATH_IMAGE039
it is each controlled quentity controlled variable
Figure 728187DEST_PATH_IMAGE038
the dominant vector formed;
Figure 841636DEST_PATH_IMAGE040
it is the moment of taking the cutting load control measure;
it is the emulation terminal moment;
it is the relative merit of the generator angle upper limit;
Figure 131300DEST_PATH_IMAGE043
it is generator number of units in system;
Figure 9126DEST_PATH_IMAGE044
it is the center inertia size of whole system generator;
Figure 156074DEST_PATH_IMAGE045
Figure 652914DEST_PATH_IMAGE046
it is the merit angle of each generator;
Figure 834497DEST_PATH_IMAGE047
Figure 162186DEST_PATH_IMAGE046
it is the inertia time constant of each generator;
3, the cutting load controlled quentity controlled variable of calculating each load is to target function sensitivity, gets the sensitivity maximum individual load (
Figure 972327DEST_PATH_IMAGE007
but) as cutting load;
Sensitivity calculations:
Figure 8416DEST_PATH_IMAGE048
In formula:
Figure 103411DEST_PATH_IMAGE049
( ) be the sensitivity of each load;
Figure 188227DEST_PATH_IMAGE050
that load is according to controlled quentity controlled variable
Figure 344402DEST_PATH_IMAGE038
after excision, the target function value calculated according to the described method of step 2; being Perturbation, is a very little parameter;
4, the parameter of initialization parallel schema search, comprise iterations
Figure 219134DEST_PATH_IMAGE004
, search precision
Figure 127048DEST_PATH_IMAGE005
with the initial ranging step-length
Figure 481937DEST_PATH_IMAGE006
, iteration point
Figure 449893DEST_PATH_IMAGE002
but in comprise each cutting load (
Figure 14866DEST_PATH_IMAGE007
) cutting load controlled quentity controlled variable information, but according to the cutting load that step 3 is determined, determine the set of the direction of search
Figure 726470DEST_PATH_IMAGE008
:
Figure 857237DEST_PATH_IMAGE009
In formula:
Figure 589570DEST_PATH_IMAGE010
but be cutting load
Figure 376260DEST_PATH_IMAGE011
corresponding unit vector, dimension with
Figure 891555DEST_PATH_IMAGE002
identical;
If 5 current search step-lengths
Figure 876829DEST_PATH_IMAGE012
be less than search precision
Figure 530795DEST_PATH_IMAGE005
, finish the parallel schema search, obtain optimum cutting load controlled quentity controlled variable, forward step 9 to; Otherwise forward step 6 to;
6, according to current iteration point
Figure 335940DEST_PATH_IMAGE013
and step-size in search
Figure 858188DEST_PATH_IMAGE012
, distribute to each direction of search and form corresponding search terms (
Figure 178628DEST_PATH_IMAGE015
), determine search listing:
Figure 330124DEST_PATH_IMAGE016
Figure DEST_PATH_IMAGE053
In formula:
Figure 390484DEST_PATH_IMAGE017
it is direction of search set
Figure 350350DEST_PATH_IMAGE008
in element;
Should judge each search terms
Figure 343189DEST_PATH_IMAGE014
whether in actual cutting load controlled quentity controlled variable scope, such as
Figure 122926DEST_PATH_IMAGE052
if, exceed the controlled quentity controlled variable scope, this search terms does not add search listing, otherwise this search terms is added to search listing, guarantees that like this in the parallel schema search procedure, each search all tallies with the actual situation;
7, each search terms is carried out to transient stability calculating, according to transient stability result of calculation calculating target function value , determine target function value minimum in the current iteration process
Figure 801349DEST_PATH_IMAGE019
:
Figure 623812DEST_PATH_IMAGE020
Owing between each search terms, there is no coupled relation, so search terms can a plurality of processes carry out Parallel Transient Stability calculating simultaneously, by parallel computing, accelerates to solve speed;
If 8 target function values
Figure 484320DEST_PATH_IMAGE019
be less than the target function value of iteration point
Figure 683220DEST_PATH_IMAGE021
, upgrade iteration point
Figure 820941DEST_PATH_IMAGE022
,
Figure 814304DEST_PATH_IMAGE023
,
Figure 568634DEST_PATH_IMAGE024
, increase step-size in search simultaneously
Figure 915433DEST_PATH_IMAGE012
, forward step 5 to; Otherwise reduce step-size in search
Figure 438818DEST_PATH_IMAGE012
, forward step 5 to;
9, the cutting load controlled quentity controlled variable obtained according to step 5 is as the scheme of electrical network cutting load control measure, in order to control the excision amount of each load, to improve entire system economy and fail safe.
Below in conjunction with accompanying drawing, embodiments of the invention are elaborated, flow chart of the present invention is as shown in Figure 1.
application example:
The inventive method is considered the electrical network of 145 nodes, include 50 generators with excitation, 64 loads, by step 1, utilize computer to read in electric power system data, the parameter of the parameter of the parameter that comprises 401 circuits, 50 generators, 64 loads etc., the generator model of system all adopts the high-order dynamic model, and have a small amount of or governing system, initial time that the short trouble between No. 6 nodes and No. 7 nodes occurs, cut fault moment and be
Figure 71925DEST_PATH_IMAGE054
=0.0583s, cutting load is controlled constantly =0.15s.
By step 2, load excision controlled quentity controlled variable under system initial state
Figure 854253DEST_PATH_IMAGE002
=0(does not take any control measure), system is carried out to initial emulation, obtain each merit angle of every generator constantly, the calculating target function value
Figure 825620DEST_PATH_IMAGE001
for 1.09277e12.
By step 3, calculate the cutting load controlled quentity controlled variable of each load to target function sensitivity, wherein Perturbation
Figure 426365DEST_PATH_IMAGE051
get 0.01, but 10 loads getting the sensitivity maximum in all loads as cutting load, its concrete Sensitirity va1ue is as shown in table 1:
The sensitivity of each load of table 1 (front 10)
Figure DEST_PATH_IMAGE055
Get emulation terminal during optimization and constantly be 3s, step-length is 0.008333s, and the maximum merit angle that allows is 160 °.
By step 4, the parameter of initialization parallel schema search, comprise iterations , search precision , the initial ranging step-length
Figure DEST_PATH_IMAGE057
.But owing to getting the cutting load number
Figure 279549DEST_PATH_IMAGE058
, each load can have and increases, subtracts two directions of search, so one has 20 unit locating vectors and forms direction of search set
Figure 520038DEST_PATH_IMAGE008
.
By step 5-8, utilize the continuous iteration of parallel schema searching method, make target function constantly reduce to upgrade iteration point information, until step 5 judgement current search step-length simultaneously
Figure 470676DEST_PATH_IMAGE012
be less than search precision
Figure 946657DEST_PATH_IMAGE005
, exit search procedure.Particularly in step 7, in order to realize that parallel schema search quickening solves speed, open altogether 8 processes in optimizing process, comprise 1 host process and 7 subprocesss, host process is responsible for allocating task, is upgraded correlated variables etc., and 7 sub-task parallelism ground calculate the target function value of each search point.In this process, subprocess need to call PSS/E and come each merit angle constantly of each generator in the acquisition system, thereby obtains target function value
Figure 299141DEST_PATH_IMAGE031
.
In step 9, the cutting load controlled quentity controlled variable after optimizing, as the scheme of electrical network cutting load control measure, is controlled the excision amount of each load, to improve entire system economy and fail safe.But concrete each cutting load excision ratio optimization result is as shown in table 2:
But the excision ratio of each cutting load of table 2
The Transient Stability Criterion that the method is used is no more than a certain angle for the deviation in each generator's power and angle and the center of inertia, and Fig. 2 has made the variation of unstability generator power-angle curve before and after cutting load is controlled in the system.Can find out intuitively after cutting load that by figure the vibration of power-angle curve constantly reduces relatively, in fact the simulation result of longer time has also illustrated that power-angle curve is to tend towards stability, and this has proved the validity of the cutting load control strategy that the method provides.
For the effect of parallel search is described, table 3 has been listed the process number
Figure 710530DEST_PATH_IMAGE060
(now being equivalent to serial search) with the time each process dynamic simulation number of times that need to carry out, Fig. 3 has made when adopting different processes to count Optimization Solution, the change curve of the dynamic simulation number of times of individual process.
Proportionate relationship in table 3 is the ratio of former and later two dynamic simulation number of times.Can find out, 10 times of left and right of individual process simulation times when the dynamic simulation number of times of serial search is 7 processes, so parallel search can significantly reduce the dynamic simulation number of times that process needs, thus significantly reduced whole computing time.
The dynamic simulation number of times contrast of a process of table 3
Figure 492673DEST_PATH_IMAGE062
Another characteristics of the method are to obtain fast a feasible cutting load control strategy.In fact, after carrying out parallel search at every turn and upgrading iteration point, this iteration point is exactly a new control strategy, and now the stability of system also can be obtained by Dynamic Simulation Results before.When the economy to cutting load is less demanding, in order to obtain more rapidly making the control strategy of system stability, can get the middle result of iteration fully as actual control strategy.In the actual optimization process, open 7 sub-task parallelism search findings, individual process only needs a transient stability to calculate just can obtain a Stable Control Strategy.

Claims (5)

1. the optimum load shedding control method based on the parallel schema search, is characterized in that, comprises the steps:
Step 1, utilize computer to read in electric power system data, comprise line parameter circuit value, generator parameter, load parameter;
Step 2, electric power system is carried out to initial emulation, obtain each merit angle of every generator constantly, the calculating target function value
Figure 2013104234649100001DEST_PATH_IMAGE002
,
Figure 2013104234649100001DEST_PATH_IMAGE004
for the primary iteration point;
Step 3, in all loads, choose
Figure 2013104234649100001DEST_PATH_IMAGE006
but individual cutting load;
The parameter of step 4, the search of initialization parallel schema, comprise iterations
Figure 2013104234649100001DEST_PATH_IMAGE008
, search precision
Figure 2013104234649100001DEST_PATH_IMAGE010
with the initial ranging step-length , iteration point but in comprise each cutting load (
Figure 2013104234649100001DEST_PATH_IMAGE014
) cutting load controlled quentity controlled variable information, but according to the cutting load that step 3 is determined, determine the set of the direction of search
Figure 2013104234649100001DEST_PATH_IMAGE016
:
Figure 2013104234649100001DEST_PATH_IMAGE018
In formula:
but be cutting load
Figure 2013104234649100001DEST_PATH_IMAGE022
corresponding unit vector, dimension with
Figure 978526DEST_PATH_IMAGE004
identical;
If step 5 current search step-length
Figure 2013104234649100001DEST_PATH_IMAGE024
be less than search precision , finish the parallel schema search, obtain optimum cutting load controlled quentity controlled variable, forward step 9 to; Otherwise forward step 6 to;
Step 6, according to current iteration point
Figure 2013104234649100001DEST_PATH_IMAGE026
and step-size in search
Figure 578058DEST_PATH_IMAGE024
, distribute to each direction of search and form corresponding search terms
Figure 2013104234649100001DEST_PATH_IMAGE028
(
Figure 2013104234649100001DEST_PATH_IMAGE030
), determine search listing:
Figure 2013104234649100001DEST_PATH_IMAGE032
In formula:
Figure 2013104234649100001DEST_PATH_IMAGE034
it is direction of search set
Figure 271208DEST_PATH_IMAGE016
in element;
Step 7, each search terms is carried out to transient stability calculating, according to transient stability result of calculation calculating target function value
Figure 2013104234649100001DEST_PATH_IMAGE036
, determine target function value minimum in the current iteration process
Figure 2013104234649100001DEST_PATH_IMAGE038
:
Figure 2013104234649100001DEST_PATH_IMAGE040
If step 8 target function value
Figure 660601DEST_PATH_IMAGE038
be less than the target function value of iteration point
Figure 2013104234649100001DEST_PATH_IMAGE042
, upgrade iteration point
Figure 2013104234649100001DEST_PATH_IMAGE044
,
Figure 2013104234649100001DEST_PATH_IMAGE046
,
Figure 2013104234649100001DEST_PATH_IMAGE048
, increase step-size in search simultaneously , forward step 5 to; Otherwise reduce step-size in search
Figure 967265DEST_PATH_IMAGE024
, forward step 5 to;
Step 9, the cutting load controlled quentity controlled variable that obtains according to step 5 are as the scheme of electrical network cutting load control measure, in order to control the excision amount of each load, to improve entire system economy and fail safe.
2. the optimum load shedding control method of searching for based on parallel schema according to claim 1, is characterized in that, in described step 2, to utilize the constraint method for transformation, Transient Stability Constraints is transformed, added in target function as penalty term, the calculating target function value
Figure 339472DEST_PATH_IMAGE002
,
Figure 521055DEST_PATH_IMAGE004
for the primary iteration point;
Former Transient Stability Constraints:
Figure 2013104234649100001DEST_PATH_IMAGE050
Figure 2013104234649100001DEST_PATH_IMAGE052
Transform and obtain penalty term by constraint
Figure 2013104234649100001DEST_PATH_IMAGE054
and middle entry
Figure 2013104234649100001DEST_PATH_IMAGE056
:
Figure 2013104234649100001DEST_PATH_IMAGE058
；
Figure 2013104234649100001DEST_PATH_IMAGE060
；
Obtain target function
Figure DEST_PATH_IMAGE062
:
Figure DEST_PATH_IMAGE064
In formula:
Figure DEST_PATH_IMAGE066
mean each load excision power sum;
be the penalty function factor, get very large number;
Figure DEST_PATH_IMAGE070
it is the number of loading in system;
(
Figure DEST_PATH_IMAGE074
) be the former meritorious power of each load;
Figure DEST_PATH_IMAGE076
(
Figure 379902DEST_PATH_IMAGE074
) for each ratio of loading and excising, as the controlled quentity controlled variable of each load;
Figure DEST_PATH_IMAGE078
it is each controlled quentity controlled variable the dominant vector formed;
Figure DEST_PATH_IMAGE080
it is the moment of taking the cutting load control measure;
Figure DEST_PATH_IMAGE082
it is the emulation terminal moment;
Figure DEST_PATH_IMAGE084
it is the relative merit of the generator angle upper limit;
Figure DEST_PATH_IMAGE086
it is generator number of units in system;
Figure DEST_PATH_IMAGE088
it is the center inertia size of whole system generator;
Figure DEST_PATH_IMAGE090
Figure DEST_PATH_IMAGE092
it is the merit angle of each generator;
Figure DEST_PATH_IMAGE094
Figure 799830DEST_PATH_IMAGE092
it is the inertia time constant of each generator.
3. the optimum load shedding control method based on parallel schema search according to claim 1, is characterized in that, described step 3 is the sensitivity to target function of cutting load controlled quentity controlled variable by calculating each load, gets the sensitivity maximum
Figure 101498DEST_PATH_IMAGE006
individual load ( but) as cutting load;
Sensitivity calculations:
Figure DEST_PATH_IMAGE096
In formula:
Figure DEST_PATH_IMAGE098
(
Figure 911508DEST_PATH_IMAGE074
) be the sensitivity of each load;
Figure DEST_PATH_IMAGE100
that load is according to controlled quentity controlled variable
Figure 750151DEST_PATH_IMAGE076
after excision, the target function value that method is calculated according to claim 2;
Figure DEST_PATH_IMAGE102
it is Perturbation.
4. the optimum load shedding control method of searching for based on parallel schema according to claim 1, is characterized in that, in described step 6, should judge each search terms whether in actual cutting load controlled quentity controlled variable scope,
Figure DEST_PATH_IMAGE104
if, exceed the controlled quentity controlled variable scope, this search terms does not add search listing, otherwise this search terms is added to search listing.
5. the optimum load shedding control method of searching for based on parallel schema according to claim 1 is characterized in that, in described step 7, described search terms is that a plurality of processes are carried out Parallel Transient Stability calculating simultaneously, realizes the parallel schema search, to accelerate to solve speed.
CN2013104234649A 2013-09-17 2013-09-17 Optimal load curtailment control method based on parallel mode searching Pending CN103457268A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2013104234649A CN103457268A (en) 2013-09-17 2013-09-17 Optimal load curtailment control method based on parallel mode searching

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2013104234649A CN103457268A (en) 2013-09-17 2013-09-17 Optimal load curtailment control method based on parallel mode searching

Publications (1)

Publication Number Publication Date
CN103457268A true CN103457268A (en) 2013-12-18

Family

ID=49739360

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2013104234649A Pending CN103457268A (en) 2013-09-17 2013-09-17 Optimal load curtailment control method based on parallel mode searching

Country Status (1)

Country Link
CN (1) CN103457268A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104573383A (en) * 2015-01-23 2015-04-29 浙江大学 Distributed evolution method suitable for comprehensive optimization model of building equipment
EP2894750A3 (en) * 2014-01-09 2015-11-04 Kabushiki Kaisha Toshiba Power system stabilizing device
CN109471100A (en) * 2018-10-16 2019-03-15 湖北航天技术研究院总体设计所 A kind of SAR doppler frequency rate estimation method and system
CN109586298A (en) * 2018-12-11 2019-04-05 国网山东省电力公司电力科学研究院 A kind of more direct current receiving end electric network synthetic load optimal control methods and system
CN115048690A (en) * 2022-05-09 2022-09-13 中存大数据科技有限公司 Cement sintering model optimization method based on pattern search

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110035077A1 (en) * 2009-08-10 2011-02-10 Korea Electric Power Corporation Distribution automation system for reactive power compensation and its voltage control method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110035077A1 (en) * 2009-08-10 2011-02-10 Korea Electric Power Corporation Distribution automation system for reactive power compensation and its voltage control method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
姚国强等: "最优切负荷控制问题的并行模式搜索算法", 《电力系统自动化》, vol. 36, no. 4, 25 February 2012 (2012-02-25), pages 11 - 15 *
毕兆东等: "基于数值积分法灵敏度的快速切负荷算法", 《电网技术》, vol. 26, no. 8, 31 August 2002 (2002-08-31) *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2894750A3 (en) * 2014-01-09 2015-11-04 Kabushiki Kaisha Toshiba Power system stabilizing device
US9768786B2 (en) 2014-01-09 2017-09-19 Kabushiki Kaisha Toshiba Power system stabilizing device
CN104573383A (en) * 2015-01-23 2015-04-29 浙江大学 Distributed evolution method suitable for comprehensive optimization model of building equipment
CN104573383B (en) * 2015-01-23 2018-02-09 浙江大学 A kind of distributed evolution method suitable for building equipment Integrated Optimization Model
CN109471100A (en) * 2018-10-16 2019-03-15 湖北航天技术研究院总体设计所 A kind of SAR doppler frequency rate estimation method and system
CN109586298A (en) * 2018-12-11 2019-04-05 国网山东省电力公司电力科学研究院 A kind of more direct current receiving end electric network synthetic load optimal control methods and system
CN109586298B (en) * 2018-12-11 2021-01-15 国网山东省电力公司电力科学研究院 Multi-direct-current receiving-end power grid comprehensive load optimization control method and system
CN115048690A (en) * 2022-05-09 2022-09-13 中存大数据科技有限公司 Cement sintering model optimization method based on pattern search

Similar Documents

Publication Publication Date Title
Feng et al. A comprehensive approach for preventive and corrective control to mitigate voltage collapse
CN103279639B (en) Receiving end Network Voltage Stability overall process Situation Assessment based on response and preventing control method
Sakthivel et al. Optimal location of SVC for voltage stability enhancement under contingency condition through PSO algorithm
CN102436534B (en) Simulating method for electric power system
CN102346442B (en) Method for carrying out parallel transient simulation computing on microgrid suitable for nodal analysis frameworks
CN103985058B (en) Available transfer capability calculation method based on improved multiple centrality-correction interior point method
CN103457268A (en) Optimal load curtailment control method based on parallel mode searching
CN103810646B (en) Improved projection integral algorithm based active power distribution system dynamic simulation method
CN104156542B (en) It is a kind of based on the active distribution system Simulation of stability method implicitly projected
Hajian et al. A chance-constrained optimization approach for control of transmission voltages
Chamorro et al. A network control system for hydro plants to counteract the non-synchronous generation integration
Bento Monitoring of the power system load margin based on a machine learning technique
CN103093037A (en) Electric system splitting fracture surface searching method based on master-slave problem alternating optimization
Rueda et al. Heuristic optimization based approach for identification of power system dynamic equivalents
Zenelis et al. A model-free sparse wide-area damping controller for inter-area oscillations
Djari et al. Voltage stability assessment using the VQ sensitivity and modal analyses methods
Wang et al. A three-machine equivalent system frequency response model and its closed-form solution
JP2017028915A (en) Arrangement calculation apparatus for power storage device, arrangement calculation method for power storage device, arrangement calculation system for power storage device, and program
Lin et al. A two-layer frequency control method for large-scale distributed energy storage clusters
Honari et al. A scalable blockchain-based smart contract model for decentralized voltage stability using sharding technique
Zhang et al. Impact of dynamic load models on transient stability-constrained optimal power flow
Melo et al. Voltage stability assessment using modal analysis based on power systems state estimation
Ding et al. Continuation power flow model for interconnected systems considering the electricity market influence and its corresponding distributed algorithm
CN104935249A (en) Photovoltaic power generation system stability verification method and apparatus
Liu et al. Comparison of multi-area reactive power optimization parallel algorithm based on Ward and REI equivalent

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20131218