CN104123683B - Electrical power system black-start scheme generation method based on dynamic programming - Google Patents
Electrical power system black-start scheme generation method based on dynamic programming Download PDFInfo
- Publication number
- CN104123683B CN104123683B CN201410377330.2A CN201410377330A CN104123683B CN 104123683 B CN104123683 B CN 104123683B CN 201410377330 A CN201410377330 A CN 201410377330A CN 104123683 B CN104123683 B CN 104123683B
- Authority
- CN
- China
- Prior art keywords
- state
- electric network
- load
- recovery
- unit
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 241001672018 Cercomela melanura Species 0.000 title claims abstract description 27
- 238000011084 recovery Methods 0.000 claims abstract description 80
- 230000008569 process Effects 0.000 claims abstract description 26
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 230000009467 reduction Effects 0.000 claims abstract description 10
- 230000005611 electricity Effects 0.000 claims description 31
- 238000012795 verification Methods 0.000 claims description 23
- 230000008859 change Effects 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 15
- NGVDGCNFYWLIFO-UHFFFAOYSA-N pyridoxal 5'-phosphate Chemical compound CC1=NC=C(COP(O)(O)=O)C(C=O)=C1O NGVDGCNFYWLIFO-UHFFFAOYSA-N 0.000 claims description 15
- 230000001052 transient effect Effects 0.000 claims description 15
- 238000002485 combustion reaction Methods 0.000 claims description 11
- 230000009194 climbing Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 5
- 206010048669 Terminal state Diseases 0.000 claims description 3
- 230000001351 cycling effect Effects 0.000 claims description 3
- 235000006508 Nelumbo nucifera Nutrition 0.000 claims 1
- 240000002853 Nelumbo nucifera Species 0.000 claims 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 claims 1
- 230000004069 differentiation Effects 0.000 claims 1
- 230000008901 benefit Effects 0.000 description 5
- 238000009825 accumulation Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000004075 alteration Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 206010016256 fatigue Diseases 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010008190 Cerebrovascular accident Diseases 0.000 description 1
- 208000006011 Stroke Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
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
- 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
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an electrical power system black-start scheme generation method based on dynamic programming. The electrical power system black-start scheme generation method adopts the idea of gradually deducing black-start recovery operations, operation sequences with low recovery efficiency or quality are reduced gradually by means of state reduction technology, so that the process of decision making after the full enumeration of recovery conditions and cohesive relations thereof of all elements of the whole grid, which is required in technology such as Petri network technology, is avoided, and efficiency of algorithm is improved while requirements for computing storage capacity are lowered. Without any hypothesis of black-start recovery phase, each recovery operation can be chosen freely from all types, optional combination of restart of a generator unit, grid recovery and load recovery is allowed to carry out according to needs, problems of over-high voltage of line nodes in early stage of black start and inadaptability of the grid to the load recovery in later stage of black start are easy to solve, and the electrical power system black-start scheme generation method is naturally adaptable to different power grid after-fault scenes.
Description
Technical field
The invention belongs to power system restoration technical field, more particularly, to a kind of electricity based on dynamic programming
Force system black-start scheme generation method.
Background technology
After power system blackstart is often referred to whole system because of the stoppage in transit of the reasons such as cascading failure, supply external network is independent of
With the help of electricity, by the startup of the unit with self-startup ability in system, the unit of non self starting is driven, progressively expanded
The recovery scope of big system power supply, finally realizes the recovery process of whole power system.The sixties in 20th century is so far all over the world
The a lot of large-scale blackouts for occurring show that there is local fault in modern power systems all the time and deal with improperly causes accident scope
The hidden danger of extension, therefore used as preventive measure, the formulation of power system blackstart prediction scheme is necessary all the time.Current this work
Still rely primarily on and be accomplished manually, can only often consider the recovery order of wherein several most important units, its efficiency and scheme are complete
Whole property is difficult in adapt to the application requirement in large-scale complex power grid, needs the reliable black-start scheme of research badly and automatically generates calculation
Method.
But actual electric network black starting-up consider factor is more and complexity, at least need during program decisions while considering large-scale fire
Group of motors subsidiary engine starts, rack is reconstructed and the problem of load restoration, therefore up to now, most of existing algorithm research
The solution of some of them subproblem is focused on, how the recovery decision-making of the whole black starting-up process of comprehensive coordinate is still difficult point.Wherein,
Distributed decision making aspect, although give based on many agencies, consider that the principal and subordinate of graded dispatching escorts Decision-making structures from one place to another, and apply Yushan Hill
Eastern electrical network, but it is not directed to the sub- power system restoration algorithm for refining；Centralized decision aspect, introduces Petri network as decision-making technic,
But slightly larger its Petri network structure of electrical network scale will become extreme complicated difficult to calculate.
Additionally, the target part for first starting by unit steepest recover rack, complete unit and restart, consider further that load restoration
Treatment principle, although substantially increase scheme search efficiency, feasible power system restoration side can be obtained in polynomial time
Case, but the algorithm is limited only to consider the recovery of most short supply path between electromotor in the early stage, therefore loading is not on path
Line node overtension is likely to result in when sufficient, its early stage has recovered rack may also be not suitable with the requirement of later stage load restoration,
Cause the prolongation of overall restoration time.
The content of the invention
For the disadvantages described above or Improvement requirement of prior art, the invention provides a kind of power train based on dynamic programming
System black-start scheme generation method, using the teaching of the invention it is possible to provide take the refinement restoration schedule scheme of all elements into account, improve efficiency of algorithm, drops
The low requirement to calculating storage, it is easy to overcome black starting-up early stage line node overtension, it is extensive that later stage rack is not suitable with load
Multiple problem, while having adapted to scene after different electric network faults naturally.
For achieving the above object, the invention provides a kind of power system blackstart scheme generation method, it is characterised in that
Comprise the steps：
(1) electric network information after read failure obtains initial electric network state s0, arrival electric network state s is represented with variable g (s)
It is most short accumulative time-consuming, initialize g (s0)=0, series k=0, original state collection S0={ s0, initialization terminal state set SEFor
Empty set, preferred plan T recovery timebestFor infinity；
(2) to kth level state set SkIn each state ski, corresponding recovery operation is determined, to these recovery operations
Transient state verification is carried out, to judge the safety of electric network state transient process；To each state skiFor, by each of verification
Kind of recovery operation can generate a new state, by these new states without being repeatedly added to+1 grade of state set S of kthk+1In；Its
In, i=1,2 ..., nk, nkFor kth level state set SkThe sum of middle electric network state；
(3) to+1 grade of state set S of kthk+1In each state carry out stable state verification, to judge that Power System Steady-state electricity refers to
Target safety, will be unable to the state by verifying and deletes from set；
(4) to+1 grade of state set S of kthk+1In each state s(k+1)j, calculate and reach the most short accumulative time-consuming of the state
g(s(k+1)j)=min { g (ski)+r(ski,s(k+1)j), and will most short accumulative time-consuming g (s(k+1)j) corresponding restoration path conduct
Arrival state s(k+1)jOptimal recovery path；Wherein, r (ski,s(k+1)j) represent from state skiTo s(k+1)jRecovery operation consumption
When, j=1,2 ..., nk+1, nk+1For+1 grade of state set S of kthk+1The sum of middle electric network state；
(5) if+1 grade of state set S of kthk+1Middle presence has completed the electric network state for recovering, then will be all these complete
SOT state of termination collection S is moved to into the electric network state for recoveringE, make TminThe electric network state for recovering has been completed to reach these most
It is short to add up time-consuming minima, if preferred plan T recovery timebest>Tmin, then T is madebest=Tmin, otherwise TbestIt is constant, perform
Step (6)；If+1 grade of state set S of kthk+1In there is no the electric network state for having completed to recover, then direct execution step (6)；
(6) if+1 grade of state set S of kthk+1For sky, or+1 grade of state set S of kthk+1The most short of middle remaining all states is tired out
Meter is time-consuming to be more than Tbest, then by SOT state of termination collection SEIn correspond to TbestOptimal recovery path as black-start scheme, eventually
Only calculate；Otherwise execution step (7)；
(7) to+1 grade of state set S of kthk+1The state of carrying out cuts down rearmounted k=k+1, return to step (2).
Preferably, in the step (2), to kth level state set SkIn each state ski, it is optional before transient state verification
Recovery operation at least includes following five kinds：
(A1) electric network state s is madekiIn a certain uncharged circuit a enter state " in charging ", the circuit must with electrical network in
The region that restores electricity be joined directly together；In the corresponding new electric network state for generating, the charging of circuit a in line status record B
State variable SBaCorresponding change, the accumulative charge period T of circuit aBaSet to 0, remaining power grid state information and skiIt is identical；
(A2) electric network state s is madekiIn a certain NBS units b do not restarted enter state " during subsidiary engine is restarted ", unit b is necessary
Meet following two conditions：One is in electric network state skiIn its generator terminal bus restored electricity；Two is electric network state skiIn
The whole network can raise generated output PupMore than the startup power of unit b；In the corresponding new electric network state for generating, electromotor record G
The starting state variable S of middle unit bGbCorresponding change, the accumulative of unit b restarts period TGbSet to 0, remaining power grid state information with
skiIt is identical；
(A3) electric network state s is madekiIn a certain subsidiary engine restart the NBS units c that finishes and enter " rise before steady combustion and exert oneself " state；
In the corresponding new electric network state for generating, the starting state variable S of unit c in electromotor record GGcCorresponding change, remaining electrical network shape
State information and skiIt is identical；
(A4) selected section load enters the state that restores electricity；In the corresponding new electric network state for generating, each node
The load power for restoring electricity is increased by self-defining allocation strategy, remaining power grid state information and skiIt is identical；
(A5) as electric network state skiIn deposit circuit in charging, restart in unit or the load that restoring electricity
When, default T is used according to issued Status Change instructionstepDuration proceeds follow-up quantity of state adjustment, generates new electricity
Net state, wherein, TstepRepresent user-defined power system restoration operating time step-length.
Preferably, in the recovery operation (A4), allocation strategy specifically includes following steps：
(B1) electric network state s is calculatedkiIn the whole network can raise generated output Pup, calculated load access limit value PLP=αLPup,
Wherein, αL≤ 1 is the default proportionality coefficient of algorithm；
(B2) according to electric network state skiLoad flow calculation is carried out, the whole network load bus is pressed into node voltage perunit value from big to small
Sequence；
(B3) the front N after sequence is selectedLIndividual load bus is used as both candidate nodes；NLValue principle be：If the whole network is not yet extensive
Multiple load total amount is less than or equal to PLP, then NL=nN, nNFor the sum of electrical network interior joint；If the still unrecovered load of the whole network is total
Amount is more than PLP, then NLValue should cause front NLThe still unrecovered load total amount of -1 node is less than or equal to PLP, front NLIndividual section
The still unrecovered load total amount of point is more than PLP；
(B4) loading that each both candidate nodes are specifically recovered is determined.
Preferably, if the burden with power increment that o-th both candidate nodes increase successively is (Do1,Do2,…DoMo), amount to MoIt is individual
The loading of grade, the step (B4) further includes following sub-step：
(C1) initialization index o=1, continues cycling through flag Fgoon=0；
(C2) if o-th both candidate nodes current loads restoration grade uo=Mo, then show that its all load has been enter into treating extensive
Multiple state, skips to step (C4)；Otherwise execution step (C3)；
(C3) the individual nodes of I (o) in o-th both candidate nodes correspondence electrical network, RNI(x)Recover for x-th both candidate nodes
The load power of power supply,For the u of o-th both candidate nodesoThe load increment of+1 stage load amount first order load amount earlier above,
IfLoad power R that then o-th both candidate nodes are being restored electricityNI(o)Increase one-level
Loading so thatJuxtaposition uo=uo+ 1, Fgoon=1, then execution step (C4)；It is otherwise straight
Connect execution step (C4)；
(C4) if o+1≤NL, then return to step after o=o+1 (C2), otherwise execution step (C5) are made；
(C5) if FgoonFor 1, then it is 1, F to reset ogoonFor return to step (C2) after 0, otherwise terminate circulation.
Preferably, the recovery operation (A5) further includes following sub-step：
(D1) the new electric network state s* for generating is initialized as and electric network state skiIdentical state；
(D2) all accumulative charging intervals for being in " in charging " status line increase a time in new electric network state s*
Step-length TstepIf the accumulative charging interval of arbitrary circuit is more than or equal to its default charging interval, its state is changed " to charge
Finish " state, add up charging interval clearing；
All accumulative reboot times for being in " during subsidiary engine is restarted " state unit increase a time in new electric network state s*
Step-length Tstep, its unit output is unit starting power PstartBy step-length PstepNegative value after discretization；If arbitrary unit is tired
Meter reboot time is more than or equal to its default starting time, changes its state for " subsidiary engine restart finish " state, accumulative to restart
Time resets；
It is all in new electric network state s* be in " rise before steady combustion and exert oneself " state unit exert oneself then by climbing rate before steady combustion with
Time step TstepProduct increase, if unit output is exerted oneself P more than or equal to its minimum technology after calculatingGminWith startup power
PstartDifference, then change its state for " unit restart finish " state, unit output is PGmin-PstartBy step-length PstepDiscretization
Value afterwards；
In the state recording L of the load of new electric network state s*, the load power for having restored electricity of q-th node is reset
PNq=PNq+RNq, and by load power R for restoring electricity of q-th nodeNqReset, wherein q=1,2 ..., nN, nNFor
The sum of electrical network interior joint.
Preferably, in the step (7), state is cut down and include successively non-dominant sequence reduction, hierarchical classification and monocular
Three steps are cut down in mark sequence；
Wherein, the execution flow process of hierarchical classification is as follows：
(E1) init state collection Sk+1In classification layer residing for all electric network states be the 1st layer；
(E2) classification layer v=1 is put, by default similar discriminant function f1() is to state set Sk+1In all electric network states
Carry out the first subseries；
(E3) if always classification number is more than classification number upper limit Cmax, then execution step (E4), otherwise execution step (E5)；
(E4) default distance function d is adoptedv() calculate between each state class of v layers and other state class away from
From summation, retain apart from the larger C of summationmaxIndividual state class, deletes other state class, and deleted state class is included
Electric network state from state set Sk+1In remove, terminate hierarchical classification flow process；
(E5) electric network state quantity is most in selected all state class one, makes v be equal to the residing classification layer of the state class,
If v<3, then v=v+1 is made, it is v layers to put classification layer in the state class residing for all electric network states, and similar is sentenced by default
Other function fv() reclassifies to this batch of electric network state, return to step (E3)；Otherwise terminate hierarchical classification flow process.
In general, by the contemplated above technical scheme of the present invention compared with prior art, with following beneficial effect
Really：
1st, flow process cover the whole network electromotor restart, rack recover and each load bus recovery decision-making, it is thus possible to
The refinement restoration schedule scheme of all elements is taken in offer into account.
2nd, using the thinking for progressively deducing black starting-up recovery operation, and recovery efficiency is constantly simplified by state reduction technology
Or the sequence of operation that quality is not good, thus need not be as the technologies such as Petri network in advance by the recovery condition of all elements of the whole network
And its be mutually linked relation and enumerated entirely and make decision again, efficiency of algorithm is improve, reduce the requirement to calculating storage.
3rd, be not based on any black starting-up Restoration stage it is assumed that often walk recovery operation can from all kinds unrestricted choice,
Therefore allow as needed to restart generating set, rack recovers and load restoration carries out combination in any, it is easy to overcome black starting-up
Early stage line node overtension, later stage rack is not suitable with the problem of load restoration, while having adapted to different electrical networks naturally
Scene after failure.
4th, the status switching instruction of element is separated with implementation procedure in recovery operation, it is thus possible to adapt to parallel recovery behaviour
Make.
5th, method flow and specific electric network state represent, recovery operation setting, transient state/steady state constraint verification content without
Close, therefore the electric network element considered in method can arbitrarily be increased and decreased on demand with verification project.
6th, the enforcement of state reduction technology avoids computation complexity excessively increases with the increase of electrical network scale, so that
This method can be efficiently applied in actual electric network, and single goal therein sequence reduction step is then provided by its index that sorts
Introduce the approach of manual decision's tendency.
Description of the drawings
Fig. 1 is that the power system blackstart scheme generation method flow process based on dynamic programming of the embodiment of the present invention is illustrated
Figure；
Fig. 2 is the generation schematic diagram of new state；
Fig. 3 is optimal recovery path schematic diagram；
Fig. 4 is the conceptual schematic view of hierarchical classification；
Fig. 5 is the line chart of the node system of IEEE5 machines 14；
Fig. 6 is the relation of the state class upper limit and Riming time of algorithm；
Fig. 7 is the relation of the state class upper limit and optimum decision making's recovery time；
Fig. 8 is the line chart of the node system of 10 machine of New England 39；
Fig. 9 is load accumulation recovery curve；
Figure 10 is power system restoration process voltage amplitude curve；
Figure 11 is circuit overload multiple block diagram；
Power system restoration situation when Figure 12 is the 42nd minute.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, and
It is not used in the restriction present invention.As long as additionally, technical characteristic involved in invention described below each embodiment
Not constituting conflict each other just can be mutually combined.
As shown in figure 1, the embodiment of the present invention is included based on the power system blackstart scheme generation method of dynamic programming
Following steps：
(1) electric network information after read failure obtains initial electric network state s0, and the target grid state after recovering is set
sT, represented with variable g (s) and reach the most short accumulative time-consuming of electric network state s, initialize g (s0)=0, adds up recovery operation number of times
(hereinafter referred to as " series ") k=0 and original state collection S0={ s0, initialization terminal state set SEFor empty set, preferred plan is extensive
Multiple time TbestFor infinity.
Wherein, electric network state s should at least include Generator Status record G, line status record B and load condition note
Record L tri- parts, i.e.,：
S=[G B L]
Notice that all of performance number and time value are required for according to given power and time step in electric network state record
PstepAnd TstepDiscretization is carried out, but participates in still being processed by continuous quantity during computing, explanation is not repeated below.Practical application
Middle PstepPreferably it is taken as the whole network NBS units (Non-Black Start Unit, not the unit with self-startup ability) startup power
And each node accesses the minima in load increment；TstepAdvise as the minima of each system element recovery time, but it is unsuitable
More than the minima of each system element recovery time, otherwise it is likely to result in recovery process and idle period occurs, be i.e. TstepRepresent
User-defined power system restoration operating time step-length.
Generator Status record recovery information I of the G by every unitGmConstitute, IGmIncluding unit starting state variable SGm、
Unit is accumulative to restart period TGmActive exert oneself P current with unitGm：
IGm=[SGm,TGm,PGm],(m∈[1,nG])
N in above formulaGFor generating set sum in electrical network.
Line status records recovery information I of the B by every circuitBpConstitute, including line charging state variable SBpAnd circuit
Accumulative charge period TBp：
IBp=[SBp,TBp],(p∈[1,nB])
N in formulaBFor the circuit sum in electrical network.
The state recording L of load by each node load restoration information vector ILqConstitute, including what is restored electricity
Load power PNqAnd load power R for restoring electricityNq：
ILi=[PNq,RNq],(q∈[1,nN])
N in formulaNFor the sum of electrical network interior joint.
If other needs are yet suffered from electrical network carries out the equipment (such as transformator etc.) of emphasis consideration in recovery process,
Above-mentioned data structure can be copied arbitrarily to be expanded.
(2) to kth level state set SkIn each state ski, corresponding recovery operation is determined, to these recovery operations
Transient state verification is carried out, to judge the safety of electric network state transient process；To each state skiFor, by each of verification
Kind of recovery operation can generate a new state, referring to Fig. 2, by these new states without being repeatedly added to+1 grade of state set of kth
Sk+1In；Wherein, i=1,2 ..., nk, nkFor kth level state set SkThe sum of middle electric network state.
Wherein, transient state verification mainly includes that switching overvoltage, self-excitation, active frequency dynamic respond three projects, need
Differentiated by electromagnetic transient simulation technology.
To kth level state set SkIn each state ski, optional recovery operation before transient state verification at least includes as follows
Five kinds：
(A1) electric network state s is madekiIn a certain uncharged circuit a enter state " in charging ", the circuit must with electrical network in
The region that restores electricity be joined directly together.In the corresponding new electric network state for generating, the charging of circuit a in line status record B
State variable SBaValue be changed to 1 (" in charging " state), and the accumulative charge period T of circuit aBaSet to 0, remaining electric network state
Information and skiIt is identical.Alteration command is only assigned in this operation to line status, is not actual execution, therefore the time-consuming r of its operation1=0.
(A2) electric network state s is madekiIn a certain NBS units b do not restarted enter state " during subsidiary engine is restarted ".Accordingly generating
New electric network state in, electromotor record G in unit b starting state variable SGbValue be changed to 1 (" during subsidiary engine is restarted " shape
State), and the accumulative of unit b restarts period TGbSet to 0, remaining power grid state information and skiIt is identical.This operation is only under set state
Up to alteration command, actual execution, therefore the time-consuming r of its operation are not2=0.
It should be noted that the selected NBS units b of subsidiary engine reboot operation has to comply with two conditions：The first is in electrical network shape
State skiIn its generator terminal bus restored electricity, it two is electric network state skiIn the whole network can raise generated output PupMore than machine
Startup power (the P of group bupFor electric network state skiMiddle the whole network unit is in time step TstepInterior maximum climbing power summation is deducted just
In loading R for restoring electricityNqAnd unit residue startup power E in " during subsidiary engine is restarted " stateGmRemaining work(afterwards
Rate).
(A3) electric network state s is madekiIn a certain subsidiary engine restart the NBS units c that finishes and enter " rise before steady combustion and exert oneself " state.
In the corresponding new electric network state for generating, the starting state variable S of unit c in electromotor record GGcValue be changed to 3 (" before steady combustion
Rise and exert oneself " state), remaining power grid state information and skiIt is identical.Alteration command is only assigned in this operation to set state, does not do reality
Perform, therefore the time-consuming r of its operation3=0.
(A4) selected section load enters the state that restores electricity.In the corresponding new electric network state for generating, each node
The load power for restoring electricity is increased by self-defining allocation strategy, remaining power grid state information and skiIt is identical.This operation
Incoming instruction is only assigned to sub-load, actual execution, therefore the time-consuming r of its operation is not4=0.
In view of actual electric network load access always be classified it is discrete, if what o-th both candidate nodes may increase successively
Burden with power increment is (Do1,Do2,…DoMo), amount to MoThe loading of individual grade, a kind of feasible load distribution strategy is basis
The higher load of the climbing capacity priority access place node voltage of the whole network unit, its detailed allocation flow is as follows：
(B1) electric network state s is calculatedkiIn the whole network can raise generated output Pup, calculated load access limit value PLPSuch as following formula：
PLP=αLPup
Wherein, αL≤ 1 is the default proportionality coefficient of algorithm, and the more little then load restoration of its value is slower, but reserve nargin
It is bigger, need to select by actual electric network service requirement.
(B2) according to electric network state skiLoad flow calculation is carried out, the whole network load bus is pressed into node voltage perunit value from big to small
Sequence.
(B3) the front N after sequence is selectedLIndividual load bus is used as both candidate nodes, NLValue principle be：If the whole network is not yet extensive
Multiple load total amount already less than or equal to PLP, then NL=nN；If the still unrecovered load total amount of the whole network is still greater than PLP, then NL's
Value should cause front NLThe still unrecovered load total amount of -1 node is less than or equal to PLP, front NLThe still unrecovered load of individual node is total
Amount is more than PLP。
(B4) loading that each both candidate nodes are specifically recovered is determined.Further include following sub-step：
(C1) initialization index o=1, continues cycling through flag Fgoon=0；
(C2) if o-th both candidate nodes current loads restoration grade uo=Mo, then show that its all load has been enter into treating extensive
Multiple state, makes and skip to after o=o+1 step (C4)；Otherwise execution step (C3)；
(C3) the individual nodes of I (o) in o-th both candidate nodes correspondence electrical network, RNI(x)Recover for x-th both candidate nodes
The load power of power supply,For the u of o-th both candidate nodesoThe load increment of+1 stage load amount first order load amount earlier above,
IfLoad power R that then o-th both candidate nodes are being restored electricityNI(o)Increase one-level
Loading, evenJuxtaposition uo=uo+ 1, Fgoon=1, then execution step (C4)；It is otherwise straight
Connect execution step (C4)；
(C4) if o+1≤NL, then return to step after o=o+1 (C2), otherwise execution step (C5) are made；
(C5) if FgoonFor 1, then it is 1, F to reset ogoonFor return to step (C2) after 0, otherwise terminate circulation.
It should be noted that electric network state s has to pass through (A5) operating all load powers for restoring electricity all clearly
Just allow to carry out (A4) operation once again after zero.
(A5) four operations more than are assigning for instruction, and change is electric network state skiIn recovery shape residing for certain element
State, and element state switching after its accumulative recovery time and exert oneself wait accumulation change then in this operation generation, thus
The physical meaning of this operation can be considered in Preset Time step-length TstepThe new command for changing grid equipment state is inside no longer sent, and
Actual execution is made to system mode change before this, so operation is taken as a time step, i.e. r5=Tstep.Simultaneously should
Work as attention, this operation and if only if electric network state skiIn deposit circuit in charging, restart in unit or recovering to supply
Just allow to carry out during the load of electricity, the flow process of the new electric network state of generation is as follows：
(D1) the new electric network state s* for generating is initialized as and electric network state skiIdentical state；
(D2) all accumulative charging intervals for being in " in charging " status line increase a time in new electric network state s*
Step-length TstepIf the accumulative charging interval of arbitrary circuit is more than or equal to its default charging interval, its state is changed " to charge
Finish " state, add up charging interval clearing；
All accumulative reboot times for being in " during subsidiary engine is restarted " state unit increase a time in new electric network state s*
Step-length Tstep, its unit output is taken as unit starting power PstartBy step-length PstepNegative value after discretization.If arbitrary unit
Accumulative reboot time is more than or equal to its default starting time, changes its state for " subsidiary engine restart finish " state, adds up weight
Open time clearing；
It is all in new electric network state s* be in " rise before steady combustion and exert oneself " state unit exert oneself then by climbing rate before steady combustion with
Time step TstepProduct increase, if unit output is exerted oneself P more than or equal to its minimum technology after calculatingGminWith startup power
PstartDifference, then change its state for " unit restart finish " state, unit output is taken as PGmin-PstartBy step-length PstepIt is discrete
Value after change；
In the state recording L of the load of new electric network state s*, the load power for having restored electricity of q-th node is reset
PNq=PNq+RNq, and by load power R for restoring electricity of q-th nodeNqReset, wherein q=1,2 ..., nN, nNFor
The sum of electrical network interior joint.
Because (A5) operation is usually directed to the change of each node injecting power of electrical network, therefore exerting oneself for each unit must be carried out
It is corresponding to adjust so that during self-defined network load changed power each unit output regulated quantity distribution principle, this specification
The feasible program of middle employing is that newly-increased load power is assigned to by the climbing capacity of unit " unit restart finish " state
Each unit.
It should be added that, due to (A1) (A2) (A3) (A4) four kinds of recovery operations are instruction issuing, only (A5)
Operation embodies the practical implementation of electric network element Status Change, therefore this method allows MIMD to perform simultaneously after assigning, energy
Enough strategies for covering multicomponent parallel recovery naturally.
If electric network state considers other electrical equipments in addition to electromotor, circuit, load, power system restoration in representing
Operation can also arbitrarily be expanded in addition at above-mentioned five kinds.
(3) to+1 grade of state set S of kthk+1In each state carry out stable state verification, to judge that Power System Steady-state electricity refers to
Target safety, will be unable to the state by verifying and deletes from set.
Wherein, stable state verification includes that the crash time verification of each unit starting and trend verify (i.e. power frequency steady state voltage width
Whether value is out-of-limit with Line Flow) two projects, wherein trend verification index can be by the stabilization result of electromagnetic transient simulation
Directly obtain.
It should be noted that transient state verification and stable state verification are relatively independent links in this method flow process, basis is thus susceptible to
Need addition or delete any number of verification project, with good expansibility.
(4) to+1 grade of state set S of kthk+1In each state s(k+1)j, by formula (1) arrival state s is calculated(k+1)jMost
Short accumulative time-consuming g (s(k+1)j) (wherein r (ski,s(k+1)j) represent from state skiTo s(k+1)jRecovery operation take), and will most
Short accumulative time-consuming g (s(k+1)j) corresponding restoration path as reach state s(k+1)jOptimal recovery path, referring to Fig. 3.Wherein,
J=1,2 ..., nk+1, nk+1For+1 grade of state set S of kthk+1The sum of middle electric network state.
g(s(k+1)j)=min { g (ski)+r(ski,s(k+1)j)} (1)
(5) if+1 grade of state set S of kthk+1Middle presence has completed the electric network state for recovering, then will be all these complete
SOT state of termination collection S is moved to into the electric network state for recoveringE, make TminThe electric network state for recovering has been completed to reach these most
It is short to add up time-consuming minima, if preferred plan T recovery timebest>Tmin, then T is madebest=Tmin, otherwise TbestIt is constant, perform
Step (6)；If+1 grade of state set S of kthk+1In there is no the electric network state for having completed to recover, then direct execution step (6)；
(6) if+1 grade of state set S of kthk+1For sky, or+1 grade of state set S of kthk+1The most short of middle remaining all states is tired out
Meter is time-consuming to be more than Tbest, then by SOT state of termination collection SEIn correspond to TbestOptimal recovery path as black-start scheme, eventually
Only calculate；Otherwise execution step (7).
(7) to+1 grade of state set S of kthk+1The state of carrying out cuts down rearmounted k=k+1, return to step (2).
State is cut down and cuts down three steps including non-dominant sequence reduction, hierarchical classification and single goal sequence successively, its
It is to provide approach to introduce decision-making tendentiousness to implement purpose, is in all the time with the newly-generated status number of season this method controllable
In the range of ensureing algorithm performs efficiency.
First in non-dominant sequence is cut down, multiple electric network state indexs are predefined according to user's decision-making tendency, to shape
State collection Sk+1In all electric network states calculate these desired values, only retain and be wherein in Pareto forward position (Pareto
Frontier state), remaining electric network state is from state set Sk+1Middle deletion.(due to Pareto forward position and the reality of non-dominated ranking
Applying method is common in existing document data, is not detailed herein), five electric network state indexs C are recommended herein1～C5It is as follows：
In above formula：1{·}For indicative function, g (s) for state s accumulation recovery time, UmaxAnd UminRespectively to save in state s
The maximum and minima of point voltage perunit value；PLLmaxTo have recovered the maximum line loss value of circuit in state s；εVWith εLLFor normal
Number, only prevents denominator from adding for 0, and its value should be fully little；Index C1～C5Physical meaning be respectively in the unit interval and open
The average line way that begins to recover, start the average unit number restarted, the loading, the voltage that recover in the unit interval in the unit interval
Deviate the degree and line losses indices of reference value, more big then electric network state s is more excellent in the index meaning for index.C1～C3More
Comprehensively characterize recovery efficiency, C4～C5The quality of electrical network current operating conditions is then weighed.
Secondly electric network state is split as by its similarity by hierarchical classification different classes of, its batch total is less than default
Classification upper limit Cmax, needing to predefine multiple " classification layers " in being embodied as, its concept is referring to Fig. 4, state in different classification layers
Classification using different finenesses sorting criterion, layer numbering be bigger, and corresponding sorting criterion is finer, and per layer default one is same
Class discriminant function remembers xth layer for judging whether two electric network states belong to same class in the criteria for classification of this layer, here
Similar discriminant function is fx(), at the same per layer also need define one with layer between class distance function to quantify same layer in different conditions
The similarity degree of class, the distance function that xth layer is remembered here is dx(), provides here a kind of implementation example of 3 classification layer as follows：
S ' represents the different electric network state of any two with s in above formula, it is noted that the 1st layer of discriminant function and distance function
The only charged state of meter and power network line, and the 2nd, 3 layers are then added successively on its basis to electrical network set state, load restoration
The comparison of state, it is thus achieved that classification resolution is stepped up.
According to defined above, the execution flow process of hierarchical classification is as follows：
(E1) init state collection Sk+1In classification layer residing for all electric network states be the 1st layer；
(E2) classification layer v=1 is put, by default similar discriminant function f1() is to state set Sk+1In all electric network states
Carry out the first subseries；
(E3) if always classification number is more than classification number upper limit Cmax, then execution step (E4), otherwise execution step (E5)；
(E4) default distance function d is adoptedv() calculate between each state class of v layers and other state class away from
From summation, retain apart from the larger C of summationmaxIndividual state class, deletes other state class, and deleted state class is included
Electric network state from state set Sk+1In remove, terminate hierarchical classification flow process.
(E5) electric network state quantity is most in selected all state class one, makes v be equal to the residing classification layer of the state class,
If v<3, then v=v+1 is made, it is v layers to put classification layer in the state class residing for all electric network states, and similar is sentenced by default
Other function fv() reclassifies to this batch of electric network state, return to step (E3)；Otherwise terminate hierarchical classification flow process.
State set S after hierarchical classification executionk+1In state will be included into quantity less than CmaxState class in, while really
The state characteristic polymorphic between these state class is protected.
The last sequence by single goal that state is cut down passes through final index CFThe electric network state of each apoplexy due to endogenous wind is carried out
Sequence, only retains the optimal front R of desired value in suchmaxA kind of individual state, feasible selection is quality of voltage index (desired value
It is the bigger the better), be shown below (U in formulaqFor the voltage perunit value of q-th node, nNrecFor the whole network, recovery nodes are total)：
In sum, the state of this method is cut down technology and is introduced decision-making tendentiousness, led to by non-dominant sequence reduction
Cross hierarchical classification and single goal sequence is cut down and guaranteeing that electric network state is diversified while guarantee state set Sk+1In state sum
Less than CmaxRmax, hereby it is ensured that the efficiency of decision method.
It should be noted that the index of three step employings can according to actual needs be adjusted or mend in state reduction
Fill, the scheme of the present embodiment is only one of which implementation.
To make those skilled in the art more fully understand the present invention, with reference to power train of the specific embodiment to the present invention
System black-start scheme generation method is described in detail.
First, algorithm efficiency of the practice analysis
Carry out validation verification to the efficiency of the practice of the inventive method below, hardware environment is Core Duo T6500
2.10Ghz, DDR3 RAM 2.00GB, algorithm implementation is C++ and Matlab hybrid programmings, electric network state, recovery operation,
Unit output allocation strategy, load restoration strategy, state are cut down each index for adopting and are consistent with the explanation in flow process above,
Other complement and amendmentses are not made.
In Analysis of operation efficiency, the example for adopting is the node system of IEEE4 machines 14, and its line chart is as shown in figure 5, concrete system
System parameter is using associated data files in the tool kits of matpower 4.0.Example all assumes that system initial state is completely black when testing
(all devices power failure), each node load presses the discrete values of 2MW in system, it is allowed at most 3 circuit parallel recoveries, algorithm parameter
Take and determine Pstep=2MW, Tstep=2min, αL=1.0 arrange constant, do not consider that transient state is verified, the Load flow calculation needed for stable state verification
Completed by calling the tool kits of matpower 4.0.
Inventive algorithm adoption status cuts down technology to ensure efficiency of algorithm, wherein there are two arrange parameters：State class
Upper limit CmaxAnd status number upper limit R of each state classmax.Pass through adjustment algorithm Verification on the node systems of IEEE 14
The impact result of the two is as shown in Fig. 6, Fig. 7 and table 1.
The dynamic programming algorithm of state reduction is not performed due to long operational time (run time was more than 10 hours), consumption
It is big with internal memory, therefore almost without practical value.As seen from Figure 6, raising of the enforcement that state is cut down to algorithm speed is served
Conclusive effect, and in general the state class upper limit is less, Riming time of algorithm is shorter.
As seen from Figure 7, the restriction of state class quantity may cause a certain degree of deterioration of result of decision appearance (known optimal
Recovery time is 76min, enforcement state cut down after time for obtaining of decision-making substantially in the range of 80～84min, much smaller than each unit
Upper limit 361min that part sequentially adds up recovery time).And due to there may be sensitive branch, too low shape in example decision making process
The state number upper limit will cause unstable (blank parts represent that algorithm search scheme fails in two block diagrams) of the result of decision.
The R of table 1maxEffect test
In upper table, Cmax=35.
From table 1, if the state classification number upper limit selects reasonable, then the state upper limit for retaining in each classification is substantially not
The result of decision is affected, but is conducive to ground further to improve speed of decision.
2nd, decision scheme benefit analysiss
Below the advantage to the inventive method decision scheme is analyzed, using the machine 39 of New England 10 as shown in Figure 8
Node system is discussed, and takes algorithm parameter for Pstep=2MW, Tstep=2min, αL=1.0, Cmax=30, Rmax=1 are right
The node system of 10 machine of New England 39 carries out black starting-up decision-making, can be calculated optimal recovery scheme and takes through 1049s and is
456min, much smaller than upper limit 1411min that each element recovery time sequentially adds up, now carries out labor as follows with regard to its result：
1) equipment restoration schedule
Each unit restoration schedule is as shown in table 2 below in system.Each unit reboot time meets the default crash time about
Beam, wherein G9～10 are because generator terminal supply line susceptance is excessive, G4～7 to black starting-up unit electrical distance due to causing too far
Reboot time is later.
The generating set reboot time of table 2
2) load restoration process
Load accumulation recovery curve (not counting each set auxiliary machinery startup power) of black starting-up process is as shown in Figure 9, it is seen that
Load restoration has run through the recovery process of whole black starting-up in Optimal Decision-making.Curve has gentle section at number, actually corresponds to
Start and carry out the process that output distribution is distributed with Optimal Power Flow between unit in climbing before unit and steady combustion：Wherein 40～64min,
It is adjusted with G2, G8 unit for G1 at 80～96min；It is then to have been turned between unit and G3～7 unit at 142～242 minutes
Regulation process.
Because part of generating units has reached EIAJ after 270 minutes, and the further load restoration of part of nodes needs
Voltage after G9 and G10 are restarted is maintained, and load access speed totally slows down.
3) voltage magnitude
The maximum of whole network voltage amplitude and minima curve are as shown in Figure 10 in recovery operation overall process, it is seen that each node
All the time can be maintained in the normal range of 0.9～1.1p.u., wherein 0～70 minute stage is due to the new light-loaded circuit for charging
More, voltage max is higher, and the voltage high level for occurring for 138 and 196 minutes is then respectively due to 43, No. 45 big susceptance circuits fill
Caused by electric.
4) operating mode after recovering
Electrical network operating mode after black starting-up recovers is as shown in table 3 below with the contrast of former operating condition.
Electrical network operating mode after the recovery of table 3
From table 3 and Figure 11, each unit output of electrical network after recovering is performed rationally using the inventive method, via net loss
It is upper close with former operating mode, it can thus be assumed that recovery operation is more successful.
5) decision scheme advantage
First, the inventive method suppresses the overvoltage of initial stage supply path by proper restoration load, otherwise if do not consider
The access of load, then will appear from serious overvoltage during supply path is set up, and its maximum value changes is as shown in Figure 10,
When 20 minutes, 4,5,6,16, No. 17 nodes start slight voltage out-of-limit occur, and 3, No. 18 nodes are also reached on voltage magnitude
Boundary；When 22 minutes, remaining node voltage is all serious out-of-limit (maximum 1.226p.u.) in addition to No. 30 nodes, and black starting-up cannot be after
It is continuous to carry out.
Second, the inventive method can be taken into account at the black starting-up initial stage recovers general outside the necessary path for restarting NBS units
Logical load, as shown in figure 12 (recovered part overstriking is represented), when 42 minutes, in addition to G2 and G8, remaining electromotor is due to power supply
During path is charged or because overvoltage problem can not obtain instant recovery, and the available active of existing unit has obvious redundancy,
Therefore power supply has been recovered to important load centers such as 7,8,18 nodes in algorithm decision-making, has not only improved and make full use of unit to power
Time slot during the foundation of path, climbing before also surely firing for unit provides more available balanced loads.
The recovery of above-mentioned analytic explanation electromotor, circuit and load advantage on demand alternately, and the present invention is proposed
Method effectively this can be coordinated.
As it will be easily appreciated by one skilled in the art that the foregoing is only presently preferred embodiments of the present invention, not to
The present invention, all any modification, equivalent and improvement made within the spirit and principles in the present invention etc. are limited, all should be included
Within protection scope of the present invention.
Claims (6)
1. a kind of power system blackstart scheme generation method, it is characterised in that comprise the steps：
(1) electric network information after read failure obtains initial electric network state s0, arrival electric network state s is represented most with variable g (s)
It is short accumulative time-consuming, initialize g (s0)=0, series k=0, original state collection S0={ s0, initialization terminal state set SEFor sky
Collection, preferred plan T recovery timebestFor infinity；
(2) to kth level state set SkIn each state ski, determine corresponding recovery operation, these recovery operations are carried out
Transient state is verified, to judge the safety of electric network state transient process；To each state skiFor, by verification each is extensive
Multiple operation can generate a new state, by these new states without being repeatedly added to+1 grade of state set S of kthk+1In；Wherein, i
=1,2 ..., nk, nkFor kth level state set SkThe sum of middle electric network state；
(3) to+1 grade of state set S of kthk+1In each state carry out stable state verification, to judge Power System Steady-state electricity index
Safety, will be unable to the state by verifying and deletes from set；
(4) to+1 grade of state set S of kthk+1In each state s(k+1)j, calculate the most short accumulative time-consuming g for reaching the state
(s(k+1)j)=min { g (ski)+r(ski,s(k+1)j), and will most short accumulative time-consuming g (s(k+1)j) corresponding restoration path is used as arriving
Up to state s(k+1)jOptimal recovery path；Wherein, r (ski,s(k+1)j) represent from state skiTo s(k+1)jRecovery operation take,
J=1,2 ..., nk+1, nk+1For+1 grade of state set S of kthk+1The sum of middle electric network state；
(5) if+1 grade of state set S of kthk+1Middle presence has completed the electric network state for recovering, then completed extensive by all these
Multiple electric network state is moved to SOT state of termination collection SE, make TminTire out to reach these the most short of electric network state for having completed to recover
The time-consuming minima of meter, if preferred plan T recovery timebest>Tmin, then T is madebest=Tmin, otherwise TbestIt is constant, execution step
(6)；If+1 grade of state set S of kthk+1In there is no the electric network state for having completed to recover, then direct execution step (6)；
(6) if+1 grade of state set S of kthk+1For sky, or+1 grade of state set S of kthk+1The most short accumulative consumption of middle remaining all states
Shi Junyi is more than Tbest, then by SOT state of termination collection SEIn correspond to TbestOptimal recovery path as black-start scheme, terminate meter
Calculate；Otherwise execution step (7)；
(7) to+1 grade of state set S of kthk+1The state of carrying out cuts down rearmounted k=k+1, return to step (2).
2. power system blackstart scheme generation method as claimed in claim 1, it is characterised in that right in the step (2)
Kth level state set SkIn each state ski, optional recovery operation is at least including following five kinds before transient state verification：
(A1) electric network state s is madekiIn a certain uncharged circuit a enter state " in charging ", the circuit must with electrical network in
The region that restores electricity is joined directly together；In the corresponding new electric network state for generating, the charged state of circuit a in line status record B
Variable SBaCorresponding change, the accumulative charge period T of circuit aBaSet to 0, remaining power grid state information and skiIt is identical；
(A2) electric network state s is madekiIn a certain NBS units b do not restarted enter state " during subsidiary engine is restarted ", unit b has to comply with
Following two conditions：One is in electric network state skiIn its generator terminal bus restored electricity；Two is electric network state skiIn the whole network
Generated output P can be raisedupMore than the startup power of unit b；In the corresponding new electric network state for generating, machine in electromotor record G
The starting state variable S of group bGbCorresponding change, the accumulative of unit b restarts period TGbSet to 0, remaining power grid state information and skiPhase
Together；
(A3) electric network state s is madekiIn a certain subsidiary engine restart the NBS units c that finishes and enter " rise before steady combustion and exert oneself " state；Corresponding
In the new electric network state for generating, the starting state variable S of unit c in electromotor record GGcCorresponding change, remaining electric network state letter
Breath and skiIt is identical；
(A4) selected section load enters the state that restores electricity；In the corresponding new electric network state for generating, each node is
The load power for restoring electricity is increased by self-defining allocation strategy, remaining power grid state information and skiIt is identical；
(A5) as electric network state skiIn deposit circuit in charging, restart in unit or restoring electricity load when,
Default T is used according to issued Status Change instructionstepDuration proceeds follow-up quantity of state adjustment, generates new electrical network shape
State, wherein, TstepRepresent user-defined power system restoration operating time step-length.
3. power system blackstart scheme generation method as claimed in claim 2, it is characterised in that the recovery operation (A4)
In, allocation strategy specifically includes following steps：
(B1) electric network state s is calculatedkiIn the whole network can raise generated output Pup, calculated load access limit value PLP=αLPup, its
In, αL≤ 1 is the default proportionality coefficient of algorithm；
(B2) according to electric network state skiLoad flow calculation is carried out, the whole network load bus is arranged from big to small by node voltage perunit value
Sequence；
(B3) the front N after sequence is selectedLIndividual load bus is used as both candidate nodes；NLValue principle be：If the whole network is still unrecovered
Load total amount is less than or equal to PLP, then NL=nN, nNFor the sum of electrical network interior joint；If the still unrecovered load total amount of the whole network is big
In PLP, then NLValue should cause front NLThe still unrecovered load total amount of -1 node is less than or equal to PLP, front NLIndividual node is still
Unrecovered load total amount is more than PLP；
(B4) loading that each both candidate nodes are specifically recovered is determined.
4. power system blackstart scheme generation method as claimed in claim 3, it is characterised in that set o-th both candidate nodes
The burden with power increment for increasing successively is (Do1,Do2,…DoMo), amount to MoThe loading of individual grade, the step (B4) is further
Including following sub-step：
(C1) initialization index o=1, continues cycling through flag Fgoon=0；
(C2) if o-th both candidate nodes current loads restoration grade uo=Mo, then show that its all load has been enter into shape to be restored
State, skips to step (C4)；Otherwise execution step (C3)；
(C3) the individual nodes of I (o) in o-th both candidate nodes correspondence electrical network, RNI(x)Restore electricity for x-th both candidate nodes
Load power,For the u of o-th both candidate nodesoThe load increment of+1 stage load amount first order load amount earlier above, ifLoad power R that then o-th both candidate nodes are being restored electricityNI(o)Increase one-level to bear
Lotus amount so thatJuxtaposition uo=uo+ 1, Fgoon=1, then execution step (C4)；It is otherwise direct
Execution step (C4)；
(C4) if o+1≤NL, then return to step after o=o+1 (C2), otherwise execution step (C5) are made；
(C5) if FgoonFor 1, then it is 1, F to reset ogoonFor return to step (C2) after 0, otherwise terminate circulation.
5. the power system blackstart scheme generation method as any one of claim 2 to 4, it is characterised in that described
Recovery operation (A5) further includes following sub-step：
(D1) the new electric network state s* for generating is initialized as and electric network state skiIdentical state；
(D2) all accumulative charging intervals for being in " in charging " status line increase a time step in new electric network state s*
TstepIf the accumulative charging interval of arbitrary circuit is more than or equal to its default charging interval, its state is changed " to have charged
Finish " state, add up charging interval clearing；
All accumulative reboot times for being in " during subsidiary engine is restarted " state unit increase a time step in new electric network state s*
Tstep, its unit output is unit starting power PstartBy step-length PstepNegative value after discretization；If the accumulative weight of arbitrary unit
The time is opened more than or equal to its default starting time, its state is changed for " subsidiary engine restart finish " state, adds up reboot time
Reset；
It is all in new electric network state s* to be in the exerting oneself then by climbing rate and time before steady combustion of " rise before steady combustion and exert oneself " state unit
Step-length TstepProduct increase, if unit output is exerted oneself P more than or equal to its minimum technology after calculatingGminWith startup power Pstart
Difference, then change its state for " unit restart finish " state, unit output is PGmin-PstartBy step-length PstepAfter discretization
Value；
In the state recording L of the load of new electric network state s*, load power P for having restored electricity of q-th node is resetNq=
PNq+RNq, and by load power R for restoring electricity of q-th nodeNqReset, wherein q=1,2 ..., nN, nNFor in electrical network
The sum of node.
6. power system blackstart scheme generation method as claimed in claim 1, it is characterised in that in the step (7), shape
State is cut down and cuts down three steps including non-dominant sequence reduction, hierarchical classification and single goal sequence successively；
Wherein, the execution flow process of hierarchical classification is as follows：
(E1) init state collection Sk+1In classification layer residing for all electric network states be the 1st layer；
(E2) classification layer v=1 is put, by default similar discriminant function f1() is to state set Sk+1In all electric network states carry out
One subseries；
(E3) if always classification number is more than classification number upper limit Cmax, then execution step (E4), otherwise execution step (E5)；
(E4) default distance function d is adoptedvIt is total with the distance between other state class that () calculates each state class of v layers
With reservation is apart from the larger C of summationmaxIndividual state class, deletes other state class, and the electricity that deleted state class is included
Net state is from state set Sk+1In remove, terminate hierarchical classification flow process；
(E5) electric network state quantity is most in selected all state class one, makes v be equal to the residing classification layer of the state class, if v<
3, then v=v+1 is made, the classification layer put in the state class residing for all electric network states is v layers, and by default similar differentiation letter
Number fv() reclassifies to this batch of electric network state, return to step (E3)；Otherwise terminate hierarchical classification flow process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410377330.2A CN104123683B (en) | 2014-08-01 | 2014-08-01 | Electrical power system black-start scheme generation method based on dynamic programming |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410377330.2A CN104123683B (en) | 2014-08-01 | 2014-08-01 | Electrical power system black-start scheme generation method based on dynamic programming |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104123683A CN104123683A (en) | 2014-10-29 |
| CN104123683B true CN104123683B (en) | 2017-04-12 |
Family
ID=51769083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410377330.2A Expired - Fee Related CN104123683B (en) | 2014-08-01 | 2014-08-01 | Electrical power system black-start scheme generation method based on dynamic programming |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104123683B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106357245B (en) * | 2015-07-17 | 2019-06-25 | 陕西千山航空电子有限责任公司 | A kind of frequency turns potential circuit and method |
| CN106643769A (en) * | 2016-12-19 | 2017-05-10 | 中国科学院微电子研究所 | Method and system for determining communication path for formation driving and formation driving method |
| CN107179688B (en) * | 2017-06-21 | 2020-06-26 | 华北电力大学 | Power system reliability analysis method considering Monte Carlo state sampling truncation |
| CN109217318B (en) * | 2018-10-30 | 2022-06-24 | 国网江苏省电力有限公司南通供电分公司 | Energy router-based off-grid island multi-microgrid coordinated operation control method |
| CN109490622B (en) * | 2018-11-10 | 2024-06-14 | 国网辽宁省电力有限公司电力科学研究院 | Electric quantity data verification technology based on tri-state information |
| CN110429606B (en) * | 2019-07-26 | 2023-04-18 | 昆明理工大学 | Method for verifying transient stability of black start of power grid |
| CN111614084B (en) * | 2020-06-09 | 2021-09-17 | 东北电力大学 | Multi-energy-storage-support black-start coordination control method based on dynamic power distribution |
| CN112506028B (en) * | 2020-11-19 | 2022-11-22 | 西安热工研究院有限公司 | Black start test node time recording device and method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102904256A (en) * | 2012-10-23 | 2013-01-30 | 广东电网公司电力调度控制中心 | Method and system for rapid self-healing of power grid |
-
2014
- 2014-08-01 CN CN201410377330.2A patent/CN104123683B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102904256A (en) * | 2012-10-23 | 2013-01-30 | 广东电网公司电力调度控制中心 | Method and system for rapid self-healing of power grid |
Non-Patent Citations (3)
| Title |
|---|
| "基于马尔可夫状态空间法的超高压电网继电保护系统可靠性分析模型";张雪松 等;《电网技术》;20080731;第32卷(第13期);94-99 * |
| "电力系统恢复相关问题及其三层Petri网建模";刘栋 等;《高电压技术》;20060731;第32卷(第7期);109-112 * |
| "考虑机组启动时限的大停电后初期恢复路径优化";韩忠晖 等;《中国电机工程学报》;20090205;第29卷(第4期);21-26 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104123683A (en) | 2014-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104123683B (en) | Electrical power system black-start scheme generation method based on dynamic programming | |
| CN107862405B (en) | Power system grid reconstruction optimization method taking microgrid as black-start power supply | |
| CN107317334B (en) | A kind of electric system rack reconstructing method and device | |
| CN112994022A (en) | Source-storage-load distributed cooperative voltage control method and system thereof | |
| CN103792924B (en) | A kind of Power System Expansion black-start method containing microgrid | |
| CN111062632B (en) | 5G energy Internet virtual power plant economic dispatching method based on edge intelligence | |
| CN106992519B (en) | A kind of network load recovery robust Optimal methods based on information gap decision theory | |
| Choobineh et al. | An automation scheme for emergency operation of a multi-microgrid industrial park | |
| CN103401241B (en) | A kind of micro-capacitance sensor service restoration method based on distributed power source ad-hoc network strategy | |
| CN108777493B (en) | Sensitivity matrix-based low-voltage microgrid secondary voltage control method | |
| CN113872198B (en) | Active power distribution network fault recovery method based on reinforcement learning method | |
| CN105337278A (en) | Method for network reconstruction double-layer optimization based on node importance evaluation matrix | |
| CN112769149B (en) | Hybrid wind-solar micro-grid rapid frequency response distributed coordination control method and system | |
| Liao et al. | An improved two-stage optimization for network and load recovery during power system restoration | |
| CN203839944U (en) | Control system for smooth switching of microgrid operation modes based on DG ad-hoc network strategy | |
| CN109066751A (en) | Power storage and the control device of distribution, method and power supply system | |
| CN106655253A (en) | Single-/three-phase multi-micro-grid region dynamic partitioning method | |
| CN110797918B (en) | Source network load system load recovery method and system based on closed-loop control | |
| CN109193725B (en) | Method for recovering load based on micro-grid | |
| CN113078633B (en) | Method for improving restoring force of power transmission and distribution coupling system containing renewable energy | |
| CN105896579B (en) | Consider the charging pile planing method that distribution voltage dip causes electric car tripping to cause overvoltage | |
| CN112688366A (en) | Optimal recovery path based on distributed power supply and black start strategy | |
| CN116418012A (en) | Micro-grid transient stability emergency control method based on parameter rolling regulation and control | |
| CN111106622B (en) | Active power distribution network power supply recovery method based on RMPC | |
| CN115021241A (en) | Self-recovery method of power distribution network based on distributed resources |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170412 Termination date: 20210801 |