CN108847684A - A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource - Google Patents
A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource Download PDFInfo
- Publication number
- CN108847684A CN108847684A CN201810691882.9A CN201810691882A CN108847684A CN 108847684 A CN108847684 A CN 108847684A CN 201810691882 A CN201810691882 A CN 201810691882A CN 108847684 A CN108847684 A CN 108847684A
- Authority
- CN
- China
- Prior art keywords
- load
- branch
- recoverable
- starting
- fault
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000011084 recovery Methods 0.000 claims abstract description 89
- DEPMSUUWSGUYKQ-IWXIMVSXSA-N (2r,3s,4r,5r)-2,3,4,5-tetrahydroxy-6-[(4-nitro-2,1,3-benzoxadiazol-7-yl)amino]hexanal Chemical compound O=C[C@H](O)[C@@H](O)[C@H](O)[C@H](O)CNC1=CC=C([N+]([O-])=O)C2=NON=C12 DEPMSUUWSGUYKQ-IWXIMVSXSA-N 0.000 claims description 33
- 230000008569 process Effects 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000005457 optimization Methods 0.000 claims description 21
- 230000000694 effects Effects 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000002596 correlated effect Effects 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 abstract description 3
- 230000004888 barrier function Effects 0.000 abstract 1
- 241001672018 Cercomela melanura Species 0.000 description 12
- 238000010248 power generation Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
Classifications
-
- H02J3/382—
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/001—Methods to deal with contingencies, e.g. abnormalities, faults or failures
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses the active distribution network black starting-up fault recovery methods based on distributed generation resource.This method determines starting path and the fail-over path of distributed generation resource according to the starting characteristic of the distribution situation of Distributed Generation in Distribution System and distributed generation resource mainly for the large area or completely black failure in active distribution network.In black starting-up recovery process, suitable boot sequence is selected based on shortest path first therefore starts the distributed generation resource in barrier region, branch rate of load condensate minimum is influenced with failure recovery time and fault recovery again, load may be restored and to the objective function that is up in non-faulting region, gradually restores the load in power distribution network.The present invention gives detailed algorithm description using 53 meshed networks as test macro, and by a series of experiments have shown that mentioned method with the distributed generation resource in effectively start fault zone and can effectively increase the resume speed of fault zone internal loading.
Description
Technical Field
The invention relates to the field of distributed power supply and fault recovery, in particular to a black start fault recovery method for an active power distribution network based on distributed power supplies.
Background
With the increasing maturity of power generation technologies of renewable energy sources (wind energy, solar energy and the like), distributed power generation based on the renewable energy sources is more and more accepted by people. The distributed power generation technology can fully utilize the power generation form of renewable energy, and the distributed power generation technology is more and more widely applied to a power system by virtue of the advantages of low investment cost, flexible power generation mode, high power supply reliability, environmental friendliness and the like.
The black start method under the all black fault is firstly applied to the power grid in south Italy in 1982, and in 2000 in China, the black start is firstly tested by using a thirteen-tomb hydropower station as a black start power supply in the power grid in North China. Along with the permeability of the DGs in the power distribution network is continuously improved, the DGs serve as backup power sources of the main network to make remarkable contribution to improving the power supply stability of the power distribution network. Meanwhile, the problem of fault recovery when the DG is applied to the large-area or all-black fault of the active power distribution network system also draws wide attention of domestic and foreign scholars. In the black start process, the DG with self-starting capability in the system or the external power supply of the system can be used for starting and recovering the normal supply of the whole power grid. However, the method for restoring the power grid through the external power supply of the system has a slow restoration speed and may cause a phenomenon that a large-area load loses power for a long time, but the fast self-starting characteristic based on the DG in the fault area can supply power to the load around the load in a short time, so that the power supply reliability and high efficiency in the case of total blackness or large-area fault can be improved. Therefore, in the research of the active power distribution network full black/large-area fault, the black start method based on the DG power supply can more effectively utilize the flexibility of the DG, and has important practical significance for improving the safety stability, the economy and the environmental friendliness of the power grid.
Based on the recognition, the invention establishes a fault recovery method for the distributed power supply as a black start power supply of the power distribution network. After a fault occurs, the self-starting distributed power supplies in the fault area are used for starting the rest non-self-starting distributed power supplies, the distributed power supplies started in the fault area are used for recovering loads as much as possible, particularly first-stage loads, in a time considered as short as possible, and the influence on other lines in a distribution network in the recovery process is ensured to be minimum, namely, the optimal compromise between the rapidity and the safety of fault recovery is finally realized.
Disclosure of Invention
Aiming at the defects of the existing fault recovery strategy, the invention aims to provide an intelligent fault recovery method for a power distribution network with a distributed power supply.
The invention is realized by the following technical means, and the specific implementation steps are as follows: a method for recovering intelligent faults of a power distribution network with distributed power supplies comprises the following steps:
a method for recovering intelligent faults of a power distribution network with distributed power supplies comprises the following steps:
step (1): detecting every time delta t, and detecting whether all nodes in the power distribution network system have faults at the current moment;
step (2): when a fault occurs at the time T, determining the output of the distributed power supply at the time, disconnecting the switch of the branch circuit affected by the fault and calculating the power flow distribution at the time, determining the state of the distributed power supply in a fault area and starting a self-starting distributed power supply (BDG) in a system;
and (3): according to the power flow distribution at the fault moment and the distribution situation of the self-starting distributed power supply (BDG), selecting a proper starting path for the non-self-starting distributed power supply (NBDG) which is not started in the system, and gradually starting the non-self-starting distributed power supply (NBDG).
And (4): according to the starting condition of the distributed power supply in the system, loads in the power distribution network are gradually recovered by the aid of the target function with the smallest fault recovery time and the smallest fault recovery influence branch load rate and the largest possible recovery load, and by the aid of the safety and stability constraint conditions of the system.
And (5) updating the output of the distributed power supply at the moment T +1, and repeating the step (3) and the step (4) until the fault in the system is cleared.
Further, the step 3 specifically includes:
(A1) obtaining the shortest starting path between all BDGs and NBDGs in the system through a dijkstra shortest path algorithm; and calculating each element value of the startup path weight matrix Pri, wherein the row number of the matrix is the number of BDGs in the system, and the column number of the matrix is the number of NBDGs in the system. The calculation method of each element value of the startup path weight matrix Pri is as follows:
judging whether the residual power of the ith BDG is more than 1.1 times of the load on all nodes where the shortest path from the ith BDG to the jth NBDG passes after the power required by starting the jth NBDG is started;
if the shortest path is larger than the shortest path, the weight of starting the jth NBDG by the ith BDG can be calculated by the equations (1) and (2).
Prii,j＝Li,j/Ti,j(1)
Wherein L isi,jThe weight of the load passing through all the nodes for the shortest starting path from the ith BDG to the jth NBDG; t isi,jIndicating the number of switches T required to operate the start-up pathi,j；ρkThe weight of the load of the kth node passing through the starting path is shown, and if the kth node is a first-level load, the weight is rhokOtherwise, the weight is ρk＝1；lkThe load size of the kth node passing through the starting path is shown; n is a radical ofloads(i,j)The number of nodes passed on the shortest path by which the ith BDG starts the jth NBDG is shown.
If the weight of the starting path is less than the load weight L, the weight of the starting path is the load weight Li,jNegative infinity (-inf), so Prii,jAlso (-inf);
(A2) pri obtained based on step A1i,jAnd solving an optimal starting scheme of the NBDG through a mixed integer programming model, and determining a node to be recovered and a switch state to be operated.
(A3) Calculating and checking whether the power flow of the optimal starting scheme meets power flow operation constraint and line safety constraint; for the
All branches have line safety constraints of:
wherein V is the lower limit of the line voltage;is the upper limit of the line voltage; v is the line voltage;is the upper limit of the line current; i is line current;
the power flow operation constraint is as follows:
Uv,nthe voltage value of the v-th node in the nth sub-optimization process is shown; gamma raybran_vw,nIt is indicated whether the v-th node and the w-th node are topologically in phase or not in the nth sub-optimization processIf adjacent to γbran_vw,nIf not adjacent to γ ═ 1bran_vw,n＝0；Gvw、BvwRespectively representing the conductance and susceptance values between the v-th node and the w-th node of the branch, and if the v-th node and the w-th node are not adjacent, Gvw＝Bvw＝inf；δvw.nThe angle difference between the v nodes and the w node in the nth sub-optimization process is shown;andrespectively representing the active load on the vth node and the active power of the DG;andrespectively representing the reactive load on the vth node and the reactive power magnitude of the DG.
(A4) The NBDG of a3 that meets the power system constraints is started.
(A5) The initiated NBDG is considered to be the BDG and the steps (A1) - (A4) are repeated until all NBDGs are initiated or there is no NBDG that can be initiated.
Further, the step 4 specifically includes:
(B1) calculating the load recovery capacity of all recoverable branches in the current state, wherein the load recovery capacity comprises the size of a direct recovery load and the position relation of each recoverable branch to the residual load to be recovered; the recoverable branch does not include a system association switch, and specifically comprises:
(B1.1) assume that N is common in the current stateswiThe branch can be recovered, if one branch is recovered, a part of load can be recovered and supplied, and the part of load is the direct recovery load of the branch; expression for direct recovery load of i-th recoverable branchEquation (3) calculates:
wherein, n _ load is the number of direct recovery loads of the ith recoverable branch; rhoload,jFor the weight of the jth direct recovery load, if it is the first-order load ρload,j100; otherwise, ρload,j＝1；PjThe magnitude of the direct recovery load for the jth;
(B1.2) Indirect restoration load P of i-th recoverable Branchindirect loads(i) Calculated using expression (4):
wherein m is the number of the residual loads to be recovered after the ith recoverable branch is selected to be recovered; pindirect loads,ijAfter the ith recoverable branch is recovered, the size of the load to be recovered; distanceilRecovering the number of the branch circuits to be recovered of the ith load to be recovered after recovering the ith recoverable branch circuit; rholAnd the weight of the load on the branch to be recovered for the ith load to be recovered.
(B1.3) comprehensively considering the direct and indirect recovery load, and calculating the load recovery capacity of the ith recoverable branch by using an expression (5):
f1,n(i)＝Pdirect loads(i)+Pindirect loads(i) (5)
(B2) calculating the operation time cost of all recoverable branches in the current state; dividing all nodes in the system into different areas according to buses connected with the nodes, wherein if the switches on the branches are related switches, the time cost is 0.4; if the switch on the branch is not the correlated switch, judging the time cost according to whether the nodes connected with the two ends of the branch are in the same region, and if the nodes are in the same region, the time cost of the breaker is 0.1; if the circuit breakers are not in the same area, the time cost of the circuit breaker is 0.2;
(B3) calculating the influence of all recoverable branches on the load flow in the non-fault area under the current state, namely calculating the calculated value of load flow change of other non-fault branches possibly caused by recovering a certain branch, wherein the risk coefficient of the ith recoverable branch is calculated by an expression (6):
wherein N ison_swiAnd Non_busRespectively representing the number of switches and nodes in the non-fault area; xi,p,nIndicating whether the current of the p branch in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Xi,p,nIf X is absent, 1i,p,n＝0；Yi,q,nIndicating whether the voltage of the q-th node in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Yi,q,nIf X is absent, 1i,p,n＝0,Yi,q,n0; whileShowing the current of the p branch in the non-fault area and the upper limit value (I) of the capacity of the p branch if the ith switch is closed in the nth optimization processupperlimit,p) Percent difference of (d);the lower limit value (U) of the voltage and the voltage capacity of the p-th node in the non-fault area in the nth optimization process is shownlowerlimit,p) Percent difference of (c).
(B3) Taking the sum of the load recovery capacity of each recoverable branch, the action time of the circuit breaker and the normalized penalty value of the load flow influence on the non-fault area as a recovery cost matrix element of the recoverable branch, and selecting a branch with the minimum cost from the matrix every time; wherein, the recovery cost of the recoverable branch is calculated by an expression (7):
Xnis a row vector with the total number of switches in the fault area as the dimension. f. ofk,nAnd fk,n *Respectively before and after the corresponding k-th sub-function is not normalized, NswiThe number of switches that can be operated in the fault area; wherein f is1,i *The normalized value is the action time of the circuit breaker; f. of2,i *The normalized value is the branch load recovery capacity; f. of3,i *As an effect on non-failure zones; and φ is a penalty function for x ∈ [0,1 ]]Is provided with
(B4) Calculating and checking the line safety constraint and the distribution network topological structure constraint of the system after the branch is recovered; if the line safety constraint condition is not met, excluding the branch which does not meet the safety constraint currently from the recoverable branches, and repeating the steps (3) and (4) until the smallest recoverable branch which meets the constraint condition is found; and recovering the minimum recoverable branch meeting the constraint condition, and updating the topological structure of the system. Wherein, to all branches, wherein, the line safety restraint and the distribution network topological structure restraint are:
path(v,w)≤1v∈[1,Nbus],w∈[1,Nbus]
wherein V is the lower limit of the line voltage;is the upper limit of the line voltage; v is the line voltage;is the upper limit of the line current; i is line current; the function path () represents the number of paths between two nodes of the argument. N is a radical ofbusAnd NsubThe number of all nodes in the system and the number of bus nodes are shown separately.
(B5) Repeating steps (B1) to (B4) until all recoverable loads are restored.
The invention has the beneficial effects that: according to the distribution condition of the distributed power supply with the fault influence range, the importance of the primary load and the influence on the safety and the stability of the system in the recovery process are fully considered, and the black start fault recovery method is realized. According to the fault recovery method, firstly, the non-self-starting distributed power supplies are respectively started according to the capacity of the self-starting distributed power supplies in the fault range, and then the starting method based on the distributed power supplies is realized according to the capacity of the started distributed power supplies in the fault range and the capacity of the feeder line in the non-fault range.
Drawings
FIG. 1 is a process flow diagram of the present invention;
FIG. 2 is a schematic view of the topology of the test system (53-node network) of the present method;
FIG. 3 is a graph of active and reactive power at each node in the test system;
fig. 4 shows a distributed power supply startup situation and a fault recovery situation in two distributed power supply output scenarios in embodiment 1 of the present invention;
fig. 5 shows a distributed power supply startup situation and a fault recovery situation in the case of different distributed power supply capacities in embodiment 1 of the present invention.
Detailed description of the preferred embodiments
The invention is described in further detail below with reference to the accompanying drawings:
fig. 1 shows a process flow diagram of the invention. Specific implementations thereof will be described below with reference to specific examples. The specific steps of the 53-node test network are described, and the topology structure is shown in fig. 2.
The load data curve of each node in the system is shown in fig. 3. Wherein, the nodes 8, 9, 12, 18, 22, 30, 33, 39 and 40 are first-level load nodes; and there is a distributed power source on nodes 20, 9, 4, 22, 24, 8, 33, 21, 30, 45, 39, 14, 40, 49 and 28. And the distributed power sources on nodes 20, 33, 30, 40 and 49 are non-self-startable distributed power sources, and the rest are self-startable distributed power sources.
Step (1): detecting every time delta t, and detecting whether all nodes in the power distribution network system have faults at the current moment;
step (2): when a fault occurs at the time T, determining the output of the distributed power supply at the time, disconnecting the switch of the branch circuit affected by the fault and calculating the power flow distribution at the time, determining the state of the distributed power supply in a fault area and starting a self-starting distributed power supply (BDG) in a system;
and (3): according to the power flow distribution at the fault moment and the distribution situation of the self-starting distributed power supply (BDG), selecting a proper starting path for the non-self-starting distributed power supply (NBDG) which is not started in the system, and gradually starting the non-self-starting distributed power supply (NBDG). The method specifically comprises the following steps:
(A1) obtaining the shortest starting path between all BDGs and NBDGs in the system through a dijkstra shortest path algorithm; and calculating each element value of the startup path weight matrix Pri, wherein the row number of the matrix is the number of BDGs in the system, and the column number of the matrix is the number of NBDGs in the system. The calculation method of each element value of the startup path weight matrix Pri is as follows:
judging whether the residual power of the ith BDG is more than 1.1 times of the load on all nodes where the shortest path from the ith BDG to the jth NBDG passes after the power required by starting the jth NBDG is started;
if the shortest path is larger than the shortest path, the weight of starting the jth NBDG by the ith BDG can be calculated by the equations (1) and (2).
Prii,j＝Li,j/Ti,j(1)
Wherein L isi,jThe weight of the load passing through all the nodes for the shortest starting path from the ith BDG to the jth NBDG; t isi,jIndicating the number of switches T required to operate the start-up pathi,j；ρkThe weight of the load of the kth node passing through the starting path is shown, and if the kth node is a first-level load, the weight is rhokOtherwise, the weight is ρk＝1；lkThe load size of the kth node passing through the starting path is shown; n is a radical ofloads(i,j)The number of nodes passed on the shortest path by which the ith BDG starts the jth NBDG is shown.
If the weight of the starting path is less than the load weight L, the weight of the starting path is the load weight Li,jNegative infinity (-inf), so Prii,jIs also (-inf))；
(A2) Pri obtained based on step A1i,jAnd solving an optimal starting scheme of the NBDG through a mixed integer programming model, and determining a node to be recovered and a switch state to be operated.
(A3) Calculating and checking whether the power flow of the optimal starting scheme meets power flow operation constraint and line safety constraint; for the
All branches have line safety constraints of:
wherein,Vis the lower limit of the line voltage;is the upper limit of the line voltage; v is the line voltage;is the upper limit of the line current; i is line current;
the power flow operation constraint is as follows:
Uv,nthe voltage value of the v-th node in the nth sub-optimization process is shown; gamma raybran_vw,nIndicating whether the nth node and the w-th node are adjacent in topology structure in the nth sub-optimization process, if the adjacent gamma isbran_vw,nIf not adjacent to γ ═ 1bran_vw,n＝0；Gvw、BvwRespectively representing the conductance and susceptance values between the v-th node and the w-th node of the branch, and if the v-th node and the w-th node are not adjacent, Gvw＝Bvw＝inf；δvw.nThe angle difference between the v nodes and the w node in the nth sub-optimization process is shown;andrespectively representing the active load on the vth node and the active power of the DG;andrespectively representing the reactive load on the vth node and the reactive power magnitude of the DG.
(A4) The NBDG of a3 that meets the power system constraints is started.
(A5) The initiated NBDG is considered to be the BDG and the steps (A1) - (A4) are repeated until all NBDGs are initiated or there is no NBDG that can be initiated.
(4) According to the starting condition of the distributed power supply in the system, loads in the power distribution network are gradually recovered by the aid of the target function with the smallest fault recovery time and the smallest fault recovery influence branch load rate and the largest possible recovery load, and by the aid of the safety and stability constraint conditions of the system. The method specifically comprises the following steps:
(B1) calculating the load recovery capacity of all recoverable branches in the current state, wherein the load recovery capacity comprises the size of a direct recovery load and the position relation of each recoverable branch to the residual load to be recovered; the recoverable branch does not include a system association switch, and specifically comprises:
(B1.1) assume that N is common in the current stateswiThe branch of the strip canRecovering, if one of the branches is recovered, a part of the load can be recovered and supplied, and the part of the load is the direct recovery load of the branch; the direct recovery load of the i-th recoverable branch is calculated using expression (3):
wherein, n _ load is the number of direct recovery loads of the ith recoverable branch; rhoload,jFor the weight of the jth direct recovery load, if it is the first-order load ρload,j100; otherwise, ρload,j＝1；PjThe magnitude of the direct recovery load for the jth;
(B1.2) Indirect restoration load P of i-th recoverable Branchindirect loads(i) Calculated using expression (4):
wherein m is the number of the residual loads to be recovered after the ith recoverable branch is selected to be recovered; pindirect loads,ijAfter the ith recoverable branch is recovered, the size of the load to be recovered; distanceilRecovering the number of the branch circuits to be recovered of the ith load to be recovered after recovering the ith recoverable branch circuit; rholAnd the weight of the load on the branch to be recovered for the ith load to be recovered.
(B1.3) comprehensively considering the direct and indirect recovery load, and calculating the load recovery capacity of the ith recoverable branch by using an expression (5):
f1,n(i)＝Pdirect loads(i)+Pindirect loads(i) (5)
(B2) calculating the operation time cost of all recoverable branches in the current state; dividing all nodes in the system into different areas according to buses connected with the nodes, wherein if the switches on the branches are related switches, the time cost is 0.4; if the switch on the branch is not the correlated switch, judging the time cost according to whether the nodes connected with the two ends of the branch are in the same region, and if the nodes are in the same region, the time cost of the breaker is 0.1; if the circuit breakers are not in the same area, the time cost of the circuit breaker is 0.2;
(B3) calculating the influence of all recoverable branches on the load flow in the non-fault area under the current state, namely calculating the calculated value of load flow change of other non-fault branches possibly caused by recovering a certain branch, wherein the risk coefficient of the ith recoverable branch is calculated by an expression (6):
wherein N ison_swiAnd Non_busRespectively representing the number of switches and nodes in the non-fault area; xi,p,nIndicating whether the current of the p branch in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Xi,p,nIf X is absent, 1i,p,n＝0；Yi,q,nIndicating whether the voltage of the q-th node in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Yi,q,nIf X is absent, 1i,p,n＝0,Yi,q,n0; whileShowing the current of the p branch in the non-fault area and the upper limit value (I) of the capacity of the p branch if the ith switch is closed in the nth optimization processupperlimit,p) Percent difference of (d);the lower limit value (U) of the voltage and the voltage capacity of the p-th node in the non-fault area in the nth optimization process is shownlowerlimit,p) Percent difference of (c).
(B3) Taking the sum of the load recovery capacity of each recoverable branch, the action time of the circuit breaker and the normalized penalty value of the load flow influence on the non-fault area as a recovery cost matrix element of the recoverable branch, and selecting a branch with the minimum cost from the matrix every time; wherein, the recovery cost of the recoverable branch is calculated by an expression (7):
Xnis a row vector with the total number of switches in the fault area as the dimension. f. ofk,nAnd fk,n *Respectively before and after the corresponding k-th sub-function is not normalized, NswiThe number of switches that can be operated in the fault area; wherein f is1,i *The normalized value is the action time of the circuit breaker; f. of2,i *The normalized value is the branch load recovery capacity; f. of3,i *As an effect on non-failure zones; and φ is a penalty function for x ∈ [0,1 ]]Is provided with
(B4) Calculating and checking the line safety constraint and the distribution network topological structure constraint of the system after the branch is recovered; if the line safety constraint condition is not met, excluding the branch which does not meet the safety constraint currently from the recoverable branches, and repeating the steps (3) and (4) until the smallest recoverable branch which meets the constraint condition is found; and recovering the minimum recoverable branch meeting the constraint condition, and updating the topological structure of the system. Wherein, to all branches, wherein, the line safety restraint and the distribution network topological structure restraint are:
path(v,w)≤1v∈[1,Nbus],w∈[1,Nbus]
wherein,Vis the lower limit of the line voltage;is the upper limit of the line voltage; v is the line voltage;is the upper limit of the line current; i is line current; the function path () represents the number of paths between two nodes of the argument. N is a radical ofbusAnd NsubThe number of all nodes in the system and the number of bus nodes are shown separately.
(B5) Repeating steps (B1) to (B4) until all recoverable loads are restored.
(5) And (5) updating the output of the distributed power supply at the moment T +1, and repeating the steps (3) to (6) until the fault in the system is cleared.
In the following embodiments, under consideration of a black start method based on a distributed power source, the start effect and the fault recovery effect of the distributed power source under different output scenes of the distributed power source (embodiment 1); considering the black start method based on the distributed power, the start effect and the fault recovery effect of the distributed power under different capacities of the distributed power (embodiment 2), it is appreciated that the advantages and features of the present invention can be easily understood by those skilled in the art, so as to make a clearer and more definite definition on the protection scope of the present invention.
Example 1
In the embodiment, a wind driven generator is taken as an example, and the starting condition, the fault recovery condition and the utilization rate of the distributed power supply are analyzed under different wind power output scenes, wherein in a 53-node active power distribution network, the system topology structure and the load of each node are as shown in fig. 2 and fig. 3, and under the assumption that a large-area fault lasting for 2 hours occurs.
System faults respectively occur in a high/low wind power output scene, the fault recovery condition in each time interval in the system is analyzed on the assumption that the output of a wind driven generator changes once every 30 minutes, and by taking the simultaneous occurrence of faults of nodes 1, 3, 11 and 14 as an example, the specific process is as follows:
(1) detecting every time delta t, and detecting whether all nodes in the power distribution network system have faults at the current moment;
(2) when the fault occurs at the time T, according to the topological structure in the power distribution network system, the branch affected by the fault is disconnected, the distributed power supply is prevented from operating in an island state, only the load on the bus 104 is in a normal operation state at the time, and the load flow distribution at the time is calculated.
(3) By the distribution of the renewable distributed power sources in the fault area at this time (nodes 9, 22, 24, 8, 4, 45, 39 and 14), and by selecting appropriate startup paths for the non-self-starting distributed power sources that are not started up in the system (nodes 33, 40, 49 and 28).
(4) According to the starting condition of the distributed power supply in the system and the load flow condition of the non-fault area, firstly, the load recovery capability, the switching operation time and the influence on the non-fault area of all recoverable branches in the current state are calculated.
(5) And calculating a penalty value by substituting the load recovery capacity, the switching operation time and the influence on the non-fault area of each recoverable branch into a penalty function through normalization, further taking the penalty value as a recovery cost matrix element of the recoverable branch, and selecting a branch with the minimum cost from the matrix every time.
(6) Calculating and checking the power flow operation constraint and the line safety constraint of the system after the branch is recovered; if all the constraint conditions are not met, repeating the step (5) until recoverable branches meeting the constraint conditions are found; and (5) if all the constraint conditions are met, updating the topological structure of the system, and repeating the steps (4) and (5) until all the recoverable loads are recovered.
(7) And (5) updating the output power of the distributed power supply at the moment T +1, and repeating the steps (3) to (6) until the fault is cleared.
And (3) repeating the experimental steps (1) - (7) in consideration of different wind power output scenes, and comparing the starting condition and the fault recovery condition of the distributed power supply and the utilization rate of the distributed power supply in two groups of experiments, as shown in fig. 4.
As can be seen from embodiment 1, in the process of random variation of the output of the distributed power supply, the black start fault recovery method can fully utilize the starting capability of the self-startable distributed power supply to quickly start the non-self-startable distributed power supply, and is used for quickly recovering a fault load in the fault recovery process and reducing the influence on a non-fault area. Compared with the distributed power supply under the condition of low output, the distributed power supply can be better started under the condition of high output, and the recovery of the load in a fault area is accelerated.
Example 2
This example analyzes the startup situation, the fault recovery situation, and the utilization rate of the distributed power sources when a large area fault occurs in the case of a capacity change (from 4.5MW to 6.0MW) of a single distributed power source in the same active power distribution network in example 1.
When the capacity of the distributed power supply in the system is 4.5MW, the specific process when the analysis system has a large-area fault is as follows:
(1) detecting at intervals, and detecting whether all nodes in the power distribution network system have faults at the current moment;
(2) when a fault occurs at a moment, according to a topological structure in a power distribution network system, a branch circuit affected by the fault is disconnected, the distributed power supply is prevented from operating in an island state, only the load on the bus 104 is in a normal operation state at the moment, and the load flow distribution at the moment is calculated.
(3) By the distribution of the renewable distributed power sources in the fault area at this time (nodes 9, 22, 24, 8, 4, 45, 39 and 14), and by selecting appropriate startup paths for the non-self-starting distributed power sources that are not started up in the system (nodes 33, 40, 49 and 28).
(4) According to the starting condition of the distributed power supply in the system and the load flow condition of the non-fault area, firstly, the load recovery capability, the switching operation time and the influence on the non-fault area of all recoverable branches in the current state are calculated.
(5) And calculating a penalty value by substituting the load recovery capacity, the switching operation time and the influence on the non-fault area of each recoverable branch into a penalty function through normalization, further taking the penalty value as a recovery cost matrix element of the recoverable branch, and selecting a branch with the minimum cost from the matrix every time.
(6) Calculating and checking the power flow operation constraint and the line safety constraint of the system after the branch is recovered; if all the constraint conditions are not met, repeating the step (5) until recoverable branches meeting the constraint conditions are found; and (5) if all the constraint conditions are met, updating the topological structure of the system, and repeating the steps (4) and (5) until all the recoverable loads are recovered.
(7) And (5) updating the output power of the distributed power supply at the moment, and repeating the steps (3) to (6) until the fault is cleared.
The above experimental steps (1) - (7) were repeated to compare the startup condition, the failure recovery condition, and the utilization ratio of the distributed power supplies in two sets of experiments, considering that the capacity of the distributed power supplies increased from 4.5MW to 6.0MW, as shown in fig. 5.
As can be seen from embodiment 2, in the process of gradually increasing the capacity of the distributed power supply, the black start fault recovery method can fully utilize the starting capability of the self-startable distributed power supply to quickly start the non-self-startable distributed power supply, and is used for quickly recovering the load which is a fault in the fault recovery process and reducing the influence on the non-fault area. When the output of the distributed power supply is insufficient (the capacity of the distributed power supply is 4.5 MW), the distributed power supply in the fault area can not be fully started; when the capacity of the distributed power supply is larger than 5.0MW, the distributed power supply can be started in the first time interval, but due to the constraint of the system topology and the capacity, the recoverable load quantity is not increased along with the increase of the capacity of the distributed power supply, so that the capacity and the position of the distributed power supply are reasonably arranged to achieve the optimal effect of fault recovery.
Claims (5)
1. A method for intelligent fault recovery of a power distribution network with distributed power supplies is characterized by comprising the following steps:
step (1): detecting every time delta t, and detecting whether all nodes in the power distribution network system have faults at the current moment;
step (2): when a fault occurs at the time T, determining the output of the distributed power supply at the time, disconnecting the switch of the branch circuit affected by the fault, calculating the power flow distribution at the time, determining the state of the distributed power supply in a fault area, and starting a self-starting distributed power supply (BDG) in a system;
and (3): according to the power flow distribution at the fault moment and the distribution situation of the self-starting distributed power supply (BDG), selecting a proper starting path for the non-self-starting distributed power supply (NBDG) which is not started in the system, and gradually starting the non-self-starting distributed power supply (NBDG).
And (4): according to the starting condition of the distributed power supply in the system, loads in the power distribution network are gradually recovered by the aid of the target function with the smallest fault recovery time and the smallest fault recovery influence branch load rate and the largest possible recovery load, and by the aid of the safety and stability constraint conditions of the system.
And (5) updating the output of the distributed power supply at the moment T +1, and repeating the step (3) and the step (4) until the fault in the system is cleared.
2. The method according to claim 1, wherein step 3 is specifically:
(A1) obtaining the shortest starting path between all BDGs and NBDGs in the system; and calculating each element value of the startup path weight matrix Pri, wherein the row number of the matrix is the number of BDGs in the system, and the column number of the matrix is the number of NBDGs in the system. The calculation method of each element value of the startup path weight matrix Pri is as follows:
judging whether the residual power of the ith BDG is more than 1.1 times of the load on all nodes where the shortest path from the ith BDG to the jth NBDG passes after the power required by starting the jth NBDG is started;
if the shortest path is larger than the shortest path, the weight of starting the jth NBDG by the ith BDG can be calculated by the equations (1) and (2).
Prii,j＝Li,j/Ti,j(1)
Wherein L isi,jThe weight of the load passing through all the nodes for the shortest starting path from the ith BDG to the jth NBDG; t isi,jIndicating the number of switches T required to operate the start-up pathi,j；ρkThe weight of the load of the kth node passing through the starting path is shown, and if the kth node is a first-level load, the weight is rhokOtherwise, the weight is ρk＝1；lkThe load size of the kth node passing through the starting path is shown; n is a radical ofloads(i,j)The number of nodes passed on the shortest path by which the ith BDG starts the jth NBDG is shown.
If the weight of the starting path is less than the load weight L, the weight of the starting path is the load weight Li,jNegative infinity (-inf), so Prii,jAlso (-inf);
(A2) pri obtained based on step A1i,jAnd solving an optimal starting scheme of the NBDG through a mixed integer programming model, and determining a node to be recovered and a switch state to be operated.
(A3) Calculating and checking whether the power flow of the optimal starting scheme meets power flow operation constraint and line safety constraint; for all branches, the line safety constraints are:
wherein V is a line voltage;is the upper limit of the line current; i is line current;
the power flow operation constraint is as follows:
Uv,nthe voltage value of the v-th node in the nth sub-optimization process is shown; gamma raybran_vw,nIndicating whether the nth node and the w-th node are adjacent in topology structure in the nth sub-optimization process, if the adjacent gamma isbran_vw,nIf not adjacent to γ ═ 1bran_vw,n＝0；Gvw、BvwRespectively representing the conductance and susceptance values between the v-th node and the w-th node of the branch, and if the v-th node and the w-th node are not adjacent, Gvw＝Bvw＝inf；δvw.nThe angle difference between the v nodes and the w node in the nth sub-optimization process is shown;andrespectively representing the active load on the vth node and the active power of the DG;andrespectively representing the reactive load on the vth node and the reactive power magnitude of the DG.
(A4) The NBDG of a3 that meets the power system constraints is started.
(A5) The initiated NBDG is considered to be the BDG and the steps (A1) - (A4) are repeated until all NBDGs are initiated or there is no NBDG that can be initiated.
3. The method according to claim 2, wherein in step a1, the shortest starting path between all BDGs and NBDGs in the system is obtained by dijkstra shortest path algorithm.
4. The method according to claim 2, wherein in step A3,Vis the lower limit of the line voltage;is the upper limit of the line voltage.
5. The method according to claim 1, wherein step 4 is specifically:
(B1) calculating the load recovery capacity of all recoverable branches in the current state, wherein the load recovery capacity comprises the size of a direct recovery load and the position relation of each recoverable branch to the residual load to be recovered; the recoverable branch does not include a system association switch, and specifically comprises:
(B1.1) assume that N is common in the current stateswiThe branch can be recovered, if one branch is recovered, a part of load can be recovered and supplied, and the part of load is the direct recovery load of the branch; the direct recovery load of the i-th recoverable branch is calculated using expression (3):
wherein, n _ load is the number of direct recovery loads of the ith recoverable branch; rhoload,jFor the weight of the jth direct recovery load, if it is the first-order load ρload,j100; otherwise, ρload,j＝1；PjThe magnitude of the direct recovery load for the jth;
(B1.2) Indirect restoration load P of i-th recoverable Branchindirect loads(i) Calculated using expression (4):
wherein m is the number of the residual loads to be recovered after the ith recoverable branch is selected to be recovered; pindirect loads,ijAfter the ith recoverable branch is recovered, the size of the load to be recovered; distanceilRecovering the number of the branch circuits to be recovered of the ith load to be recovered after recovering the ith recoverable branch circuit; rholAnd the weight of the load on the branch to be recovered for the ith load to be recovered.
(B1.3) comprehensively considering the direct and indirect recovery load, and calculating the load recovery capacity of the ith recoverable branch by using an expression (5):
f1,n(i)＝Pdirect loads(i)+Pindirect loads(i) (5)
(B2) calculating the operation time cost of all recoverable branches in the current state; dividing all nodes in the system into different areas according to buses connected with the nodes, wherein if the switches on the branches are related switches, the time cost is 0.4; if the switch on the branch is not the correlated switch, judging the time cost according to whether the nodes connected with the two ends of the branch are in the same region, and if the nodes are in the same region, the time cost of the breaker is 0.1; if the circuit breakers are not in the same area, the time cost of the circuit breaker is 0.2;
(B3) calculating the influence of all recoverable branches on the load flow in the non-fault area under the current state, namely calculating the calculated value of load flow change of other non-fault branches possibly caused by recovering a certain branch, wherein the risk coefficient of the ith recoverable branch is calculated by an expression (6):
wherein N ison_swiAnd Non_busRespectively representing the number of switches and nodes in the non-fault area; xi,p,nIndicating whether the current of the p branch in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Xi,p,nIf X is absent, 1i,p,n＝0；Yi,q,nIndicating whether the voltage of the q-th node in the non-fault area is influenced by closing the ith switch operation in the nth optimization process, and if so, Yi,q,nIf X is absent, 1i,p,n＝0,Yi,q,n0; whileShowing the current of the p branch in the non-fault area and the current of the p branch in the non-fault area after the ith switch is closed in the nth optimization processUpper limit of strip branch capacity (I)upperlimit,p) Percent difference of (d);the lower limit value (U) of the voltage and the voltage capacity of the p-th node in the non-fault area in the nth optimization process is shownlowerlimit,p) Percent difference of (c).
(B4) Taking the sum of the load recovery capacity of each recoverable branch, the action time of the circuit breaker and the normalized penalty value of the load flow influence on the non-fault area as a recovery cost matrix element of the recoverable branch, and selecting a branch with the minimum cost from the matrix every time; wherein, the recovery cost of the recoverable branch is calculated by an expression (7):
Xnis a row vector with the total number of switches in the fault area as the dimension. f. ofk,nAnd fk,n *Respectively before and after the corresponding k-th sub-function is not normalized, NswiThe number of switches that can be operated in the fault area; wherein f is1,i *The normalized value is the action time of the circuit breaker; f. of2,i *The normalized value is the branch load recovery capacity; f. of3,i *As an effect on non-failure zones; and φ is a penalty function for x ∈ [0,1 ]]Comprises the following steps:
(B5) calculating and checking the line safety constraint and the distribution network topological structure constraint of the system after the branch is recovered; if the line safety constraint condition is not met, excluding the branch which does not meet the safety constraint currently from the recoverable branches, and repeating the steps (B3) and (B4) until the smallest recoverable branch which meets the constraint condition is found; and recovering the minimum recoverable branch meeting the constraint condition, and updating the topological structure of the system. Wherein, to all branches, wherein, the line safety restraint and the distribution network topological structure restraint are:
path(v,w)≤1 v∈[1,Nbus],w∈[1,Nbus]
wherein,Vis the lower limit of the line voltage;is the upper limit of the line voltage; v is the line voltage;is the upper limit of the line current; i is line current; the function path () represents the number of paths between two nodes of the argument. N is a radical ofbusAnd NsubThe number of all nodes in the system and the number of bus nodes are shown separately.
(B5) Repeating steps (B1) to (B4) until all recoverable loads are restored.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810691882.9A CN108847684A (en) | 2018-06-28 | 2018-06-28 | A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810691882.9A CN108847684A (en) | 2018-06-28 | 2018-06-28 | A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108847684A true CN108847684A (en) | 2018-11-20 |
Family
ID=64201660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810691882.9A Pending CN108847684A (en) | 2018-06-28 | 2018-06-28 | A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108847684A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109754334A (en) * | 2019-01-14 | 2019-05-14 | 华北电力大学（保定） | The packet recovery method and device in power transmission path |
| CN112117761A (en) * | 2020-09-14 | 2020-12-22 | 西安交通大学 | New energy power system recovery path searching method based on improved network weight |
| CN112271727A (en) * | 2020-10-15 | 2021-01-26 | 北京交通大学 | Fault recovery method for flexible power distribution network containing flexible soft switch |
| CN112688366A (en) * | 2020-12-29 | 2021-04-20 | 国网辽宁省电力有限公司电力科学研究院 | Optimal recovery path based on distributed power supply and black start strategy |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105977970A (en) * | 2016-06-20 | 2016-09-28 | 浙江大学 | Distributed power-containing distribution network intelligent fault recovery method |
-
2018
- 2018-06-28 CN CN201810691882.9A patent/CN108847684A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105977970A (en) * | 2016-06-20 | 2016-09-28 | 浙江大学 | Distributed power-containing distribution network intelligent fault recovery method |
Non-Patent Citations (2)
| Title |
|---|
| QIANG YANG: "Robust power supply restoration for self-healing active distribution networks considering the availability of distributed generation", 《ENERGIES》 * |
| 姚骏: "《基于DG的配电网黑启动方案研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109754334A (en) * | 2019-01-14 | 2019-05-14 | 华北电力大学（保定） | The packet recovery method and device in power transmission path |
| CN112117761A (en) * | 2020-09-14 | 2020-12-22 | 西安交通大学 | New energy power system recovery path searching method based on improved network weight |
| CN112117761B (en) * | 2020-09-14 | 2021-12-28 | 西安交通大学 | New energy power system recovery path searching method based on improved network weight |
| CN112271727A (en) * | 2020-10-15 | 2021-01-26 | 北京交通大学 | Fault recovery method for flexible power distribution network containing flexible soft switch |
| CN112688366A (en) * | 2020-12-29 | 2021-04-20 | 国网辽宁省电力有限公司电力科学研究院 | Optimal recovery path based on distributed power supply and black start strategy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108847684A (en) | A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource | |
| CN107862405B (en) | Power system grid reconstruction optimization method taking microgrid as black-start power supply | |
| CN108155663B (en) | Method for controlling a distribution microgrid | |
| CN108988322B (en) | Microgrid operation strategy optimization method considering system time-varying property | |
| CN104753087B (en) | A kind of black-start method of the wind-light storage micro-capacitance sensor containing a variety of energy storage | |
| CN110783946A (en) | Method for locating phase faults in microgrid | |
| Jafari et al. | Study of smart grid protection challenges with high photovoltaic penetration | |
| Mohammadi et al. | Adaptive voltage-based load shedding scheme for the DC microgrid | |
| CN105977970B (en) | A kind of power distribution network intelligent trouble restoration methods containing distributed generation resource | |
| CN110391657B (en) | Method for improving toughness of power system for dealing with short-circuit fault caused by extreme weather | |
| CN114050573A (en) | Fault recovery control method, device, equipment and medium for active power distribution network | |
| Farmer et al. | Power system dynamic performance advancement from history of North American blackouts | |
| Xu et al. | Study on black start strategy for multi-microgrids | |
| CN217388240U (en) | Black start system and power supply system of microgrid | |
| Wang et al. | Early warning of distribution transformer based on bp neural network considering the influence of extreme weather | |
| CN116526559A (en) | Distribution network self-adaptive reclosing method considering DG low voltage ride through characteristics | |
| TW201724685A (en) | Operation scenarios plan based multi-zone microgrid protection system and method | |
| CN111106622B (en) | Active power distribution network power supply recovery method based on RMPC | |
| Alshareef | A new protection scheme for low voltage DC microgrids | |
| Khezri et al. | An Intelligent Fuzzy Control Approach for a Back-Pressure Autonomous Industrial Microgrid | |
| Mohanty et al. | Detection of symmetrical and unsymmetrical fault in a PV connected power system | |
| Alshahir et al. | Improving the reconfiguration of hybrid power networks by combining Genetic Algorithm (GA) with Particle Swarm Optimization (PSO) | |
| CN111355266B (en) | Power distribution network reliability calculation method comprising power distribution automation terminal | |
| CN114865776B (en) | Spare power automatic switching device and method without switching new energy unit and load and application | |
| Ke et al. | A Method of Distribution Network Protection and Application with Penetration of Renewable Generation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181120 |