CN110086153A - A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method - Google Patents
A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method Download PDFInfo
- Publication number
- CN110086153A CN110086153A CN201910301463.4A CN201910301463A CN110086153A CN 110086153 A CN110086153 A CN 110086153A CN 201910301463 A CN201910301463 A CN 201910301463A CN 110086153 A CN110086153 A CN 110086153A
- Authority
- CN
- China
- Prior art keywords
- load
- distribution network
- agent
- isolated island
- power distribution
- 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
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
- H02H7/26—Sectionalised protection of cable or line systems, e.g. for disconnecting a section on which a short-circuit, earth fault, or arc discharge has occured
- H02H7/261—Sectionalised protection of cable or line systems, e.g. for disconnecting a section on which a short-circuit, earth fault, or arc discharge has occured involving signal transmission between at least two stations
- H02H7/262—Sectionalised protection of cable or line systems, e.g. for disconnecting a section on which a short-circuit, earth fault, or arc discharge has occured involving signal transmission between at least two stations involving transmissions of switching or blocking orders
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a kind of, and the active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method.This method each distributed generation resource (Distributed generations) is handled first, it is specified that after failure each DG the method for operation, and determine isolated island division principle, to after failure carry out the division of DG isolated island, formulate isolated operation range.Finally consider topological constraints, node voltage constraint, the constraint of DG units limits, tributary capacity, with turn at least be for load maximum and switch number of operations it is optimal, propose a kind of active power distribution network load transfer method based on MAPSO.By emulation and network analysis show can ground quickly and effectively realize active power distribution network take fault zone load quickly turn to supply, thus reduce failure or maintenance bring loss of outage.
Description
Technical field
The invention belongs to load transfer fields, more particularly to consider that node voltage, DG power and tributary capacity constrain item
Part realizes load transfer to fault zone to lose load minimum and switch operation at least for target.
Background technique
In recent years, with the constant growth of Chinese national economy, demand of all trades and professions to electric power is increased rapidly, but
Since Construction of Intercity Network comparatively lags, lead to the unreasonable than more prominent of electric network composition, this, which is difficult to meet, uses
Requirement of the family to power quality, power supply reliability etc..Especially because the rise of the tertiary industry, the structure of electric load also occurs
Variation, medium and small user includes that residential electricity consumption ratio rises year by year, and causes the route in city electric distribution system increasingly longer, node
It is more and more.
With the construction and development of intelligent distribution network, accessed close to a large amount of DG of load side so that distribution network topological structure becomes
It obtains more complicated.When power distribution network breaks down, needs to cut off faulty equipment as early as possible, then should restore as soon as possible to customer charge
Power supply reduces the range for having a power failure and influencing as far as possible, thus greatly reduce as power failure and caused by social influence and economic loss
Deng, meanwhile, it should also reduce the loss of supply network after failure as far as possible after troubleshooting, improve power supply reliability, Yi Jijun
The performances such as weighing apparatus distribution load.
Power distribution network can cut off route after breaking down by adjusting the folding condition of interconnection switches and block switch
Failure shifts failure zone of influence internal loading, changes the method for operation etc. of power grid, the as far as possible range of reduction failure influence, thus
The economy and safety of operation of power networks are improved, guarantees the power supply quality to customer charge.Load transfer can obviously be dropped because of it
Low failure bring loss, improves system power supply reliability, and becomes one of the important core function in Automation System of Power Network.
Distribution network load turns to realize by network reconfiguration to non-event for referring in the case where meeting power network security service condition
Hinder outage area power failure load quickly turns confession, belongs to complex nonlinear combinatorial optimization problem.Scheme in the prior art does not have
There is solution above-mentioned technical problem.
Summary of the invention
The purpose of the present invention is to propose to a kind of, and the active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns
For method, the program effectively improves the service efficiency of equipment, realizes that distribution network load turns to supply and reduce failure or maintenance is brought
Loss of outage provide decision-making foundation.
To achieve the goals above, technical scheme is as follows:
A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method, which is characterized in that
It the described method comprises the following steps:
When power distribution network containing DG breaks down, load transfer step are as follows:
Step a: after failure occurs, check whether non-faulting power supply interrupted district includes DG, if containing DG, determined according to its type
Whether isolated operation or off-grid operation, turn if isolated operation walk b, otherwise turn walk d；
Step b: according to the DG isolated island division principle pre-established, dividing isolated operation range, remains important negative in isolated island
Lotus power supply；
Step c: the block switch that need to be disconnected when the formation of DG isolated island and outage area branch block switch position are searched, is transferred to
Isolated operation mode；
It establishes the distribution network load containing DG to turn to adjust network structure information for model, propose excellent using intelligent body population
Change the non-faulting outage area restoration path outside algorithm search isolated island,
Power loss load is transferred to normal power supply region to the maximum extent to continue to power, to realize quickly turning for load
For；
Step d: it executes MAPSO load transfer algorithm and entirely solves process；
Step e: after major network load transfer, according to frequency, the amplitude of voltage and the phase of power distribution network, to load transfer model
It encloses interior contained isolated island and carries out simultaneous interconnecting operation, otherwise the method for operation before being restored to failure continues isolated operation.
The present invention is according to the practical operation situation and isolated island division principle rational isolated island of DG different types of after failure
Range of operation makes full use of DG generating capacity to maintain important load normal power supply in isolated island；It is negative that the power distribution network containing DG is established simultaneously
Lotus turns to adjust network structure information for model, proposes using the non-faulting outside Multi-agent Particle Swarm Optimization Algorithm search isolated island
Power loss load is transferred to normal power supply region to the maximum extent and continues to power, to realize load by outage area restoration path
Quickly turn supply.The emulation and analysis method can effectively solve the problems, such as that active power distribution network failure afterload is specialized in, and improve and calculate
Efficiency realizes load transfer, to guarantee the power supply reliability of power distribution network.
As an improvement of the present invention, the step a is specific as follows, when power distribution network breaks down, first to distribution
Net carries out region division, is broadly divided into fault zone, normal power supply region and non-faulting power supply interrupted district.Existed according to distributed generation resource
Physical location in topology determines the DG method of operation after failure.It is broadly divided into direct off-grid, grid-connected or can not isolated operation, simultaneously
Net or three kinds of isolated operation, are indicated with A, B, C type respectively.
As an improvement of the present invention, the isolated island division principle in the step b, specific as follows:
1) meet total power load in isolated island to be no more than under the premise of DG power generation total amount, it is preferential to guarantee important load power supply；
2) guarantee isolated island in device line it is in a safe condition, avoid overload with it is out-of-limit；
3) containing the c-type DG for having island operating capacity in non-faulting outage area, original is divided according to isolated island is pre-established
It is then translated into isolated operation mode, maintains important load power supply in isolated island；If DG containing Type B, it is divided into tool as far as possible
It is run in the isolated island for thering is the c-type DG of frequency modulation and voltage modulation ability to constitute, otherwise follows A type DG off-grid.
As an improvement of the present invention, the distribution network load containing DG is established in the step c to turn to adjust network for model
Structural information is proposed using the non-faulting outage area restoration path outside intelligent body particle swarm optimization algorithm search isolated island, specifically
It is as follows:
For breaking down containing distributed power distribution network, by network reconfiguration to non-faulting power supply interrupted district load transfer,
It during load transfer, needs to maximize load transfer, and requires switch number of operations minimum, extend switch and use year
Limit.
At least turn the objective function supplied as distribution network load to lose load minimum and switch motion number:
Wherein, ω1Indicate loss function of load weight coefficient, ω2To switch number of operations weight coefficient, ω1And ω2Root
According to actual conditions selection and ω1+ω2=1；LBFor fault zone total losses load capacity；γ is non-former for what is obtained after Fault Isolation
Hinder power supply interrupted district set；LiTo turn the capacity for load i；N is total number of switches；XjAnd X'jIt respectively indicates before failure and turns to open after supplying
The state for closing j takes 0 or 1, indicates that switch is in and is opened or closed.
Distribution network load turn supply solution be meet power network security operation and search the smallest solution of target function value to
Amount, so according to the actual situation, solution needs to include following constraint condition:
(1) not network topology constraint (not including DG)；
gk∈G； (2)；
In formula: gkFor the network structure in the region that restored electricity；G is all feasible network structure set.
(2) node voltage constrains；
Node voltage should be maintained in fixed range；
Uimin≤Ui≤Uimax, i=1,2,3 ..., n (3)；
In formula: n is node summary；UiminAnd UimaxRespectively node voltage UiUpper lower limit value；
(3) DG units limits
PDGt,min≤PDGt≤PDGt,max； (4)；
In formula: PDGt,minAnd PDGt,maxIndicate the upper lower limit value of t-th of DG power output；
(4) tributary capacity constrains
Sj< Sjmax, j=1,2,3 ..., m (5)；
In formula: SjmaxFor jth branch road apparent energy maximum capacity；
It according to above-mentioned four kinds of constraint condition, is added in load transfer objective function, is constituted without about by penalty function of crossing the border
The augmented objective function of beam condition:
F=f+ λ1σ1； (6)；
In formula: λ1Referred to as penalty factor；σ1For inequality constraints condition.
As an improvement of the present invention, MAPSO load transfer algorithm is executed in the step d entirely to solve process specific
It is as follows:
Power distribution network network reconfiguration change-over switch state is to realize the means of non-faulting power supply interrupted district load transfer, due to distribution
Net uses closed loop design, the mode of open loop operation, so switch only exists disconnection and closure two ways.Using intelligent granule group
Optimization algorithm (MAPSO), the algorithm combine binary particle swarm algorithm (binary particle swarm
Optimization, BPSO) and multiple agent (Multi-Agent) concept, it realizes and solves；
Each Agent indicates a physics or abstract entity, can cooperate and compete, complete challenge
It solves；Assuming that in intelligent body mesh grid environment, Li,jBe coordinate be (i, j), wherein i, j=1,2 ..., Lsize, LsizeTable
Show network coordinate maximum value.
Define the neighborhood of particleAre as follows:
Wherein:
In BPSO algorithm, a group random particles are initialized, is updated by particle iteration, seeks optimal solution, is located at a n dimension
Search space in, particle xiIts speed and position are updated according to formula and formula；
In formula: subscript k indicates current iteration number；Indicate particles spatial position when kth time iteration；Indicate kth
Particle rapidity when secondary iteration；ω indicates inertia weight；c1、c2Indicate Studying factors；r1、r2Expression is evenly distributed on [0,1]
Random number；Individual extreme value and global extremum when respectively indicating kth time iteration；
In MAPSO, each particle regards an Agent as, and is fixed in mesh grid environment, by with its neighborhood grain
It is at war between sub- Agent and is operated with cooperation, constitute local optimum, and each particle also needs to execute changing for BPSO algorithm
For evolutionary mechanism, information exchange is carried out with global optimum's particle, is updated using position of the formula 9 to each particle, is finally obtained
Obtain globally optimal solution；
Intelligent granule colony optimization algorithm (MAPSO) solution procedure is as follows:
In searching process, guaranteeing the objective function of load transfer to lose load and switch number of operations is at least most
It is excellent, it is minimum optimization problem, Agent (L should be madei,j) corresponding fitness function value is high, determine Agent (Li,j) adaptation
Spend function are as follows:
Wherein, F is augmented objective function；CmaxFor definite value；
Load transfer solution procedure is as follows:
(1) block switch and interconnection switch can be operated by inputting distribution network initial information and inequality constraints condition, determination
State determines Agent Grid scale Lsize×Lsize, that is, determine population number selected in BPSO algorithm；
(2) control parameter of MAPSO system is set, maximum allowable the number of iterations T, Inertia Weight ω, Studying factors c are set
Etc. parameters；
(3) intelligent body grid environment L is constructedsize×Lsize, initial population is randomly generated under the conditions of control variables constraint,
Lsize×LsizeA Agent；
(4) using back substitution tidal current computing method progress Load flow calculation is pushed forward, the fitness function of current each Agent is assessed
fitness(Li,j)；
(5) according to the neighborhood information in grid environment, each position Agent is updated.Current Agent is set as Li,j, Mi,j
It is current Li,jThe maximum Agent of fitness function value in neighborhood, if meeting fitness (Mi,j)≤fitness(Li,j), then
Agent(Li,j) position of solution space remains unchanged, otherwise press L'i,j=Mi,j+rand(-1,1)·(Mi,j-Li,j) to tending to Mi,j
Position scan for updating, and still retain original information；
(6) iterative formula of BPSO algorithm is directly utilized, the current location for setting each particle is currently optimal as particle
Solve pBesti=fitness (Li,j), take gBest=minfitness (Li,j) it is the current optimal solution of group, to update each
Position and speed of the Agent in solution space.After updating every time, check whether speed exceeds [- 4,4], if exceeding the range,
It is the extreme value by rate limitation；
(7) if reaching maximum number of iterations or meeting the condition of convergence, stop iteration, export globally optimal solution, i.e., one
The combination of group switch state, otherwise returns to (4) step.
Compared with the existing technology, beneficial effects of the present invention are as follows:
1) firstly the need of the method for operation after the failure of regulation DG, then according to the practical operation situation of power distribution network, fault point
The DG method of operation behind position and failure considers the balance of isolated island internal loading capacity and DG power generation capacity, carries out the division of DG isolated island；It builds
The vertical distribution network load containing DG turns for model, distribution network load turn include for constraint condition in need of consideration network topology about
Beam, node voltage constraint, DG units limits, tributary capacity constraint.On the basis of particle swarm optimization algorithm, in conjunction with multiple agent
(Multi-Agent) system concept is constituted intelligent granule colony optimization algorithm (MAPSO).Algorithm objective function is to lose load most
Small and switch motion number is at least optimal progress path optimizing, obtains optimal solution.Being shown by emulation and network analysis can be with
Ground quickly and effectively realizes that active power distribution network expense fault zone load quickly turns to supply, to reduce failure or maintenance bring power failure
Loss.
2) this method considers a variety of constraint conditions, is operated based on MAPSO to switch and loss two targets of load carry out
Optimization, algorithm the convergence speed is fast, and accuracy rate is high, can restore the outer intelligent distribution network non-faulting outage area of isolated island to the maximum extent
Load quickly turns to supply.
Detailed description of the invention
Distribution network failure afterload turning solution schematic diagram of the Fig. 1 containing DG；
Fig. 2 the method for the invention flow chart.
Specific embodiment
In order to further enhance the appreciation and understanding of the invention, the present embodiment is further described with open source software clone below,
Clone is the trivial games using MVC pattern exploitation.
Embodiment 1: referring to Fig. 1-Fig. 2, a kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm
Turn to the described method comprises the following steps for method:
When power distribution network containing DG breaks down, load transfer step are as follows:
After step a. failure occurs, check whether non-faulting power supply interrupted district includes DG, if containing DG, is determined according to its type
Whether isolated operation or off-grid operation, turn if isolated operation walk b, otherwise turn walk d；
Step b. divides isolated operation range according to the DG isolated island division principle pre-established, remains important negative in isolated island
Lotus power supply；
Step c. searches the block switch that need to be disconnected when the formation of DG isolated island and outage area branch block switch position, is transferred to
Isolated operation mode；
It establishes the distribution network load containing DG to turn to adjust network structure information for model, propose excellent using intelligent body population
Change the non-faulting outage area restoration path outside algorithm search isolated island,
Power loss load is transferred to normal power supply region to the maximum extent to continue to power, to realize quickly turning for load
For；
Step d. executes MAPSO load transfer algorithm and entirely solves process；
Step e. is after major network load transfer, according to frequency, the amplitude of voltage and the phase of power distribution network, to load transfer model
It encloses interior contained isolated island and carries out simultaneous interconnecting operation, otherwise the method for operation before being restored to failure continues isolated operation.
The present invention is according to the practical operation situation and isolated island division principle rational isolated island of DG different types of after failure
Range of operation makes full use of DG generating capacity to maintain important load normal power supply in isolated island；It is negative that the power distribution network containing DG is established simultaneously
Lotus turns to adjust network structure information for model, proposes using the non-faulting outside Multi-agent Particle Swarm Optimization Algorithm search isolated island
Power loss load is transferred to normal power supply region to the maximum extent and continues to power, to realize load by outage area restoration path
Quickly turn supply.The emulation and analysis method can effectively solve the problems, such as that active power distribution network failure afterload is specialized in, and improve and calculate
Efficiency realizes load transfer, to guarantee the power supply reliability of power distribution network.
The step a is specific as follows, when power distribution network breaks down, carries out region division to power distribution network first, main to divide
For fault zone, normal power supply region and non-faulting power supply interrupted district.According to physical location of the distributed generation resource in topology, determine
The DG method of operation after failure.Be broadly divided into direct off-grid, it is grid-connected or can not isolated operation, grid-connected or three kinds of isolated operation, respectively
It is indicated with A, B, C type.
Isolated island division principle in the step b, specific as follows:
1) meet total power load in isolated island to be no more than under the premise of DG power generation total amount, it is preferential to guarantee important load power supply；
2) guarantee isolated island in device line it is in a safe condition, avoid overload with it is out-of-limit；
3) containing the c-type DG for having island operating capacity in non-faulting outage area, original is divided according to isolated island is pre-established
It is then translated into isolated operation mode, maintains important load power supply in isolated island；If DG containing Type B, it is divided into tool as far as possible
It is run in the isolated island for thering is the c-type DG of frequency modulation and voltage modulation ability to constitute, otherwise follows A type DG off-grid.
The distribution network load containing DG is established in the step c to turn to adjust network structure information for model, is proposed using intelligence
Body particle swarm optimization algorithm searches for the non-faulting outage area restoration path outside isolated island, specific as follows:
For breaking down containing distributed power distribution network, by network reconfiguration to non-faulting power supply interrupted district load transfer,
It during load transfer, needs to maximize load transfer, and requires switch number of operations minimum, extend switch and use year
Limit.
At least turn the objective function supplied as distribution network load to lose load minimum and switch motion number:
Wherein, ω1Indicate loss function of load weight coefficient, ω2To switch number of operations weight coefficient, ω1And ω2Root
According to actual conditions selection and ω1+ω2=1；LBFor fault zone total losses load capacity；γ is non-former for what is obtained after Fault Isolation
Hinder power supply interrupted district set；LiTo turn the capacity for load i；N is total number of switches；XjAnd X'jIt respectively indicates before failure and turns to open after supplying
The state for closing j takes 0 or 1, indicates that switch is in and is opened or closed.
Distribution network load turn supply solution be meet power network security operation and search the smallest solution of target function value to
Amount, so according to the actual situation, solution needs to include following constraint condition:
(1) not network topology constraint (not including DG)；
gk∈G； (2)；
In formula: gkFor the network structure in the region that restored electricity；G is all feasible network structure set.
(2) node voltage constrains；
Node voltage should be maintained in fixed range；
Uimin≤Ui≤Uimax, i=1,2,3 ..., n (3)；
In formula: n is node summary；UiminAnd UimaxRespectively node voltage UiUpper lower limit value；
(3) DG units limits
PDGt,min≤PDGt≤PDGt,max； (4)；
In formula: PDGt,minAnd PDGt,maxIndicate the upper lower limit value of t-th of DG power output；
(4) tributary capacity constrains
Sj< Sjmax, j=1,2,3 ..., m (5)；
In formula: SjmaxFor jth branch road apparent energy maximum capacity；
It according to above-mentioned four kinds of constraint condition, is added in load transfer objective function, is constituted without about by penalty function of crossing the border
The augmented objective function of beam condition:
F=f+ λ1σ1； (6)；
In formula: λ1Referred to as penalty factor；σ1For inequality constraints condition.
It is specific as follows entirely to solve process for execution MAPSO load transfer algorithm in the step d:
Power distribution network network reconfiguration change-over switch state is to realize the means of non-faulting power supply interrupted district load transfer, due to distribution
Net uses closed loop design, the mode of open loop operation, so switch only exists disconnection and closure two ways.Using intelligent granule group
Optimization algorithm (MAPSO), the algorithm combine binary particle swarm algorithm (binary particle swarm
Optimization, BPSO) and multiple agent (Multi-Agent) concept, it realizes and solves；
Each Agent indicates a physics or abstract entity, can cooperate and compete, complete challenge
It solves；Assuming that in intelligent body mesh grid environment, Li,jBe coordinate be (i, j), wherein i, j=1,2 ..., Lsize, LsizeTable
Show network coordinate maximum value.
Define the neighborhood of particleAre as follows:
Wherein:
In BPSO algorithm, a group random particles are initialized, is updated by particle iteration, seeks optimal solution, is located at a n dimension
Search space in, particle xiIts speed and position are updated according to formula and formula；
In formula: subscript k indicates current iteration number；Indicate particles spatial position when kth time iteration；Indicate kth
Particle rapidity when secondary iteration；ω indicates inertia weight；c1、c2Indicate Studying factors；r1、r2Expression is evenly distributed on [0,1]
Random number；Individual extreme value and global extremum when respectively indicating kth time iteration；
In MAPSO, each particle regards an Agent as, and is fixed in mesh grid environment, by with its neighborhood grain
It is at war between sub- Agent and is operated with cooperation, constitute local optimum, and each particle also needs to execute changing for BPSO algorithm
For evolutionary mechanism, information exchange is carried out with global optimum's particle, is updated using position of the formula 9 to each particle, is finally obtained
Obtain globally optimal solution；
Intelligent granule colony optimization algorithm (MAPSO) solution procedure is as follows:
In searching process, guaranteeing the objective function of load transfer to lose load and switch number of operations is at least most
It is excellent, it is minimum optimization problem, Agent (L should be madei,j) corresponding fitness function value is high, determine Agent (Li,j) adaptation
Spend function are as follows:
Wherein, F is augmented objective function；CmaxFor definite value；
Load transfer solution procedure is as follows:
(1) block switch and interconnection switch can be operated by inputting distribution network initial information and inequality constraints condition, determination
State determines Agent Grid scale Lsize×Lsize, that is, determine population number selected in BPSO algorithm；
(2) control parameter of MAPSO system is set, maximum allowable the number of iterations T, Inertia Weight ω, Studying factors c are set
Etc. parameters；
(3) intelligent body grid environment L is constructedsize×Lsize, initial population is randomly generated under the conditions of control variables constraint,
Lsize×LsizeA Agent；
(4) using back substitution tidal current computing method progress Load flow calculation is pushed forward, the fitness function of current each Agent is assessed
fitness(Li,j)；
(5) according to the neighborhood information in grid environment, each position Agent is updated.Current Agent is set as Li,j, Mi,j
It is current Li,jThe maximum Agent of fitness function value in neighborhood, if meeting fitness (Mi,j)≤fitness(Li,j), then
Agent(Li,j) position of solution space remains unchanged, otherwise press L'i,j=Mi,j+rand(-1,1)·(Mi,j-Li,j) to tending to Mi,j
Position scan for updating, and still retain original information；
(6) iterative formula of BPSO algorithm is directly utilized, the current location for setting each particle is currently optimal as particle
Solve pBesti=fitness (Li,j), take gBest=minfitness (Li,j) it is the current optimal solution of group, to update each
Position and speed of the Agent in solution space.After updating every time, check whether speed exceeds [- 4,4], if exceeding the range,
It is the extreme value by rate limitation；
(7) if reaching maximum number of iterations or meeting the condition of convergence, stop iteration, export globally optimal solution, i.e., one
The combination of group switch state, otherwise returns to (4) step.
Application Example:
Using IEEE distribution network system containing DG as simulation example, in Fig. 1, node 7,22,23,36,41 is power supply point, load section
Point 110,111,121,122,133,135,144 is important load, remaining is common load.Interconnection switch indicates with dot, just
It is often in an off state when operation, contact effect has been closed when abnormal；Block switch indicates with diamond shape, is closure when normal operation
State disconnects isolated fault section when abnormal, hollow expression disconnects, when solid expression closure power distribution network operates normally.As shown,
Interconnection switch 1,4,12,14,24 is in disjunction state, remaining block switch is in "on" position, wherein DG installation site and
Operating parameter such as table 1.
Sizing grid L is chosen hereinsize=6, maximum number of iterations Tmax=100, inertia weight factor ω=
0.7298, Studying factors c1=c2=1.496 2；Weight coefficient ω1=0.6, ω2=0.4；Penalty factor λ1=λ2=800.
When power distribution network containing DG breaks down, load transfer step are as follows:
A. after failure occurs, check whether non-faulting power supply interrupted district includes DG, if containing DG, is decided whether according to its type
Isolated operation or off-grid operation, turn to walk b if isolated operation, otherwise turn to walk d；
B. according to the DG isolated island division principle pre-established, isolated operation range is divided, important load in isolated island is maintained to supply
Electricity；
C. the block switch that need to be disconnected when the formation of DG isolated island and outage area branch block switch position are searched, isolated island is transferred to
Operational mode；
D. it executes MAPSO load transfer algorithm and entirely solves process；
E. after major network load transfer, according to frequency, the amplitude of voltage and the phase of power distribution network, within the scope of load transfer
Contained isolated island carries out simultaneous interconnecting operation, otherwise the method for operation before being restored to failure continues isolated operation.
1 DG installation site of table and operating parameter
| Number | Installation site | Type | The method of operation after failure | Capacity/kw | Power factor |
| DG1 | 2 | A | Direct off-grid | 600 | 0.85 |
| DG2 | 10 | C | Grid-connected or isolated operation | 1000 | 0.80 |
| DG3 | 19 | C | Grid-connected or isolated operation | 2800 | 0.85 |
| DG4 | 29 | C | Grid-connected or isolated operation | 2500 | 0.80 |
| DG5 | 32 | B | It is grid-connected or can not isolated operation | 1200 | 0.85 |
| DG6 | 39 | C | Grid-connected or isolated operation | 800 | 0.90 |
| DG7 | 46 | B | It is grid-connected or can not isolated operation | 2000 | 0.90 |
It is assumed that permanent fault occurs and causes load power loss simultaneously at the bus of power supply point 7 and 23, segmentation is disconnected rapidly and is opened
11,42,26,30 are closed, power supply interrupted district is as shown in Figure 1, specific load condition is shown in Table 2.Since 7 downstream of node includes DG2, node 23
Downstream includes DG1 and DG4, therefore need to carry out isolated island division to DG2, DG1, DG4, and DG1 off-grid, DG2, DG4 off-the-line are at single user
Isolated operation divides range and is shown in Table 3.
2 outage area load parameter of table
3 isolated island of table divides region
It is iterated solution by MAPSO algorithm, is operated according to the switch of solution after carrying out load transfer, the load of recovery
Situation is as shown in table 4；
4 MAPSO load transfer strategy of table
Finally, it solves by the above method, breaks down simultaneously at 7,23, then the outer non-faulting outage area load of isolated island
By power supply point 36 and 22 by closure interconnection switch 1,4,12, power loss load transfer is realized；After major network service restoration, DG2,
DG4 is incorporated into the power networks by the realization of closure switch 9,24, to restore the common load being removed power supply.
Be simulation example by IEEE distribution network system containing DG, carry out specific emulation and analysis shows, it can be seen that institute
The active power distribution network failure afterload based on intelligent granule colony optimization algorithm mentioned turns effectively provide operation strategy for method, this
Method considers a variety of constraint conditions, is operated based on MAPSO to switch and loss two targets of load optimize, algorithmic statement
Speed is fast, and accuracy rate is high, can restore quickly turning for the outer intelligent distribution network non-faulting outage area load of isolated island to the maximum extent
For.
It should be noted that above-described embodiment is only presently preferred embodiments of the present invention, there is no be used to limit the present invention
Protection scope, the equivalent substitution or substitution made based on the above technical solution belongs to protection model of the invention
It encloses.
Claims (5)
1. a kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method, which is characterized in that institute
State method the following steps are included:
When power distribution network containing DG breaks down, load transfer step are as follows:
Step a: after failure occurs, check whether non-faulting power supply interrupted district includes DG, if containing DG, is decided whether according to its type
Isolated operation or off-grid operation, turn to walk b if isolated operation, otherwise turn to walk d；
Step b: according to the DG isolated island division principle pre-established, dividing isolated operation range, and important load in isolated island is maintained to supply
Electricity；
Step c: the block switch that need to be disconnected when the formation of DG isolated island and outage area branch block switch position are searched, isolated island is transferred to
Operational mode；
It establishes the distribution network load containing DG to turn to adjust network structure information for model, proposes to use intelligent body Particle Swarm Optimization
Method searches for the non-faulting outage area restoration path outside isolated island,
Step d: it executes MAPSO load transfer algorithm and entirely solves process；
Step e: after major network load transfer, according to frequency, the amplitude of voltage and the phase of power distribution network, within the scope of load transfer
Contained isolated island carries out simultaneous interconnecting operation, otherwise the method for operation before being restored to failure continues isolated operation.
2. the active power distribution network failure afterload according to claim 1 based on intelligent granule colony optimization algorithm turns supplier
Method, which is characterized in that
The step a is specific as follows, when power distribution network breaks down, carries out region division to power distribution network first, is broadly divided into event
Barrier region, normal power supply region and non-faulting power supply interrupted district determine failure according to physical location of the distributed generation resource in topology
The DG method of operation afterwards, be broadly divided into direct off-grid, it is grid-connected or can not isolated operation, grid-connected or three kinds of isolated operation, respectively with A,
B, C type indicates.
3. the active power distribution network failure afterload according to claim 1 based on intelligent granule colony optimization algorithm turns supplier
Method, which is characterized in that
Isolated island division principle in the step b, specific as follows:
1) meet total power load in isolated island to be no more than under the premise of DG power generation total amount, it is preferential to guarantee important load power supply；
2) guarantee isolated island in device line it is in a safe condition, avoid overload with it is out-of-limit；
It 3), will according to isolated island division principle is pre-established containing the c-type DG for having island operating capacity in non-faulting outage area
It is converted into isolated operation mode, maintains important load power supply in isolated island；If DG containing Type B, it is divided into have as far as possible and is adjusted
It is run in the isolated island that the c-type DG of frequency modulation pressure energy power is constituted, otherwise follows A type DG off-grid.
4. the active power distribution network failure afterload according to claim 1 based on intelligent granule colony optimization algorithm turns supplier
Method, which is characterized in that
The distribution network load containing DG is established in the step c to turn to adjust network structure information for model, proposes to use intelligent body grain
Subgroup optimization algorithm searches for the non-faulting outage area restoration path outside isolated island, specific as follows:
For breaking down containing distributed power distribution network, by network reconfiguration to non-faulting power supply interrupted district load transfer, with damage
It loses load minimum and switch motion number is at least used as distribution network load to turn the objective function supplied:
Wherein, ω1Indicate loss function of load weight coefficient, ω2To switch number of operations weight coefficient, ω1And ω2According to reality
Situation chooses and ω1+ω2=1；LBFor fault zone total losses load capacity；γ has a power failure for the non-faulting obtained after Fault Isolation
Regional ensemble；LiTo turn the capacity for load i；N is total number of switches；XjAnd X'jRespectively indicate before failure and turn switch j after supplying
State takes 0 or 1, indicates that switch is in and is opened or closed.
The solution that distribution network load turns to supply is to meet power network security operation and the search the smallest solution vector of target function value, institute
With according to the actual situation, solution needs to include following constraint condition:
(1) not network topology constraint (not including DG)；
gk∈G； (2)；
In formula: gkFor the network structure in the region that restored electricity；G is all feasible network structure set.
(2) node voltage constrains；
Node voltage should be maintained in fixed range；
Uimin≤Ui≤Uimax, i=1,2,3 ..., n (3)；
In formula: n is node summary；UiminAnd UimaxRespectively node voltage UiUpper lower limit value；
(3) DG units limits；
PDGt,min≤PDGt≤PDGt,max； (4)；
In formula: PDGt,minAnd PDGt,maxIndicate the upper lower limit value of t-th of DG power output；
(4) tributary capacity constrains；
Sj< Sjmax, j=1,2,3 ..., m (5)；
In formula: SjmaxFor jth branch road apparent energy maximum capacity；
It according to above-mentioned four kinds of constraint condition, is added in load transfer objective function, is constituted without constraint item by penalty function of crossing the border
The augmented objective function of part:
F=f+ λ1σ1； (6)；
In formula: λ1Referred to as penalty factor；σ1For inequality constraints condition.
5. the active power distribution network failure afterload according to claim 4 based on intelligent granule colony optimization algorithm turns supplier
Method, which is characterized in that
It is specific as follows entirely to solve process for execution MAPSO load transfer algorithm in the step d:
Using intelligent granule colony optimization algorithm (MAPSO), which combines binary particle swarm algorithm (binary
Particle swarm optimization, BPSO) and multiple agent (Multi-Agent) concept, it realizes and solves；
Each Agent indicates a physics or abstract entity, can cooperate and compete, complete asking for challenge
Solution；Assuming that in intelligent body mesh grid environment, Li,jBe coordinate be (i, j), wherein i, j=1,2 ..., Lsize, LsizeIt indicates
Network coordinate maximum value.
Define the neighborhood of particleAre as follows:
Wherein:
In BPSO algorithm, a group random particles are initialized, is updated by particle iteration, seeks optimal solution, is located at searching for a n dimension
In rope space, particle xiIts speed and position are updated according to formula and formula；
In formula: subscript k indicates current iteration number；Indicate particles spatial position when kth time iteration；Indicate kth time repeatedly
For when particle rapidity；ω indicates inertia weight；c1、c2Indicate Studying factors；r1、r2Expression is evenly distributed on the random of [0,1]
Number；Individual extreme value and global extremum when respectively indicating kth time iteration；
In MAPSO, each particle regards an Agent as, and is fixed in mesh grid environment, by with its neighborhood particle
It is at war between Agent and is operated with cooperation, constitute local optimum, and each particle also needs to execute the iteration of BPSO algorithm
Evolutionary mechanism is carried out information exchange with global optimum's particle, is updated using position of the formula 9 to each particle, final to obtain
Globally optimal solution；
Intelligent granule colony optimization algorithm (MAPSO) solution procedure is as follows:
In searching process, guarantee the objective function of load transfer with lose load and switch number of operations be at least it is optimal, be
Minimum optimization problem should make Agent (Li,j) corresponding fitness function value is high, determine Agent (Li,j) fitness letter
Number are as follows:
Wherein, F is augmented objective function；CmaxFor definite value；
Load transfer solution procedure is as follows:
(1) block switch and interconnection switch state can be operated by inputting distribution network initial information and inequality constraints condition, determination,
Determine Agent Grid scale Lsize×Lsize, that is, determine population number selected in BPSO algorithm；
(2) control parameter of MAPSO system is set, maximum allowable the number of iterations T, the ginseng such as Inertia Weight ω, Studying factors c are set
Number；
(3) intelligent body grid environment L is constructedsize×Lsize, initial population, L are randomly generated under the conditions of control variables constraintsize
×LsizeA Agent；
(4) using back substitution tidal current computing method progress Load flow calculation is pushed forward, the fitness function of current each Agent is assessed
fitness(Li,j)；
(5) according to the neighborhood information in grid environment, each position Agent is updated.Current Agent is set as Li,j, Mi,jIt is to work as
Preceding Li,jThe maximum Agent of fitness function value in neighborhood, if meeting fitness (Mi,j)≤fitness(Li,j), then Agent
(Li,j) position of solution space remains unchanged, otherwise press L'i,j=Mi,j+rand(-1,1)·(Mi,j-Li,j) to tending to Mi,jPosition
It sets and scans for updating, and still retain original information；
(6) iterative formula for directly utilizing BPSO algorithm, sets the current location of each particle as the current optimal solution of particle
pBesti=fitness (Li,j), take gBest=minfitness (Li,j) it is the current optimal solution of group, to update each Agent
Position and speed in solution space.After updating every time, check whether speed exceeds [- 4,4], if exceeding the range, by speed
It is limited to the extreme value；
(7) if reaching maximum number of iterations or meeting the condition of convergence, stop iteration, export globally optimal solution, i.e., one group is opened
Otherwise the combination of off status returns to (4) step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910301463.4A CN110086153A (en) | 2019-04-15 | 2019-04-15 | A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910301463.4A CN110086153A (en) | 2019-04-15 | 2019-04-15 | A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110086153A true CN110086153A (en) | 2019-08-02 |
Family
ID=67415166
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910301463.4A Pending CN110086153A (en) | 2019-04-15 | 2019-04-15 | A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110086153A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111313419A (en) * | 2019-12-13 | 2020-06-19 | 山东大学 | Power distribution network elastic lifting method and system based on dynamic reconfiguration in extreme weather |
| CN111682525A (en) * | 2020-05-28 | 2020-09-18 | 天津大学 | Load transfer method based on optimal flow method and Mayeda spanning tree method |
| CN112132354A (en) * | 2020-09-28 | 2020-12-25 | 国网江苏省电力有限公司苏州供电分公司 | Urban power grid load advanced transfer method considering network toughness and system thereof |
| CN112149347A (en) * | 2020-09-16 | 2020-12-29 | 北京交通大学 | Power distribution network load transfer method based on deep reinforcement learning |
| CN112583633A (en) * | 2020-10-26 | 2021-03-30 | 东北大学秦皇岛分校 | Distributed optimization method of directed multi-agent network based on rough information |
| CN112886617A (en) * | 2021-02-03 | 2021-06-01 | 华南理工大学 | Commutation control method and system based on BPSO algorithm |
| CN113507116A (en) * | 2021-07-08 | 2021-10-15 | 国网河北省电力有限公司电力科学研究院 | Power distribution network load transfer method, device, equipment and storage medium |
| CN113673065A (en) * | 2021-08-12 | 2021-11-19 | 国网浙江义乌市供电有限公司 | Loss reduction method for automatic reconstruction of power distribution network |
| CN114285161A (en) * | 2021-12-06 | 2022-04-05 | 南京国电南自电网自动化有限公司 | Distributed power distribution protection fault self-healing overload processing method |
| CN114865625A (en) * | 2022-06-09 | 2022-08-05 | 国网湖北省电力有限公司鄂州供电公司 | Power distribution network fault recovery method comprising microgrid |
| CN116298686A (en) * | 2023-03-16 | 2023-06-23 | 广东电网有限责任公司广州供电局 | Fault positioning method, device, equipment and medium applied to power distribution network |
| CN116973694B (en) * | 2023-09-22 | 2023-12-12 | 国网浙江宁波市鄞州区供电有限公司 | Power distribution network fault diagnosis optimization method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103078391A (en) * | 2013-01-10 | 2013-05-01 | 长兴县供电局 | Power distribution network power supply power restoration method based on photovoltaic power generation system |
| CN104112165A (en) * | 2014-05-19 | 2014-10-22 | 浙江工业大学 | Intelligent power distribution network fault recovery method based on multi-target discrete particle swarm |
| CN104362623A (en) * | 2014-11-10 | 2015-02-18 | 国家电网公司 | Multi-target network reestablishing method for active power distribution network |
| CN104578427A (en) * | 2015-01-27 | 2015-04-29 | 国家电网公司 | Fault self-healing method for power distribution network containing microgrid power source |
| CN105069517A (en) * | 2015-07-14 | 2015-11-18 | 浙江工业大学 | Power distribution network multi-objective fault recovery method based on hybrid algorithm |
| CN104092211B (en) * | 2014-07-14 | 2016-03-02 | 国家电网公司 | A kind of switching optimization method adapting to power distribution network self-healing requirement |
-
2019
- 2019-04-15 CN CN201910301463.4A patent/CN110086153A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103078391A (en) * | 2013-01-10 | 2013-05-01 | 长兴县供电局 | Power distribution network power supply power restoration method based on photovoltaic power generation system |
| CN104112165A (en) * | 2014-05-19 | 2014-10-22 | 浙江工业大学 | Intelligent power distribution network fault recovery method based on multi-target discrete particle swarm |
| CN104092211B (en) * | 2014-07-14 | 2016-03-02 | 国家电网公司 | A kind of switching optimization method adapting to power distribution network self-healing requirement |
| CN104362623A (en) * | 2014-11-10 | 2015-02-18 | 国家电网公司 | Multi-target network reestablishing method for active power distribution network |
| CN104578427A (en) * | 2015-01-27 | 2015-04-29 | 国家电网公司 | Fault self-healing method for power distribution network containing microgrid power source |
| CN105069517A (en) * | 2015-07-14 | 2015-11-18 | 浙江工业大学 | Power distribution network multi-objective fault recovery method based on hybrid algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 赵凤贤等: "基于MAPSO优化的智能配电网大面积断电供电恢复", 《中国电力》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111313419A (en) * | 2019-12-13 | 2020-06-19 | 山东大学 | Power distribution network elastic lifting method and system based on dynamic reconfiguration in extreme weather |
| CN111682525A (en) * | 2020-05-28 | 2020-09-18 | 天津大学 | Load transfer method based on optimal flow method and Mayeda spanning tree method |
| CN112149347B (en) * | 2020-09-16 | 2023-12-26 | 北京交通大学 | Power distribution network load transfer method based on deep reinforcement learning |
| CN112149347A (en) * | 2020-09-16 | 2020-12-29 | 北京交通大学 | Power distribution network load transfer method based on deep reinforcement learning |
| CN112132354A (en) * | 2020-09-28 | 2020-12-25 | 国网江苏省电力有限公司苏州供电分公司 | Urban power grid load advanced transfer method considering network toughness and system thereof |
| CN112132354B (en) * | 2020-09-28 | 2022-06-28 | 国网江苏省电力有限公司苏州供电分公司 | Urban power grid load advanced transfer method considering network toughness and system thereof |
| CN112583633A (en) * | 2020-10-26 | 2021-03-30 | 东北大学秦皇岛分校 | Distributed optimization method of directed multi-agent network based on rough information |
| CN112886617A (en) * | 2021-02-03 | 2021-06-01 | 华南理工大学 | Commutation control method and system based on BPSO algorithm |
| CN113507116A (en) * | 2021-07-08 | 2021-10-15 | 国网河北省电力有限公司电力科学研究院 | Power distribution network load transfer method, device, equipment and storage medium |
| CN113507116B (en) * | 2021-07-08 | 2022-07-22 | 国网河北省电力有限公司电力科学研究院 | Power distribution network load transfer method, device, equipment and storage medium |
| CN113673065A (en) * | 2021-08-12 | 2021-11-19 | 国网浙江义乌市供电有限公司 | Loss reduction method for automatic reconstruction of power distribution network |
| CN114285161A (en) * | 2021-12-06 | 2022-04-05 | 南京国电南自电网自动化有限公司 | Distributed power distribution protection fault self-healing overload processing method |
| CN114285161B (en) * | 2021-12-06 | 2024-03-19 | 南京国电南自电网自动化有限公司 | Distributed power distribution protection fault self-healing overload processing method |
| CN114865625A (en) * | 2022-06-09 | 2022-08-05 | 国网湖北省电力有限公司鄂州供电公司 | Power distribution network fault recovery method comprising microgrid |
| CN116298686A (en) * | 2023-03-16 | 2023-06-23 | 广东电网有限责任公司广州供电局 | Fault positioning method, device, equipment and medium applied to power distribution network |
| CN116973694B (en) * | 2023-09-22 | 2023-12-12 | 国网浙江宁波市鄞州区供电有限公司 | Power distribution network fault diagnosis optimization method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110086153A (en) | A kind of active power distribution network failure afterload based on intelligent granule colony optimization algorithm turns for method | |
| Mishra et al. | A comprehensive review on power distribution network reconfiguration | |
| CN104820865B (en) | Intelligent distribution network fault recovery intelligent optimization method based on graph theory | |
| CN106410808B (en) | Universal micro-capacitance sensor group distributed control method comprising invariable power and droop control | |
| CN105046022B (en) | A kind of intelligent distribution network self-healing method based on improvement ant group algorithm | |
| CN109768573A (en) | Var Optimization Method in Network Distribution based on multiple target difference grey wolf algorithm | |
| Ustun et al. | Implementation of Dijkstra's algorithm in a dynamic microgrid for relay hierarchy detection | |
| CN106777449A (en) | Distribution Network Reconfiguration based on binary particle swarm algorithm | |
| CN108182498A (en) | The restorative reconstructing method of distribution network failure | |
| CN111817345A (en) | Reconstruction method for power distribution network with distributed power supply after serious fault | |
| CN108462194A (en) | A kind of wide area optimization method for low-voltage network three-phase load unbalance | |
| Karimianfard et al. | An initial-point strategy for optimizing distribution system reconfiguration | |
| CN108832615A (en) | A kind of reconstruction method of power distribution network and system based on improvement binary particle swarm algorithm | |
| CN107147102A (en) | Direct-current grid networking distributed and coordinated control method based on multiple agent | |
| CN109768546A (en) | The active power distribution network service restoration method coordinated based on more intelligent Sofe Switch | |
| CN104218681B (en) | A kind of control method for reducing isolated island micro-capacitance sensor cutting load cost | |
| CN109004639B (en) | Power distribution network partition power supply recovery strategy optimization method based on completely distributed algorithm | |
| CN106611966B (en) | Multi-inverter type exchanges micro-capacitance sensor distribution economy Automatic Generation Control algorithm | |
| Zhang et al. | A fault reconfiguration strategy based on adjustable space operator discrete state transition algorithm for ship microgrid system | |
| He et al. | Multi-objective operation mode optimization of medium voltage distribution networks based on improved binary particle swarm optimization | |
| Armbruster et al. | The maximum flow algorithm applied to the placement and distributed steady-state control of UPFCs | |
| CN109390971A (en) | A kind of power distribution network multiple target active reconstructing method based on genetic algorithm well-matched in social and economic status | |
| CN105117796B (en) | Piconet island division methods based on quantum evolutionary algorithm | |
| Fan et al. | An integrated power restoration method based on improved genetic algorithm for active distribution network | |
| Yang et al. | Optimal configuration method of distributed generation based on load transfer in distribution network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190806 Address after: Four pailou Nanjing Xuanwu District of Jiangsu Province, No. 2 210096 Applicant after: Southeast University Applicant after: Suzhou power supply branch, Jiangsu Electric Power Co., Ltd. Address before: Four pailou Nanjing Xuanwu District of Jiangsu Province, No. 2 210096 Applicant before: Southeast University |
|
| TA01 | Transfer of patent application right | ||
| 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 |
Application publication date: 20190802 |
|
| RJ01 | Rejection of invention patent application after publication |