CN107508280B - A kind of reconstruction method of power distribution network and system - Google Patents
A kind of reconstruction method of power distribution network and system Download PDFInfo
- Publication number
- CN107508280B CN107508280B CN201710671270.9A CN201710671270A CN107508280B CN 107508280 B CN107508280 B CN 107508280B CN 201710671270 A CN201710671270 A CN 201710671270A CN 107508280 B CN107508280 B CN 107508280B
- Authority
- CN
- China
- Prior art keywords
- power
- node
- harmony
- function
- indicate
- 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.)
- Active
Links
Classifications
-
- 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
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Abstract
The present invention discloses a kind of reconstruction method of power distribution network and system.This method comprises: establishing the interval model of load and distributed energy；Active power loss expense is established respectively, it is expected to lack power supply volume and is switched the first object majorized function of operating cost；Determine the constraint condition of the first object majorized function；According to the interval model of the load and distributed energy, the weight of the constraint condition and each first object majorized function, the first object majorized function is normalized, the second objective optimization function is obtained；Set the node of the power distribution network switches on-off initial value；The second objective optimization function is solved using harmonic search algorithm；Initial value is switched on-off according to the node of the corresponding power distribution network of solution of the smallest second objective optimization function, carries out power distribution network reconfiguration.The present invention has comprehensively considered the random fluctuation bring uncertainty of load, distributed energy, has high consistency, accuracy with higher with actual state.
Description
Technical field
The present invention relates to distribution network planning technical fields, more particularly to a kind of reconstruction method of power distribution network and system.
Background technique
Distribution Networks Reconfiguration is one and switches over adjustment to the state of cut-offfing of interconnection switch in network and block switch
Process.By network reconfiguration, new topological network is enabled to reduce network loss, improves operational reliability.In restructuring procedure, timing
It influences and environmental factor will inevitably lead to data and generate uncertainty.These uncertainties are usually expressed as the wave of load
Dynamic, fluctuation of plant maintenance parameter etc..By taking negative rules as an example, usually needed in load prediction with uncertain number
Value section describes the load of following a period of time.Such numerical value more meets objective demand.And uncertain numerical value is utilized to obtain
Prediction result, Electric Power Network Planning, risk analysis, in terms of it is more reliable and scientific.
The reconstruct factor that traditional reconstruction method of power distribution network considers is limited so that the reconstruction result generated have it is certain inclined
Difference.
Summary of the invention
The embodiment of the present invention provides a kind of reconstruction method of power distribution network and system, is considered with solving the reconstructing method of the prior art
Reconstruct factor it is limited so that reconstruction result has the problem of certain deviation.
In a first aspect, providing a kind of reconstruction method of power distribution network, comprising: establish the interval model of load and distributed energy；
Active power loss expense is established respectively, it is expected to lack power supply volume and is switched the first object majorized function of operating cost；Determine described
The constraint condition of one objective optimization function；According to the interval model of the load and distributed energy, the constraint condition and every
The weight of the one first object majorized function, is normalized the first object majorized function, obtains the second mesh
Mark majorized function；Set the node of the power distribution network switches on-off initial value；Described second is solved using harmonic search algorithm
Objective optimization function；According to the switch of the node of the corresponding power distribution network of solution of the smallest second objective optimization function
On-off initial value carries out power distribution network reconfiguration.
Second aspect provides a kind of power distribution network reconfiguration system, comprising: first establishes module, for establishing load and distribution
The interval model of the formula energy；Second establishes module, for establishing active power loss expense respectively, it is expected to lack power supply volume and switch operation
The first object majorized function of expense；Determining module, for determining the constraint condition of the first object majorized function；Normalization
Module, for according to the interval model of the load and distributed energy, the constraint condition and each objective optimization
The first object majorized function is normalized in the weight of function, obtains the second objective optimization function；Set mould
Block, the node for setting the power distribution network switch on-off initial value；Module is solved, for solving using harmonic search algorithm
The second objective optimization function；Reconstructed module, for the corresponding institute of solution according to the smallest second objective optimization function
The initial value that switches on-off of the node of power distribution network is stated, power distribution network reconfiguration is carried out.
In this way, it is uncertain to have comprehensively considered load, the random fluctuation bring of distributed energy in the embodiment of the present invention
Property, there are high consistency, accuracy with higher with actual state；Devise power distribution network reconfiguration multi-objective Model, meter and
Operation network loss, power supply reliability, switch operating cost, optimization aim covering comprehensively, are conducive to optimize power distribution network supply path,
Power distribution network operational efficiency is improved, realizing reduces network loss, eliminate overload, balanced load, improve quality of voltage, promote power supply reliability
The purpose of with economic benefit；The consideration factor in reconstruct decision process has directly been reacted, there is stronger practicability；Based on use
Harmonic search algorithm solves multiple target, the nonlinear power distribution network reconfiguration model calculated containing bounded-but-unknown uncertainty, and it is good to solve performance
Good, convergence is fast, precision is high.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of the reconstruction method of power distribution network of the embodiment of the present invention；
Fig. 2 is the operation characteristic schematic diagram of induction conductivity；
Fig. 3 is the power curve schematic diagram of wind-driven generator；
Fig. 4 is the schematic diagram of the topologies change for the network that the harmonic search algorithm of the embodiment of the present invention generates；
Fig. 5 is the step flow chart that the second objective optimization function is solved using harmonic search algorithm of the embodiment of the present invention；
Fig. 6 is the structural block diagram of the power distribution network reconfiguration system of the embodiment of the present invention；
Fig. 7 is the schematic diagram before IEEE33 node power distribution network reconfiguration；
Fig. 8 is the reconstruction method of power distribution network using the embodiment of the present invention to the signal after IEEE33 node power distribution network reconfiguration
Figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
The embodiment of the present invention discloses a kind of reconstruction method of power distribution network.As shown in Figure 1, this method comprises the following steps that
Step S10: the interval model of load and distributed energy is established.
Step S20: active power loss expense is established respectively, it is expected to lack power supply volume and switchs the first object optimization of operating cost
Function.
Step S30: the constraint condition of first object majorized function is determined.
Step S40: according to the interval model of load and distributed energy, constraint condition and per objective optimization function one by one
First object majorized function is normalized in weight, obtains the second objective optimization function；
Step S50: set the node of power distribution network switches on-off initial value.
Step S60: the second objective optimization function is solved using harmonic search algorithm.
Step S70: it is switched on-off according to the node of the corresponding power distribution network of solution of the smallest second objective optimization function
Initial value carries out power distribution network reconfiguration.
Specifically, in step S10:
The interval model of the load and distributed energy includes: negative rules model, wind-power electricity generation active power area
Between model, wind-power electricity generation reactive power interval model, photovoltaic power generation active power interval model and photovoltaic power generation reactive power area
Between model.
(1) negative rules model
Negative rules model are as follows:
Wherein, [S] indicates the waving interval of the practical complex power S of induction conductivity.The reality of [P] expression induction conductivity
The waving interval of active-power P.[Q] indicates the waving interval of the practical reactive power Q of induction conductivity.Indicate the power factor of induction conductivityWaving interval.
The maximum load of distribution network load accounting is induction conductivity.The parameter for characterizing induction conductivity operation characteristic is main
Efficient η, power factorWith revolving speed n.These parameters are the functions of induction conductivity load factor β.As shown in Fig. 2, being
The operation characteristic schematic diagram of induction conductivity, wherein M and s is respectively the electromagnetic torque and revolutional slip of induction conductivity.Induced electricity
Motivation power factorDecline with the reduction of load factor β, and load factor is lower, power factorDecline faster.From
From the point of view of efficiency and power factor, the load factor β of induction conductivity is optimal operating condition in 0.75~0.85 range.By reality
The influence of load change can set induction conductivity and generally operate in load factor β as in 0.3~0.9 range.Work as induction electric
When the load factor β of machine changes within this range, power factor can be seen that from the power factor characteristic of induction conductivity
Amplitude of variation very little, can be approximately considered and float up and down ± 1%.
Therefore, as the rated active power P of given induction conductivityNAnd power factorWhen, it may be assumed that it is practical to have
Function power P is in section [0.3PN, 0.9PN] the interior fluctuation of range, it is denoted as [P]；And power factor?It changes, is denoted as in rangeThen the practical complex power S of induction conductivity is believed that
It is changed on section [S].
(2) wind-power electricity generation active power interval model
Wind-power electricity generation is the technology that traditional wind energy is converted to electric energy.Wind is the motive power of wind power generating set.Blower
Power output and wind speed it is closely related.Wind speed has randomness, and the present invention describes wind speed using typical Weibull distribution.Prestige
Boolean's distribution is specific as follows:
Wherein, k is form factor, and value is embodied in Weibull distribution probability density function generally between 1.8~2.3
Curve shape on.μ is scale coefficient, indicates the mean wind speed of institute sector of observation in a certain amount of time.
It is generally believed that the power output of wind power generating set is associated with the cube of wind speed at axial fan hub height,
I.e.
Wherein, PRFor the nominal output of wind power generating set.PW(v) wind power generating set power when be wind speed being v goes out
Power.N is the quantity of wind-driven generator.ρ is atmospheric density.CpFor energy conversion efficiency.R is the radius of the inswept area of wind wheel.vRFor
Rated wind speed.vciTo cut wind speed.vcoFor cut-out wind speed.
When actual wind speed is greater than incision wind speed vciWhen, wind-driven generator starting issues electric energy；When actual wind speed is greater than specified
Wind speed vRWhen, wind-driven generator keeps rated power output；When wind speed is less than incision wind speed vciOr it is higher than cut-out wind speed vcoWhen, wind
Power generator is out of service and and grid disconnection.The power curve of wind-driven generator is substantially as shown in Figure 3.
The section modeling of wind-driven generator power output, that is, specify a section [Pw down, Pw up], so that it is met following two item
Part: 1) power output of wind-driven generator has sufficiently large probability to be in section [Pw down, Pw up] in；2) section [Pw down, Pw up] width
Degree is answered as small as possible, and at least it is the section of a limited width, is not a unlimited range.Therefore, wind-force can be passed through
The requirement met needed for the Probability Forms characterization wind-power electricity generation active power interval model of the active power interval model that generates electricity description.
Specifically, wind-power electricity generation active power interval model are as follows:
Wherein,Expression eventFor genuine probability, εwExpression and wind
The parameter of probability correlation of the power output of power generator in interval model, PWIndicate wind-power electricity generation active power, Pw downAnd Pw up
The respectively lower and upper limit of wind-power electricity generation active power.
(3) wind-power electricity generation reactive power interval model
Wind-driven generator also issues reactive power while issuing active power, it is generally recognized that the power of wind-driven generator
Factor be it is determining, then pass through wind-power electricity generation active power interval model [Pw down, Pw up] its reactive power area can be calculated
Between model [Qw down, Qw up]。
Specifically, wind-power electricity generation reactive power interval model are as follows:
Wherein, cos θwFor the power factor of wind-driven generator, Qw downAnd Qw upRespectively wind-power electricity generation reactive power
Lower and upper limit.
(4) photovoltaic power generation active power interval model
For photovoltaic power generation, the output power of photovoltaic generating system and suffered Intensity of the sunlight are closely related, therefore,
The randomness of Intensity of the sunlight variation brings randomness to the output power of photovoltaic generating system.In inventive embodiments,
Intensity of illumination is calculated using beta distribution, probability density function is as follows:
Wherein, s is Intensity of the sunlight, and value range is a≤s≤b, and a, b are respectively the sun in the time of somewhere section
The minimum value and maximum value of intensity of illumination.Г is Gamma function.
Based on Intensity of the sunlight, the output power of photovoltaic generating system can be calculated according to the following formula:
PPV=ξ × cos θ × ηm×Ap×ηp。
Wherein, ξ is Intensity of the sunlight.θ is the incidence angle of sunlight.ηmFor MPPT (Maximum Power Point
Tracking, MPPT maximum power point tracking) efficiency.ApFor photovoltaic cell plate suqare.ηpFor the transfer efficiency of photovoltaic cell.
Using sectionThe output power of photovoltaic generating system is characterized, which need to meet photovoltaic power generation
The requirement of active power interval model.
Specifically, photovoltaic power generation active power interval model are as follows:
Wherein,Expression eventFor genuine probability, εPVIt indicates
The parameter of the probability correlation in interval model, P are in the power output of photovoltaic generating systemPVIndicate photovoltaic power generation active power,WithThe respectively lower and upper limit of photovoltaic power generation active power.
For determining εPV, due to parameterWithIt is difficult definite by the probability distribution of Intensity of the sunlight
It obtains, therefore, photovoltaic hair can be obtained by the method for Monte Carlo simulation according to Intensity of the sunlight probability density function
The interval model of electric system output power.
(5) photovoltaic power generation reactive power interval model
It is identical as wind-powered electricity generation, it is generally recognized that the power factor of grid-connected photovoltaic system be it is determining, then pass through photovoltaic power generation
Active power interval model [Pw down, Pw up] its reactive power interval model [Q can be calculatedw down, Qw up]。
Specifically, photovoltaic power generation reactive power interval model are as follows:
Wherein, cos θPVFor the power factor of photovoltaic generating system,WithThe respectively idle function of photovoltaic power generation
The lower and upper limit of rate.
Specifically, negative rules model, wind-power electricity generation active power interval model, wind-power electricity generation reactive power section
The section of model, photovoltaic power generation active power interval model and photovoltaic power generation reactive power interval model, which can be used, is pushed forward back substitution
Tidal current computing method obtains.
Specifically, power uncertainty interval computation formula are as follows:
Wherein, { x } and { y } indicates two set.Interval computation can be replaced with bound calculating, i.e.,x, y,
Interval computation is applied to power distribution network and is pushed forward back substitution Load flow calculation in the embodiment of the present invention.Wherein, iteration is divided into two steps:
1) it is pushed forward process.Start to calculate to head end from the node farthest apart from power supply point, if whole network voltage is all specified electricity
Pressure successively calculates the power distribution of each section of route against the direction of power transmission, does not calculate voltage landing.Firstly, using following formula
Calculate each node Injection Current:
In formula: Ija、Ijb、IjcIndicate that node j respectively mutually injects electric sea, Sja、Sjb、SjcIndicate each phase load power of node j,
Uja、Ujb、UjcIndicate each phase voltage of node j, Yja、Yjb、YjcIndicate node j respectively relatively admittance, k indicates the number of iterations.
Then, calculate branch current: since feeder terminal, direction against the grain, according to KCL, (kirchhoff electric current is fixed
Rule) each branch current is calculated, and acquire root node electric current:
In formula: Ija、Ijb、IjcIndicate that each phase Injection Current of node j, M indicate all lower layer's branch being connected directly with node j
The set on road, Ima、Imb、ImcIndicate the electric current of each lower layer's branch m.
2) back substitution process.It is distributed using the power that given beginning voltage and the first step acquire, since power supply point, along
Power direction of transfer is calculated the voltage landing of each section of route by beginning terminad paragraph by paragraph, finds out each node voltage, but at this moment no longer
Recalculate power loss.Since feeder line head end, along direction of tide, by known root node voltage and the three-phase electricity acquired
Stream, according to each phase voltage of KVL (Kirchhoff's second law) calculate node j:
In formula: Uia、Uib、UicIndicate each phase voltage of node i.Indicate nodal impedance matrix, Ila、Ilb、
IlcIndicate the electric current of each root node l.
By be pushed forward above with two processes of back substitution, just complete the iterative calculation of a wheel.Next round is pushed forward in iteration, meter
The node voltage that can be acquired using last round of back substitution process when calculating power loss.It is pushed forward in iteration, is acquired by network voltage every time
Trend distribution；Be distributed in back substitution iteration by power and calculate voltage's distribiuting, such iteration several times, until the node of front and back iteration twice
The modulus value of voltage difference meets required precision, can stop iteration.
Sometimes for comparing superiority and inferiority between section.Two sections [E] and [F] are taken, when [E] and [F] does not have overlapping region,
Bound, which need to only be compared, can distinguish the two superiority and inferiority.For example, [0.1,0.2] and [0.5,0.8] is compared, the mobility scale of the latter
Interior minimum value 0.5 is greater than the maximum value 0.2 in the former mobility scale, and therefore, the latter is larger.
When [E] and [F] is there are when overlapping region, following comparative approach is needed: assuming that from the section [E] of Normal Distribution
A stochastic variable ξ is taken in [F] respectivelyE, ξF.Remember ξE> ξFFor event P, it is excellent that probability of happening just represents the value from [E]
In the probability of [F], i.e., [E] is better than [F].
For example, [0.3,0.8] and [0.4,0.7] compares size, sample 100 times.It samples, is taken out in the two sections every time
Carry out a number than size, remember that the probability for being greater than the number taken out in second section from the number taken out in first section is P,
If P > 0.5, then it is assumed that first section is big, on the contrary then anti-.
Specifically, in step S20:
(1) the first object majorized function of active power loss expense are as follows:
f1(x)=Ptlossd。
Wherein, f1(x) active power loss expense is indicated.D is unit electricity price.
And
Wherein, PtlossIndicate network total losses.PLoss, kAnd QLoss, kActive loss and idle damage before respectively node k
Consumption.UkFor node voltage.PkAnd QkNot Wei node active outflow power and idle outflow power.RkIndicate the electricity of downstream leg
Resistance.N indicates number of nodes.
The power relation of two adjacent nodes are as follows:
Wherein, PL, kAnd QL, kRespectively node active injection power and idle injecting power.
Trend is calculated using forward-backward sweep method (FBSM) in the present invention, wherein being pushed forward as follows with the iterative equation of back substitution process
Formula:
Wherein,WithRespectively indicate the node Injection Current and node voltage in kth time iteration.WithPoint
It Biao Shi not branch current and voltage vector in kth time iteration.Y and Z is respectively admittance and impedance matrix.F expression is sweared with voltage
Amount is the function of independent variable, represents the node Injection Current as caused by constant load.A expression contains related to by calculate node
The incidence matrix submatrix of all nodes of connection.
(2) it is expected the first object majorized function of scarce power supply volume are as follows:
Wherein, wherein f2(x) it indicates it is expected to lack power supply volume.PA, iIndicate the average load at node i, TiIndicate annual
Power-off time, andM is the equipment sum at node i.λ is equipment annual failure rate.γ is flat for equipment
The equal repair time.
(3) the first object majorized function of operating cost is switched are as follows:
f3(x)=Nop×q。
Wherein, q is that single switchs operating cost.Switch operating cost contains a variety of expenses.Operation itself will be to switch
It causes to be lost, is allowed to service life decline.Meanwhile the closure of interconnection switch will be so that network instantaneously enters loop-net operation state, around here
The extra charge of generation is also required to be included in operating cost.Therefore, these above-mentioned expenses are quantified as single switch operating cost
Use q.NopFor number of operations.
Specifically, in step S30:
Constraint condition includes: the constraint of power distribution network topological structure, trend Constraints of Equilibrium and voltage deviation constraint.
(1) power distribution network topological structure, which constrains, includes:
g∈G。
Wherein, g is the topological structure after reconstruct.G is the set of all topological structures for meeting constraint condition.
(2) trend Constraints of Equilibrium includes:
Wherein, PiAnd QiThe respectively burden with power and load or burden without work of node i.PDG, iAnd QDG, iRespectively distributed generation resource
Active output and idle output.UiFor the voltage of node i.GijAnd BijThe respectively real and imaginary parts of admittance matrix.δijFor section
The power-factor angle of point i.
(3) voltage deviation, which constrains, includes:
Wherein,WithThe respectively upper and lower bound of the voltage of node i.For the maximum outflow electric current of node i.For the active output of maximum of the distributed generation resource of node i.
Specifically, in step S40:
The second objective optimization function is obtained by following mode:
Above-mentioned power distribution network topological constraints differentiate that memory guarantees by the connectivity in following step S603.Above-mentioned tide
Mobile equilibrium constraint is guaranteed by Load flow calculation.Above-mentioned voltage deviation, the constraint of branch power limit pass through penalty function amendment
Model for Multi-Objective Optimization.Therefore, it by peer-to-peer and the inequality constraints additional penalty factor and is integrated into objective function, makes original
This optimization problem with Prescribed Properties can be converted into unconstrained optimization problem, as follows:
After the section numerical value for considering the interval model of the load and distributed energy in step S10, the second objective optimization letter
Number is as follows:
Wherein, w1、w2And w3It respectively indicates the first object majorized function of active power loss expense, it is expected to lack the of power supply volume
The weight factor of one objective optimization function and the first object majorized function of switch operating cost.w4And w5Respectively indicate voltage and
The penalty factor of restriction of current item.WithU iRespectively indicate the bound of node i voltage range value.Indicate node i electric current section
The upper bound of value.
Wherein, in embodiment, PLoss, k~(k+1)The active power loss value on the weary road between node k and node k+1；N is indicated
Number of nodes in power distribution network network, n indicate the number of nodes in power distribution network network；According to sin θw 2+cosθw 2=1, therefore,Similarly, cos θPVFor the power factor of photovoltaic generating system, then
B is correction factor, and τ is year constant (taking 8750 hours).
Specifically, as shown in figure 4, step S60 is comprised the following steps that
Step S601: construction harmony objective function initializes harmony memory and relevant parameter.
Wherein, relevant parameter includes: the number of iterations, musical instrument (i.e. switch) number, harmony data base HM, harmony data base guarantor
Stay probability HMCR, data base disturbance probability P AR.The number of iterations can be set according to operand and actual demand.
The harmony objective function is the second objective optimization function.
The harmony memory of the initialization includes: at least one harmony.Each harmony all includes harmony solution vector.Each harmony
Harmony solution vector indicate that a kind of node of preset power distribution network switches on-off initial value (i.e. in advance to musical instrument assignment).
For example, disconnecting initial value is 0, conducting initial value is 1.
Step S602: new harmony is created.
Wherein, the corresponding harmony solution vector of the new harmony is the power distribution network indicated with the harmony solution vector in harmony memory
The node for switching on-off the different power distribution network of initial value of node switches on-off initial value.
Step S603: the corresponding harmony solution vector of new harmony is subjected to topological structure verifying.
It is verified by topological structure, it may be determined that the switch of the node for the power distribution network that newly the corresponding harmony solution vector of harmony indicates
On-off initial value so that power distribution network is met the constraint condition of radiativity and connectivity.
Topological structure verifying are as follows: whether the network that detection harmony generates contains looped network or isolated island.Due to above-mentioned harmony
The particle generating process of searching algorithm determines that network generated can not contain looped network, and topological analysis only needs to detect network
In whether contain isolated island.
The network generated to one by harmonic search algorithm remembers that its node incidence matrix is B.The every a line of the matrix all represents
The connection relationship of one node and other nodes: 1 representative is connected, and 0 representative is not attached to.The summation of each row (column) element is referred to as should
The Connected degree of node indicates the tightness degree that the node is connected with other nodes.The degree of isolated node is 0, network first and last node
Degree be 1, the degree of other nodes is not less than 2.
By taking the system shown in Fig. 5 as an example, the topology analyzing method of the embodiment of the present invention is briefly described.Normal condition
Under Fig. 5 (a) system node incidence matrix it is as follows:
The topological analysis step of the embodiment of the present invention are as follows:
1) in calculating matrix B all nodes Connected degree.If the node that degree of having is 0, illustrates that network contains isolated node,
It is i.e. infeasible, it can not be verified by topological structure.
If 2) node that network is 0 without degree, deleting the network moderate is 1 and the largest number of node, that is, deleting should
Corresponding row and column is put, then return step 1), until system is only left two nodes.
If 3) degree of remaining two nodes of system is 1, illustrate that the network is feasible, Ke Yitong without isolated island
Cross topological structure verifying；Otherwise infeasible, it cannot be verified by topological structure.
Each circular matrix B can reduce single order.Such as after 4 circulations, matrix B will be reduced to shape shown in following formula
Formula:
In matrix B after reduction, spending for 1 row (column) is 1,5 and 7, the node 1,5 and 7 in corresponding original system.Cause
This, step 2) is only to reduce the scale of original system, does not destroy its topological property.As the process continues, finally
Remaining node 1 and 2, their degree are 1 by matrix B.This illustrates that original system has passed through validating topology, that is, is free of isolated island.
If carrying out step S604 by verifying；Otherwise step S602 is carried out.
Step S604: using tidal current computing method, obtains the solution of the corresponding harmony objective function of the new harmony.
Step S605: judge whether the solution of the harmony objective function of new harmony is better than the worst harmony pair in harmony memory
The solution for the harmony objective function answered.
If being better than, step S606 is carried out；Otherwise, step S602 is carried out.
Step S606: new harmony is substituted into the worst harmony in harmony memory.
Step S607: judge whether to reach maximum number of iterations.
If reaching maximum number of iterations, step S608 is carried out；Otherwise step S602 is carried out, and the number of iterations is increased by one
It is secondary.
Step S608: output harmony memory.
Specifically, in step S70:
The harmony memory of output includes the solution of at least one the second objective optimization function.In these solutions, obtain the smallest
Solution.If the smallest solution has multiple, one of minimal solution can be selected according to the actual situation.Match using minimal solution is corresponding
Power distribution network is reconstructed in the initial value that switches on-off of the node of power grid.
To sum up, the reconstruction method of power distribution network of the embodiment of the present invention has comprehensively considered the random fluctuation of load, distributed energy
Bring is uncertain, has high consistency, accuracy with higher with actual state；Devise the more mesh of power distribution network reconfiguration
Model, meter and operation network loss, power supply reliability, switch operating cost are marked, optimization aim covering comprehensively, is conducive to optimize distribution
Net supply path improves power distribution network operational efficiency, and realizing reduces network loss, eliminate overload, balanced load, improve quality of voltage, mention
The purpose of rising power supply reliability and economic benefit；The consideration factor in reconstruct decision process has directly been reacted, there is stronger reality
The property used；Based on the power distribution network reconfiguration mould for solving multiple target, the nonlinear calculating containing bounded-but-unknown uncertainty using harmonic search algorithm
Type, solution is functional, and convergence is fast, precision is high.
The embodiment of the present invention also provides a kind of power distribution network reconfiguration system.As shown in fig. 6, the power distribution network reconfiguration system includes:
First establishes module 601, for establishing the interval model of load and distributed energy.
Second establishes module 602, for establishing active power loss expense respectively, it is expected to lack power supply volume and switching operating cost
First object majorized function.
Determining module 603, for determining the constraint condition of first object majorized function.
Module 604 is normalized, for according to the interval model of load and distributed energy, constraint condition and per target one by one
First object majorized function is normalized in the weight of majorized function, obtains the second objective optimization function.
Setting module 605, the node for setting power distribution network switch on-off initial value.
Module 606 is solved, for solving the second objective optimization function using harmonic search algorithm.
Reconstructed module 607, for opening for the node for solving corresponding power distribution network according to the smallest second objective optimization function
The on-off initial value of pass carries out power distribution network reconfiguration.
To sum up, the power distribution network reconfiguration system of the embodiment of the present invention has comprehensively considered the random fluctuation of load, distributed energy
Bring is uncertain, has high consistency, accuracy with higher with actual state；Devise the more mesh of power distribution network reconfiguration
Model, meter and operation network loss, power supply reliability, switch operating cost are marked, optimization aim covering comprehensively, is conducive to optimize distribution
Net supply path improves power distribution network operational efficiency, and realizing reduces network loss, eliminate overload, balanced load, improve quality of voltage, mention
The purpose of rising power supply reliability and economic benefit；The consideration factor in reconstruct decision process has directly been reacted, there is stronger reality
The property used；Based on the power distribution network reconfiguration mould for solving multiple target, the nonlinear calculating containing bounded-but-unknown uncertainty using harmonic search algorithm
Type, solution is functional, and convergence is fast, precision is high.
With a specific application example, technical scheme is described further below.
As shown in fig. 7, being 33 meshed network of IEEE containing distributed generation resource, wherein DG1~DG3 uses photovoltaic power generation,
DG4~DG5 uses wind turbine power generation.Section fluctuation is [0.3,0.9], and distributed energy is limited to [0.3P up and downrated,
0.9Prated], i.e., the distributed photovoltaic of DG1-DG5, wind-powered electricity generation waving interval be 30%-90% power-handling capability.Load fluctuation
Consider 10% fluctuation, therefore, section is [0.9PL, 1.1PL], i.e., what the waving interval of load obtained is the volume of 90%-110%
Definite value.33 meshed network of IEEE is a 12.66kV Standard Ratio power distribution network, is opened comprising 33 interconnection switches and 5 segmentations
It closes.As shown in figure 8, being the reconstruction method of power distribution network using the embodiment of the present invention to showing after IEEE33 node power distribution network reconfiguration
It is intended to.Dotted line in Fig. 7 indicates interconnection, there is the connection of closure between two load bus, also has normally opened route to connect
It connects, the route of normal open switch can choose closure if necessary, change supply path.It as shown in table 1, is relevant device reliability
Parameter.It as shown in table 2, is the Comparative result before reconstruct and after reconstruct.
1 relevant device dependability parameter of table
| Device type | λ×r(h/year) |
| Overhead line | [0.282,0.358] |
| Breaker | [0.0418,0.0508] |
| Switch | [0.036,0.044] |
According to different device types, the dependability parameter of different λ × r, and then available annual can be obtained
Power-off time Ti, and according to TiIt can be calculated
IEEE 33 node system of the table 2 containing distributed energy reconstructs forward and backward Comparative result
From table 1, it can be seen that reconstruct before network loss be [108.85,266.06], it is expected that lack power supply volume be [2438.5,
2880.3], minimum voltage per unit value is [0.8718,0.8178], is significantly greater than network loss [75.41,161.33] after reconstructing, phase
It hopes and lacks power supply volume [2425.1,2864.5], minimum voltage per unit value [0.8282,0.8536], to demonstrate weight of the invention
The validity of structure method.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (9)
1. a kind of reconstruction method of power distribution network characterized by comprising
Establish the interval model of load and distributed energy；
Active power loss expense is established respectively, it is expected to lack power supply volume and is switched the first object majorized function of operating cost；
Determine the constraint condition of the first object majorized function；
According to the interval model of the load and distributed energy, the constraint condition and each first object majorized function
Weight, the first object majorized function is normalized, the second objective optimization function is obtained；
Set the node of the power distribution network switches on-off initial value；
The second objective optimization function is solved using harmonic search algorithm；
Initial value is switched on-off according to the node of the corresponding power distribution network of solution of the smallest second objective optimization function,
Carry out power distribution network reconfiguration；
The first object majorized function of the active power loss expense are as follows: f1(x)=PtlossD, wherein f1(x) indicate that active power loss takes
With, d is unit electricity price, PtlossIndicate that network always damages
Consumption, PLoss, kAnd QLoss, kActive loss and reactive loss before respectively node k, UkFor node voltage, PkAnd QkIt not Wei node
Active outflow power and idle outflow power；RkIndicate that the resistance of downstream leg, N indicate number of nodes；The function of two adjacent nodes
Rate relationship are as follows:Wherein, PL, kAnd QL, kRespectively node active injection power and idle note
Enter power；
The first object majorized function for it is expected to lack power supply volume are as follows:Wherein, f2(x) indicate that expectation lacks
Power supply volume, PA, iIndicate the average load at node i, TiIndicate annual power-off time, andM is node i
The equipment sum at place, λ are equipment annual failure rate, and γ is equipment mean time to overhaul；
The first object majorized function of the switch operating cost are as follows: f3(x)=Nop× q, wherein f3(x) switch operating cost is indicated
With q is that single switchs operating cost, NopFor number of operations；
PLoss, k~(k+1)Between node K and node k+1 road active power loss value；N indicates the number of nodes in power distribution network network
Amount.
2. the method according to claim 1, wherein the interval model of the load and distributed energy includes:
Negative rules model, wind-power electricity generation active power interval model, wind-power electricity generation reactive power interval model, photovoltaic power generation have
Function power interval model and photovoltaic power generation reactive power interval model.
3. according to the method described in claim 2, it is characterized in that, the negative rules model, wind-power electricity generation wattful power
Rate interval model, wind-power electricity generation reactive power interval model, photovoltaic power generation active power interval model and the idle function of photovoltaic power generation
The section of rate interval model is obtained using back substitution tidal current computing method is pushed forward.
4. according to the method described in claim 2, it is characterized in that,
The negative rules model are as follows:Wherein, [S] indicates induction conductivity
The waving interval of practical complex power S, [P] indicate that the waving interval of the practical active-power P of induction conductivity, [Q] indicate induction
The waving interval of the practical reactive power of motor, Indicate the power of induction conductivity
FactorWaving interval；
The wind-power electricity generation active power interval model are as follows:Wherein,Expression eventFor genuine probability, εwExpression goes out with wind-driven generator
Power is in the parameter of the probability correlation in interval model, PWIndicate wind-power electricity generation active power, Pw downAnd Pw upRespectively wind-force is sent out
The lower and upper limit of electric active power；
The wind-power electricity generation reactive power interval model are as follows:Wherein, cos θwFor wind
The power factor of power generator；Qw downAnd Qw upThe respectively lower limit of wind-power electricity generation reactive power
And the upper limit；
The photovoltaic power generation active power interval model are as follows:Wherein,Expression eventFor genuine probability, εPVExpression and photovoltaic generating system
Probability correlation of the power output in the interval model parameter, PPVIndicate photovoltaic power generation active power,WithRespectively
The lower and upper limit of photovoltaic power generation active power；
The photovoltaic power generation reactive power interval model are as follows:Wherein, cos θPVFor light
The power factor of photovoltaic generating system； WithRespectively photovoltaic power generation reactive power
Lower and upper limit.
5. the method according to claim 1, wherein the constraint condition include: power distribution network topological structure constraint,
Trend Constraints of Equilibrium and voltage deviation constraint.
6. according to the method described in claim 5, it is characterized in that,
The power distribution network topological structure constraint includes: g ∈ G, wherein g is the topological structure after reconstruct, and G is the constraint of all satisfactions
The set of the topological structure of condition；
The trend Constraints of Equilibrium includes:Wherein, PiAnd QiRespectively
The burden with power and load or burden without work of node i, PDG, iAnd QDG, iThe respectively active output and idle output of distributed generation resource, UiFor
The voltage of node i, GijAnd BijThe respectively real and imaginary parts of admittance matrix, δijFor the power-factor angle of node i；
The voltage deviation constrainsWherein,WithRespectively node
The upper and lower bound of the voltage of i,Maximum for node i flows out electric current,Maximum for the distributed generation resource of node i has
Function output, n indicate the number of nodes in power distribution network network.
7. the method according to claim 1, wherein the second objective optimization function are as follows:
Wherein, w1、w2And w3It respectively indicates the first object majorized function of active power loss expense, it is expected to lack the first mesh of power supply volume
It marks majorized function and switchs the weight factor of the first object majorized function of operating cost, w4And w5Respectively indicate voltage and current
The penalty factor of bound term,WithU iThe bound of node i voltage range value is respectively indicated,Indicate node i electric current interval value
The upper limit；N indicates that the number of nodes in power distribution network network, b are correction factor, and τ is year constant.
8. the method according to claim 1, wherein described solve second target using harmonic search algorithm
The step of majorized function, comprising:
Harmony objective function is constructed, harmony memory and relevant parameter are initialized；Wherein, the harmony objective function is the second target
Majorized function；The harmony memory includes: at least one harmony, and each harmony includes harmony solution vector, it is each described and
What the harmony solution vector of sound indicated a kind of node of preset power distribution network switches on-off initial value；
Create new harmony；
The corresponding harmony solution vector of the new harmony is subjected to topological structure verifying；
If using tidal current computing method by verifying, the solution of the corresponding harmony objective function of the new harmony is obtained；Otherwise it returns
The step of back into the row creation new harmony；
The worst harmony for judging whether the solution of the harmony objective function of the new harmony is better than in the harmony memory is corresponding
The solution of harmony objective function；
If being better than, the new harmony is substituted into the worst harmony in the harmony memory；Otherwise it is new to come back for the creation
The step of harmony；
The new harmony is substituted into the worst harmony in the harmony memory；
Judge whether to reach maximum number of iterations；
If reaching maximum number of iterations, harmony memory is exported；Otherwise the step of coming back for the creation new harmony, and will be repeatedly
Generation number increases primary.
9. a kind of power distribution network reconfiguration system characterized by comprising
First establishes module, for establishing the interval model of load and distributed energy；
Second establishes module, for establishing active power loss expense respectively, it is expected to lack power supply volume and switching the first mesh of operating cost
Mark majorized function；
The first object majorized function of the active power loss expense are as follows: f1(x)=PtlossD, wherein f1(x) indicate that active power loss takes
With, d is unit electricity price, PtlossIndicate that network always damages
Consumption, PLoss, kAnd QLoss, kActive loss and reactive loss before respectively node k, UkFor node voltage, PkAnd QkIt not Wei node
Active outflow power and idle outflow power；RkIndicate that the resistance of downstream leg, N indicate number of nodes；The function of two adjacent nodes
Rate relationship are as follows:Wherein, PL, kAnd QL, kRespectively node active injection power and idle note
Enter power；
The first object majorized function for it is expected to lack power supply volume are as follows:Wherein, f2(x) indicate that expectation lacks
Power supply volume, PA, iIndicate the average load at node i, TiIndicate annual power-off time, andM is node i
The equipment sum at place, λ are equipment annual failure rate, and γ is equipment mean time to overhaul；
The first object majorized function of the switch operating cost are as follows: f3(x)=Nop× q, wherein f3(x) switch operating cost is indicated
With q is that single switchs operating cost, NopFor number of operations；
Determining module, for determining the constraint condition of the first object majorized function；
Module is normalized, for according to the interval model of the load and distributed energy, the constraint condition and each described
The first object majorized function is normalized in the weight of one objective optimization function, obtains the second objective optimization letter
Number；
Setting module, the node for setting the power distribution network switch on-off initial value；
Module is solved, for solving the second objective optimization function using harmonic search algorithm；
Reconstructed module, for opening for the node for solving the corresponding power distribution network according to the smallest second objective optimization function
The on-off initial value of pass carries out power distribution network reconfiguration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710671270.9A CN107508280B (en) | 2017-08-08 | 2017-08-08 | A kind of reconstruction method of power distribution network and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710671270.9A CN107508280B (en) | 2017-08-08 | 2017-08-08 | A kind of reconstruction method of power distribution network and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107508280A CN107508280A (en) | 2017-12-22 |
| CN107508280B true CN107508280B (en) | 2019-10-08 |
Family
ID=60689151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710671270.9A Active CN107508280B (en) | 2017-08-08 | 2017-08-08 | A kind of reconstruction method of power distribution network and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107508280B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108183475B (en) * | 2017-12-27 | 2021-02-09 | 国网河南省电力公司电力科学研究院 | Low-voltage distribution network topology reconstruction method and device |
| CN110348048B (en) * | 2019-05-31 | 2022-09-30 | 国网河南省电力公司郑州供电公司 | Power distribution network optimization reconstruction method based on consideration of heat island effect load prediction |
| CN110610059A (en) * | 2019-09-26 | 2019-12-24 | 东南大学 | Method suitable for topology reduction of single-power-supply tree-shaped power distribution network |
| CN110794220A (en) * | 2019-10-28 | 2020-02-14 | 云南电网有限责任公司 | Method and device for evaluating oscillation risk of wind power plant convergence system |
| CN111277004B (en) * | 2020-02-11 | 2020-09-29 | 天津大学 | Power distribution network source-network-load two-stage multi-target control method and system |
| CN111614081A (en) * | 2020-05-22 | 2020-09-01 | 国网宁夏电力有限公司电力科学研究院 | Urban distribution network loss reduction optimization method based on load prediction model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069704A (en) * | 2015-08-14 | 2015-11-18 | 中国电力科学研究院 | Quick traversing power distribution network reconstruction method for improving distributed power supply permeability |
| CN105117517A (en) * | 2015-07-28 | 2015-12-02 | 中国电力科学研究院 | Improved particle swarm algorithm based distribution network reconfiguration method |
| CN106026187A (en) * | 2016-08-10 | 2016-10-12 | 广东工业大学 | Distributed-power-source-containing power distribution network reconfiguration method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2014033892A1 (en) * | 2012-08-31 | 2016-08-08 | 株式会社日立製作所 | Power accommodation route creation method and power accommodation route creation device |
-
2017
- 2017-08-08 CN CN201710671270.9A patent/CN107508280B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117517A (en) * | 2015-07-28 | 2015-12-02 | 中国电力科学研究院 | Improved particle swarm algorithm based distribution network reconfiguration method |
| CN105069704A (en) * | 2015-08-14 | 2015-11-18 | 中国电力科学研究院 | Quick traversing power distribution network reconstruction method for improving distributed power supply permeability |
| CN106026187A (en) * | 2016-08-10 | 2016-10-12 | 广东工业大学 | Distributed-power-source-containing power distribution network reconfiguration method and system |
Non-Patent Citations (4)
| Title |
|---|
| An Improved Multi-Objective Harmony Search for Optimal Placement of DGs in Distribution Systems;Komail Nekooei, Malihe M. Farsangi, Hossein Nezamabadi-Pour, and;《IEEE Transactions on Smart Grid》;20130208;557-566 * |
| Load model effect assessment on optimal distributed generation (DG) sizing and allocation using improved harmony search algorithm;H. Nasiraghdam , S. Jadid;《2012 Conference on Smart Electric Grids Technology》;20121219;210-218 * |
| 基于改进和声搜索算法的配电网重构研究;黄哲;《中国优秀硕士学位论文全文数据库工程科技II辑》;20140315;正文第16页第2段-第21页第1段，第22页第4段-32页第2段，第35页第2段-第47页第2段 * |
| 改进和声搜索算法用于配电网重构;梁杉;《电力系统及其自动化学报》;20170331;第29卷(第3期);90-95 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107508280A (en) | 2017-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107508280B (en) | A kind of reconstruction method of power distribution network and system | |
| Uniyal et al. | Optimal network reconfiguration and DG allocation using adaptive modified whale optimization algorithm considering probabilistic load flow | |
| Syahputra et al. | Performance improvement of radial distribution network with distributed generation integration using extended particle swarm optimization algorithm | |
| CN103353979B (en) | The Optimizing Site Selection of a kind of distributed power source and constant volume method | |
| Liang et al. | Probability-driven transmission expansion planning with high-penetration renewable power generation: A case study in northwestern China | |
| CN107069814B (en) | The Fuzzy Chance Constrained Programming method and system that distribution distributed generation resource capacity is layouted | |
| Rezaeian-Marjani et al. | Optimal allocation of D-STATCOM in distribution networks including correlated renewable energy sources | |
| CN109274095A (en) | Low-voltage distribution network users topology estimation method and system based on mutual information | |
| CN108306303A (en) | A kind of consideration load growth and new energy are contributed random voltage stability assessment method | |
| CN106532710B (en) | The micro-capacitance sensor tide optimization method of meter and Voltage Stability Constraints | |
| Nayak | Optimal Feeder Reconfiguration of Distribution System with Distributed Generation Units using HC-ACO. | |
| CN105069236B (en) | Consider the broad sense load joint probability modeling method of wind power plant node space correlation | |
| CN107492908A (en) | A kind of feeder line Method for optimized planning based on distributed power source access | |
| CN108599239A (en) | A kind of droop control type isolated island micro-capacitance sensor voltage quality probability evaluation method of failure | |
| CN108281959A (en) | A kind of bulk transmission grid optimization method of high proportion type power system of renewable energy | |
| Zhang et al. | Data-driven sizing planning of renewable distributed generation in distribution networks with optimality guarantee | |
| Arabali et al. | Optimum sizing and siting of renewable-energy-based DG units in distribution systems | |
| Tian et al. | Optimal feeder reconfiguration and distributed generation placement for reliability improvement | |
| CN115017854A (en) | Method for calculating maximum allowable capacity of DG (distributed generation) of power distribution network based on multidimensional evaluation index system | |
| CN110311372A (en) | Sub-area division method based on spectral clustering | |
| Merzoug et al. | Optimal placement of wind turbine in a radial distribution network using PSO method | |
| Zhang et al. | Multi–objective Cluster Partition Method for Distribution Network Considering Uncertainties of Distributed Generations and Loads | |
| Yu et al. | Optimization of an offshore oilfield multi-platform interconnected power system structure | |
| Conti et al. | Probabilistic load flow for distribution networks with photovoltaic generators part 1: Theoretical concepts and models | |
| Ali et al. | Reconfiguration with simultaneous DG installation to improve the Voltage Profile in Distribution Network using Harmony Search Algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |