CN104866921A - Power distribution system network reconstruction method based on safety domain - Google Patents

Abstract

The invention discloses a power distribution system network reconstruction method based on a safety domain. The power distribution system network reconstruction method comprises the following steps: according to a power distribution system safety domain, a safety boundary equation, constructing a safety index of the power distribution system safety domain; according to the safety index of the power distribution system safety domain, combining a genetic algorithm so as to acquire a network reconstruction model; carrying out reconstruction on a power distribution system network through the network reconstruction model so as to acquire a final switch state of the whole network. In the method, the improved genetic algorithm is used to solve the network reconstruction model so that individual ''precocity'' can be avoided; a convergence effect is good and the algorithm is stable. By using the method, a new thought is provided for application of a power distribution network safety domain theory; a good guidance effect on optimization operation of a N-1 demanding large-scale connected city power distribution network is realized.

Description

A kind of feeder reconfiguration methods in distribution system based on security domain

Technical field

The present invention relates to distribution system optimizing operation field, particularly relate to a kind of feeder reconfiguration methods in distribution system based on security domain.

Background technology

Power distribution network network reconfiguration (Distribution Network Reconfiguration, DNR) refers to by switching manipulation to adjust network structure, is the important means improving distribution system safety in operation and economy.

DNR mathematics is the combinatorial optimization problem of a multiple constraint in essence.In the optimization aim and constraint of model, the target of traditional DNR mainly comprises: reduce network loss, balanced load, raising quality of voltage, and constraint is then for normal condition runs the irradiation structure constraint of constraint and network.In algorithm, because the combinatorial optimization problem of multiple constraint solves difficulty, large quantity research is devoted to the accuracy and runtime improving derivation algorithm, to meet the practical application request of network reconfiguration; Be particularly the intelligent algorithm of representative with genetic algorithm, achieve good effect.

What require power supply reliability along with user improves constantly, safety, reliable target or the constraint becoming DNR gradually.The essence of reliability be system each time discontinuity surface the integration of security, index is more and system ripe, the convenient objective function being used for network reconfiguration, such as: the average frequency of power cut of system (SAIFI) etc.DNR based on reliability is less demanding to computing velocity, is more applicable to being applied in Electric Power Network Planning field; And the running status of security discontinuity surface when more attention location system is some, the system-oriented realtime scheduling and monitoring of the DNR based on security, requires higher to computing velocity.

For a long time, DNR research based on security is less, mainly due to: 1) classical secrecy analysis adopts N-1 pointwise simulation method, when power distribution network is larger, point by point method computing velocity is slower, if directly N-1 analytic process is added DNR model, then can increase the difficulty that network reconfiguration calculates further; 2) system after existing safety indexes mostly is the generation of N-1 fault loses load, and when normally running, the target of DNR is prevented or optimal control electrical network, needs the index that can reflect current electric grid security information.

Summary of the invention

The invention provides a kind of feeder reconfiguration methods in distribution system based on security domain, security domain theory is applied in power distribution system network reconstruct by the present invention, for towards the power distribution network Real-Time Scheduling of security and optimal control supplying method, described below:

Based on a feeder reconfiguration methods in distribution system for security domain, described feeder reconfiguration methods in distribution system comprises the following steps:

According to the safety indexes of distribution system security region, secure border equation structure distribution system security region;

Safety indexes according to described distribution system security region obtains network reconfiguration model, is reconstructed power distribution system network by described network reconfiguration model, obtains the on off state that the whole network is final.

The step of the described safety indexes according to distribution system security region, secure border equation structure distribution system security region is specially:

Obtain wall scroll feeder line safe distance;

By described wall scroll feeder line safe distance obtain respectively average feeder line safe distance, and minimum feeder line safe distance.

The described safety indexes according to described distribution system security region, the step obtaining network reconfiguration model in conjunction with genetic algorithm are specially:

Using all switch sets on ring connection as a gene block, the gene block composition chromosome of each bar ring connection;

Using the network loss number percent of power distribution network as fitness function, by described wall scroll feeder line safe distance the value of effect of constraint value fitness function, chromosome is screened;

Produce new individuality by the operation of restriction crossover and mutation, make the topological structure of electric after network reconfiguration keep radiation operation characteristic simultaneously;

Convergence of algorithm speed is improved by optimal individual conserving method.

The beneficial effect of technical scheme provided by the invention is: DSSR theory is applied to power distribution system network reconstruct by this method first, establishes the network reconfiguration model taking into account DSSR safety indexes, is the theoretical trial first in network reconfiguration field of security domain.

First, DSSR theory can before electrical network actual motion, preferentially provides the N-1 secure border equation that can run working point and is stored in the computing machine of dispatching center, directly calling, thus eliminate the N-1 node-by-node algorithm process of complicated and time consumption in order to during observation.Therefore, this method fundamentally can improve the computing velocity of network reconfiguration, the demand making it meet power distribution network to monitor in real time and run.

Secondly, model take system losses as objective function, and safe distance SD etc. is constraint condition.Wherein utilize vectorial SD _mincan distinguish the margin of safety constraint arranging each bar feeder line, this is conducive to arranging different reconstruction strategy flexibly, to meet different economy and the demand of security.Compared to the reconstructing method of traditional simple consideration economy, this method can carry out security constraint for important feeder line, the margin of safety of the important feeder line of maximized raising, reduces the whole network network loss preferably simultaneously.

Finally, this method adopts Revised genetic algorithum to solve network reconfiguration model, can avoid individuality " precocity ", and convergence effect is better, and algorithm is comparatively stable.This method provides new approaches for the application of power distribution network security domain theory, for the optimizing operation (network reconfiguration) that N-1 requires strict extensive interconnected urban power distribution network, especially has good directive function.

Accompanying drawing explanation

Fig. 1 is Revised genetic algorithum network reconfiguration process flow diagram;

Fig. 2 is 11 node interconnect power distribution network figure;

Fig. 3 is example configuration of power network;

Fig. 4 is that the safe distance of different constraints policy is with network loss contrast schematic diagram;

Fig. 5 is the Swarm Evolution situation under security constraint strategy 1;

Fig. 6 is a kind of process flow diagram of the feeder reconfiguration methods in distribution system based on security domain.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly, below embodiment of the present invention is described further in detail.

Distribution system security region (Distribution System Security Region, DSSR) based on N-1 criterion is the new tool analyzing security of distribution network, can effectively solve the problem ^[1].First, DSSR before electrical network actual motion, directly can provide the N-1 secure border equation of system, and it is stored in advance in the computing machine of dispatching center according to electric network composition and configuration, need can directly call during observation, eliminate the N-1 computation process of complicated and time consumption.Secondly, according to the relative position of working point to secure border, can to the safety indexes made new advances---safe distance (Security Distance, SD), this index can the security information of quantificational description system present operating point, provides the foundation of prevention and optimal control.

Embodiment 1

Based on a feeder reconfiguration methods in distribution system for security domain, this feeder reconfiguration methods in distribution system comprises the following steps:

101: according to the safety indexes of distribution system security region, secure border equation structure distribution system security region;

102: the safety indexes according to distribution system security region obtains network reconfiguration model, is reconstructed power distribution system network by network reconfiguration model, obtains the on off state that the whole network is final.

Wherein, be specially according to the step of the safety indexes of distribution system security region, secure border equation structure distribution system security region:

Obtain wall scroll feeder line safe distance;

By described wall scroll feeder line safe distance obtain respectively average feeder line safe distance, and minimum feeder line safe distance.

Wherein, be specially according to the safety indexes of distribution system security region, the step that obtains network reconfiguration model in conjunction with genetic algorithm:

Using all switch sets on ring connection as a gene block, the gene block composition chromosome of each bar ring connection;

Using the network loss number percent of power distribution network as fitness function, by wall scroll feeder line safe distance the value of effect of constraint value fitness function, chromosome is screened;

Produce new individuality by the operation of restriction crossover and mutation, make the topological structure of electric after network reconfiguration keep radiation operation characteristic simultaneously;

Convergence of algorithm speed is improved by optimal individual conserving method.

The computing velocity of network reconfiguration is fundamentally improved, the demand making it meet power distribution network to monitor in real time and run by above-mentioned operation steps.

Embodiment 2

Below in conjunction with concrete test figure, the operating process of a kind of feeder reconfiguration methods in distribution system based on security domain of the description that computing formula is detailed, described below:

201: build distribution system security region and secure border equation thereof;

Distribution system security region DSSR is defined as and meets N-1 safe operation criterion and other various operations constraint (such as: Kirchhoff's law retrains, power flow equation retrains, node voltage amplitude constraint or irradiation structure run constraint, during specific implementation, the embodiment of the present invention does not limit this) the set of all working point W.Wherein, working point W is the n-dimensional vector of all load bus compositions of distribution system, and namely distribution system security region is:

W＝(F ₁,F ₂,...,F _i,...,F _n) ^T

(1)

Ω _DSSR＝{W|f(V,θ)＝W,g(W)＜0}

In formula (1), Ω _dSSRfor the mathematic sign of DSSR; V, θ are amplitude and the phase angle of node voltage.F (V, θ)=W is trend Constraints of Equilibrium; G (W) <0 is safe operation constraint, mainly comprises: the branch current constraint after N-1 fault, main transformer capacity constraint and node voltage constraint.F _irepresent node load; T represents transposition.For common Single-ring network structure distribution network, W can be reduced to and represent with feeder line outlet power.

There are some researches show, the DSSR border strictly taking into account voltage network loss also has the feature of approximately linear, when result accuracy requirement is not high, the secure border simplifying network loss and voltage constraint DSSR only directly can be resolved according to failure reconfiguration path and drawn, the secure border equation that now i-th feeder line is corresponding is designated as B (i):

$B\left(i\right):F_i-\min \{{\mathrm{RF}}_{bi}-F_{bi},R_{bi}-\underset{j\&Element;{\mathrm{\&Phi;}}^{\left(bi\right)}}{\&Sigma;}{F}_j-\underset{k\&NotEqual;i,k\&Element;{\mathrm{\&Theta;}}^{\left(i\right)}}{\&Sigma;}{F}_k\}=0---\left(2\right)$

In formula (2), bi is F _istandby power supply (feeder line) is numbered, RF _bifor this feeder line max cap.; F _bifor the total load under feeder line bi normal condition; R _bifor F _ithe main transformer capacity of standby power supply upstream; Φ ^(bi)indicate the feeder line set from main transformer bi; Θ ⁽ⁱ⁾represent the set of one group of feeder line, these feeder lines and F _icome from same main transformer, and after N-1 fault, its load and F _itransfer to same main transformer.

202: the safety indexes building distribution system security region;

1) wall scroll feeder line safe distance SD (Security Distance) is obtained:

When other feed connection node load powers of distribution system remain unchanged, the peak load undulate quantity SD that i-th feeder line can bear _i, formula (2) is done equivalence transformation and obtains:

${\mathrm{SD}}_i=\min \{{\mathrm{RF}}_{bi}-F_{bi},R_{bi}-\underset{j\&Element;{\mathrm{\&Phi;}}^{\left(bi\right)}}{\&Sigma;}{F}_j-\underset{k\&NotEqual;i,k\&Element;{\mathrm{\&Theta;}}^{\left(i\right)}}{\&Sigma;}{F}_k\}-F_i---\left(3\right)$

After network switching state is determined, SD _ialso uniquely determine.Due to each bar feeder line SD _icalculating independent of one another, therefore utilize SD _ican in the algorithm to different feeder line settings security constraint in various degree.Needed for the feeder line of important load place, margin of safety is often greater than other feeder lines, and the margin of safety of these important feeder lines is designated as SD_IF.Common important load comprises primary user load, essential industry load, hospital, bank server etc., and during specific implementation, the embodiment of the present invention does not limit this.

2) average feeder line safe distance ASD (Average Security Distance) is obtained, for showing the whole network comprehensive safety nargin horizontal safety situation, if electrical network feeder line adds up to n: in full simply connected network electrical network, feeder line sum equals outlet load sum, also equals the dimension of working point W.

$ASD=\frac{1}{n}\underset{i=1}{\overset{n}{\&Sigma;}}{\mathrm{SD}}_i---\left(4\right)$

3) minimum feeder line safe distance (Minimum Security Distance, MSD) is obtained.

As MSD>0, all feeder lines of expression electrical network, main transformer all meet N-1 security; As MSD<0, after the arbitrary main transformer of its absolute value representation or feeder line break down, the peak load that electrical network may be lost.

MSD＝min{SD ₁,SD ₂,...SD _n} (5)

203: the network reconfiguration model building the safety indexes based on distribution system security region;

Distribution system network loss is taked to account for the ratio of total burden with power as objective function:

$\min F=\left(\underset{j=1}{\overset{m}{\&Sigma;}}{k}_j{r}_j\right|I_j{|}^2)/{\mathrm{\&Sigma;}}_n{P}_f^i---\left(6\right)$

Wherein, m is power distribution network circuitry number; r _jfor the branch impedance of jth bar; I _jfor flowing through the load current of jth bar branch road; k _jfor on off state, 0 is off, and 1 is closed. for node i load active power.System losses F is obtained by Load flow calculation, and because the ratio R/X value of distribution line resistance and reactance is comparatively large, algorithm quicksort possibly cannot be restrained, and the embodiment of the present invention takes forward-backward sweep method.

Formula (6) need meet wall scroll feeder line safe distance SD and retrain.Because SD is the index coming from DSSR, the feature modeling process of DSSR has covered the branch current constraint after N-1 fault and main transformer capacity constraint, therefore formula (6) all constraint can be reduced to:

Wherein, V _{i, min}with V _{i, max}represent the bound of node i voltage respectively; G needs after representing network reconfiguration to keep power distribution network radiation to run; SD _{i, min}for the safe distance lower limit that feeder line i need meet.For ease of mark, follow-up with vectorial SD=[SD ₁, SD ₂..., SD _n] represent each bar feeder line safe distance, SD _min=[SD _{1, min}, SD _{2, min}..., SD _{n, min}] represent safe distance lower limit.To SD _minin element get different numerical value and can realize arranging different security constraints to different feeder line.

Adopt Revised genetic algorithum to solve network reconfiguration model, the improved adaptive GA-IAGA flow process that the embodiment of the present invention is taked as shown in Figure 1, is described below the committed step in algorithm.

1) individual UVR exposure;

Adopt binary coding strategy to encode to the on off state on chromosome, 0 expression is opened, and 1 represents closed.In order to ensure radial operation, every bar ring connection only has a switch opened.The embodiment of the present invention is using all switch sets on a ring connection as a gene block, and the gene block of each bar ring connection forms chromosome.As shown in Figure 2.

Simple interconnected power distribution network shown in Fig. 2, has 6 feeder line F ₁f ₆, total number of switches is 8, and therefore chromosome length is 8; Article 6, feeder line forms three simply connected networks (Single-ring network) altogether, and therefore gene block has 3.

If cut-off switch is 2,4,7, then current individual is:

2) fitness function;

The network loss number percent (ratio of network loss and total burden with power) of general power distribution network is between 0.01 ~ 0.08, and its inverse, between 10 ~ 100, does fitness function proper:

$f=\frac{1}{F}---\left(8\right)$

F is the target function value of formula (6), is obtained by Load flow calculation.To the individuality not meeting constraint condition (namely needing to meet wall scroll feeder line safe distance SD), its fitness is reduced to 0, thus this individuality can never be genetic in the next generation go:

F'=f (meeting constraint), f'=0 (not meeting constraint) (9)

Wherein, f is former fitness, and f' is the new fitness considering constraint.Constraint is herein formula (7).

3) genetic manipulation;

Selection operation is the basic step of all genetic algorithms, between algorithms of different and indifference, its objective is all objects providing interlace operation or mutation operation.Regulation interlace operation only occurs between the gene block of same position, does not have an impact to the gene block of other position, thus avoids a large amount of infeasible solution.

Mutation operation is still and occurs in gene block.For to morph operation individuality, the gene block that random selecting morphs, find genic value be 0 gene position, this gene position is become 1, will thereafter a gene (in same gene block) become 0 from 1.

The object of such restriction crossover and mutation working rule is: make the topological structure of electric after network reconfiguration still keep radiation operation characteristic, otherwise will have a large amount of infeasible solutions in newborn individual.

Crossover probability P _mwith variation P _cthe value of probability has obvious impact to algorithm convergence.In order to avoid there is the evolution Premature Convergence i.e. situation of " precocity ", when colony is when comparatively early convergent tendency appears in the stage, (before cross and variation operation) raising P after selection operation _c, reduce P _mto increase new individuality.Initial p _c, P _mvalue is:

P _m＝0.95,P _c＝0.05 (10)

If during iterations t<40 and the maximal value of fitness f (x) continuous 5 generations remain unchanged, then select new P _c', P _m':

P _m'＝0.80,P _c'＝0.15 (11)

4) optimal individual conserving method

After whole individuality procreations of certain generation terminate, need to carry out optimal individual conserving method.T for time, a ^*t () is optimized individual in colony, t+1 for time a ^*t () replaces the individuality that in A (t+1) colony, fitness is minimum.Often all carry out above-mentioned optimal save strategy operation for colony, ensure that the best individuality of fitness is not destroyed, can raising solving speed to a certain degree.Distribution feeder number of switches is huge, and practical application requires very high to the solving speed of network reconfiguration, be therefore necessary to introduce be applicable to have individual conversation strategy most.

If there is the individuality that multiple fitness is minimum, replacing whole may destroy the gene diversity of population, causes algorithm Premature Convergence.The embodiment of the present invention establishes the individual amount participating in conversation strategy to be no more than 20% of population at individual sum.

When group property meets the condition of convergence after some generations evolve, evolve and terminate, export optimum individual, the final plan of map network reconstruct.

The computing velocity of network reconfiguration is fundamentally improved, the demand making it meet power distribution network to monitor in real time and run by above-mentioned operation steps.

Embodiment 3

Below with a power distribution network example, embodiments of the present invention are described.Example takes the IEEE RBTSBUS4:1 transformed) add feeder line quantity and feeder line contact scale; 2) load slightly increases, and embodies the sight that high capacity runs lower electrical network convergence N-1 secure border; 3) adjust feeder line and main transformer capacity, such that N-1 security is feeder line, main transformer retrains jointly; 4) feeder line F7 and F16 is containing important load, needs larger margin of safety.Each section of line length is 0.5km.

Table 1 main transformer data

Table 2 feeder line data

1) secure border calculates;

Example secure border equation is calculated as follows.

$B_{DSSR}=\left\{\begin{array}{c}{F}_1\&le;\min ({\mathrm{RF}}_7-F_7,R_6-F_7-(F_{19}+F_{20)})\\ F_2\&le;\min ({\mathrm{RF}}_6-F_6,R_5-F_6-(F_{17}+F_{18}))\\ F_3\&le;\min ({\mathrm{RF}}_4-F_4,R_3-F_4-(F_{12}+F_{13}))\\ F_4\&le;\min ({\mathrm{RF}}_3-F_3,R_2-F_3-(F_{10}+F_{11}))\\ F_5\&le;\min ({\mathrm{RF}}_{17}-F_{17},R_5-F_{17}-(F_6+F_{18}))\\ F_6\&le;\min ({\mathrm{RF}}_2-F_2,R_1-F_2-(F_1+F_8+F_9))\\ F_7\&le;\min ({\mathrm{RF}}_1-F_1,R_1-F_1-(F_2+F_8+F_9))\\ F_8\&le;\min ({\mathrm{RF}}_{10}-F_{10},R_2-F_{10}-(F_3+F_{11}))\\ F_9\&le;\min ({\mathrm{RF}}_{16}-F_{16},R_4-F_{16}-(F_5+F_{14}+F_{15}))\\ F_{10}\&le;\min ({\mathrm{RF}}_8-F_8,R_1-F_8-(F_1+F_2+F_9))\\ F_{11}\&le;\min ({\mathrm{RF}}_{19}-F_{19},R_6-F_{19}-(F_7+F_{20}))\\ F_{12}\&le;\min ({\mathrm{RF}}_{14}-F_{14},R_4-F_{14}-(F_5+F_{15}+F_{16}))\\ F_{13}\&le;\min ({\mathrm{RF}}_{18}-F_{18},R_5-F_{18}-(F_6+F_{17}))\\ F_{14}\&le;\min ({\mathrm{RF}}_{12}-F_{12},R_3-F_{12}-(F_4+F_{13}))\\ F_{15}\&le;\min ({\mathrm{RF}}_{20}-F_{20},R_6-F_{20}-(F_7+F_{19}))\\ F_{16}\&le;\min ({\mathrm{RF}}_9-F_9,R_1-F_9-(F_1+F_2+F_8))\\ F_{17}\&le;\min ({\mathrm{RF}}_5-F_5,R_4-F_5-(F_{14}+F_{15}+F_{16}))\\ F_{18}\&le;\min ({\mathrm{RF}}_{13}-F_{13},R_3-F_{13}-(F_4+F_{12}))\\ F_{19}\&le;\min ({\mathrm{RF}}_{11}-F_{11},R_2-F_{11}-(F_3+F_{10}))\\ F_{20}\&le;\min ({\mathrm{RF}}_{15}-F_{15},R_4-F_{15}-(F_5+F_{14}+F_{16}))\end{array}\right.$

2) safe distance vector SD _minarrange

Due to SD _minbe the quantizating index describing security constraint, according to index size, the reconstructing method in the embodiment of the present invention can have multiple security constraint strategy.The overall safety constraint strength of strategy 1-4 constantly increases: strategy 1,3 only strengthens security constraint for important feeder line, and strategy 2,4 is security constraints of each feeder line of unified the whole network, refers to table 3 1-3 capable.

Table 3 security constraint strategy 1-4 describes and reconstruct front and back Comparative result (underscore italic is best, and italic is the poorest)

3) operational development genetic algorithm program solves

According to flow process shown in accompanying drawing 1, realize based on MATLAB coding.

4) example reconstruction result is analyzed

The 4-8 of table 1 is capable illustrates safe distance and network loss situation before and after reconstruct, and wherein SD_IF (important feeder line safe distance) is SD ₇and SD ₁₆mean value.In table 3, the variation tendency directly perceived of each index as shown in Figure 4, and after reconstruct, the safe distance of each feeder line as shown in Figure 4.

From table 3 and Fig. 4:

1) before reconstruct, the security of electrical network is better, and important feeder line F7, F16 can bear the load increment of 0.5 and 1.3MVA respectively, but network loss is larger.The reconstruct of 4 kinds of different constraints policies of the embodiment of the present invention all can reduce network loss, and overall system security (ASD) does not reduce.

2) variation tendency of MSD and SD _minthe maximal value (table 1 the 3rd row) of element is consistent, and this shows by adjustment SD _minvalue can improve the security of system bottleneck element.

3) strategy 1 is when obtaining minimum network loss, ensure that the margin of safety (SD_IF) of important feeder line.Because the breakdown loss of important feeder line is larger, improve economy so be also equivalent to the raising of SD_IF, therefore strategy 1 is well suited for for the scene higher to cost-effectiveness requirement; Although strategy 4 network loss are comparatively large, its every safety index is the best, the scene that applicable security requirement is higher.

Fig. 5 gives the situation of change of this method average fitness and optimal fitness function value during evolution.

Because the algorithm flow of Different Strategies is identical, the situation of an exhibition strategy 1 herein.Continuous running optimizatin program 20 times, wherein obtain optimal result (network loss 1.299%) for 11 times, arithmetic result is more stable.Colony forms stable optimum individual when 40 generation, evolves and restrain gradually after 50 generations.Speed of convergence is very fast, the average 355s consuming time of single calculation (CPU2.66GHz, RAM 2GB) when Population is 100, does not have precocious situation.

List of references

[1]Xiao，J.，Gu，W.，Wang，C.，Li，F..Distribution system security region：definition，model and security assessment[J].IET Generation，Transmission&Distribution，2012，6(10)：1029-1035.

It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment, the invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (3)

1. based on a feeder reconfiguration methods in distribution system for security domain, it is characterized in that, described feeder reconfiguration methods in distribution system comprises the following steps:

According to the safety indexes of distribution system security region, secure border equation structure distribution system security region;

According to the safety indexes of described distribution system security region, obtain network reconfiguration model in conjunction with genetic algorithm, by described network reconfiguration model, power distribution system network is reconstructed, obtain the on off state that the whole network is final.

2. a kind of feeder reconfiguration methods in distribution system based on security domain according to claim 1, is characterized in that, the step of the described safety indexes according to distribution system security region, secure border equation structure distribution system security region is specially:

Obtain wall scroll feeder line safe distance;

By described wall scroll feeder line safe distance obtain respectively average feeder line safe distance, and minimum feeder line safe distance.

3. a kind of feeder reconfiguration methods in distribution system based on security domain according to claim 2, is characterized in that, the described safety indexes according to described distribution system security region, the step obtaining network reconfiguration model in conjunction with genetic algorithm are specially:

Using all switch sets on ring connection as a gene block, the gene block composition chromosome of each bar ring connection;

Using the network loss number percent of power distribution network as fitness function, by described wall scroll feeder line safe distance the value of effect of constraint value fitness function, chromosome is screened;

Produce new individuality by the operation of restriction crossover and mutation, make the topological structure of electric after network reconfiguration keep radiation operation characteristic simultaneously;

Convergence of algorithm speed is improved by optimal individual conserving method.