CN104764980A - Positioning method for distribution circuit fault section based on BPSO and GA - Google Patents

Abstract

The invention relates to a positioning method for a distribution circuit fault section based on BPSO and a GA. The mixture of a binary system particle swarm and the genetic algorithm is achieved by a bi-population evolution and information interchange strategy utilized by the method, and a binary system mixed algorithm is formed; two subpopulations are each provided with an individual scale, an evolutionary process of each generation is not affected, after evolution of each generation is completed, information interchange transmission is conducted, an optimal individual is selected, optimizing searches are carried out on the next generation of the two populations respectively until an optimal solution is found out. The probability of immature convergence occurred in a fault positioning process can be reduced by the algorithm, a certain fault tolerance is obtained, and compared with fault section positioning conducted by single binary system particle swarm or the single genetic algorithm, the convergence speed is obviously improved.

Description

A kind of distribution line failure Section Location based on BPSO and GA

Technical field

The present invention relates to a kind of distribution line failure Section Location based on BPSO and GA.

Background technology

Along with increasing the weight of of world energy sources crisis and environmental pollution, people start to pay close attention to cleaner, renewable, the efficient power supply of one---distributed power source (Distributed Generator, DG).Wind turbines obtains extensive grid-connected use as one of common DG.Distribution line is directly connected to customer charge, guarantees that reliability when it runs, security and power supply quality tool are of great significance.But the access of DG makes distribution system become the complicated multiterminal electric power network of to and fro flow of power from single supply radial networks.In addition, the structure of distribution system and trend etc. also change thereupon, and many parameters of power distribution network are provided with dynamic perfromance, and this exerts a certain influence to the fault section location of distribution line.Therefore, the fault section location method of the distribution network based on single supply is radial needs Improvement and perfection because being subject to the impact of DG.

The method being applicable to locate containing the distribution line section of DG at present is mainly divided into two large classes: a class take matrix algorithms as the direct algorithm of Typical Representative, the method is based on graph theory knowledge, pass through fault signature analysis, topological structure in conjunction with power distribution network carries out localization of fault, but the method needs complicated matrix multiple, also to be normalized to prevent erroneous judgement, operand is excessive, processing time is long, comparatively loaded down with trivial details to the process of Complicated Distribution Network, also there is limitation in the unified criterion provided under the localization of fault under solution dual power supply or many power supplys paired running situation, another kind of is based on artificial intelligent type fault location algorithm, mainly comprise neural network, ant group algorithm and genetic algorithm, but use ubiquity " immature oils " phenomenon of intelligent algorithm and slow two contradictory problems of speed of convergence, this two problems of how compromising has become the main direction of studying of innovatory algorithm.

Particle cluster algorithm (Particle Swarm Optimization, PSO) is a kind of new intelligent optimization algorithm that development in recent years is got up, plurality of advantages such as having simply, easily realize, constringency performance is better.Ultimate principle the process finding optimum solution in optimization problem is regarded as the process that bird (particle) finds optimal location in space.During evolution, utilize multiple particle to search optimal location in space to improve the search efficiency of algorithm simultaneously, the search of the every generation of each particle is all upgraded in time to particle rapidity and particle position by certain formula, parameter according to self optimal location and population optimal location, after experienced by the evolution of many generations, finally search optimal location, i.e. the optimum solution of optimization problem.That this patent adopts is Binary Particle Swarm Optimization (Binary Particle Swarm Optimization, BPSO), its difference is the history of every one dimension of particle and particle itself optimum, global optimum to be restricted to 1 or 0, and speed does not do this restriction.

Genetic algorithm (GeneticAlgorithm, GA) is a kind of random global search and optimized algorithm of using for reference organic sphere " survival of the fittest " natural selection process.It utilizes certain coding techniques, problem may solution be coded on chromosomal numeric string, simulate the evolutionary process of the colony's (likely separating) be made up of these strings, by finally searching out optimum solution after selection, intersection, mutation operation in colony.GA has strong robustness, the extensive and efficient advantage of adaptability, can carry out optimizing simultaneously, thus draw globally optimal solution to the multiple points in solution space.

Because GA has quick global search capability, chromosome shares information mutually, and whole population more equably to there being most region to move, but does not make full use of for feedback information in system, and often cause redundancy iteration, speed of convergence is slow.PSO, by the renewal to position and speed, makes full use of individual optimal value and colony's optimal value converges on optimal value, has the advantage of fast convergence rate, but is easily absorbed in " immature oils " phenomenon simultaneously.So, two algorithms combine by this patent, have complementary advantages, form a kind of fast convergence rate and can reduce the scale-of-two hybrid optimization algorithm (BPSOGA) of appearance " immature oils " probability, and during the distribution line failure section be used in containing Wind turbines locates.

Summary of the invention

The object of the present invention is to provide a kind of distribution line failure Section Location based on BPSO and GA, it is characterized in that comprising the following steps:

Step S1: obtain the SOT state of termination coding I treating section location distribution line _j, j=1,2,3..., D, D are total number of terminals;

Step S2: the dimension separated using BPSOGA search volume, as the sum treating total number of terminals D and section circuit in the distribution line of section location, using the dimension state value of Space Solutions as sector status encoded radio, and carries out initialization to parameter: the sub-population scale N of BPSO ₁, inertia weight ω, Studying factors c ₁and c ₂; GA population scale N ₂, crossover probability PX, mutation probability PM, heavily insert factor GGAP, maximum iteration time T, get N ₁=N ₂=N;

The initialization of the sub-population of step S3:BPSO and GA population: particle position pop1 (i) and particle rapidity v in the sub-population of random initializtion BPSO _i, and calculate particle fitness value fit1 (i), particle optimal location popbest (i) and particle maximum adaptation angle value p _best(i), colony optimal location gpopbest and colony maximum adaptation angle value g _best; Random initializtion GA population at individual solution pop2 (i) and ideal adaptation angle value fit2 (i), i=1,2,3...N; Wherein individually in particle position pop1 (i) and GA in BPSO separates pop2 (i) and all uses binary coding, there is identical dimension, all characterize respective segments in distribution line sector status encode;

Step S4: make k=1, carries out interative computation;

Step S5: to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, form the sub-population of BPSO of a new generation, and particle fitness value fit1 (i) of the sub-population of the BPSO upgrading described a new generation; Meanwhile, genetic manipulation is carried out to GA population, obtain the GA population of a new generation, and upgrade ideal adaptation angle value fit2 (i) of the GA population of described a new generation;

Ideal adaptation angle value fit2 (i) of particle fitness value fit1 (i) of step S6: BPSO more of new generation sub-population and the GA population of corresponding a new generation, get parent particle that the large particle of fitness value or individuality evolve as two populations next generations and parent individual: make j=1, carry out interative computation;

Step S7: if judge fit1 (j) < fit2 (j), then fit1 (j)=fit2 (j), pop1 (j)=pop2 (j), otherwise fit2 (j)=fit1 (j), pop2 (j)=pop1 (j), continues to perform step S8;

Step S8: the renewal carrying out BPSO population particle maximum adaptation angle value: the size comparing BPSO population particle fitness value and particle maximum adaptation angle value: if fit1 (j) > is p _best(j), then p _bestj ()=fit1 (j), popbest (j)=pop1 (j), continues to perform step S9, otherwise proceed to execution step S10;

Step S9: the renewal carrying out BPSO population maximum adaptation angle value: the size comparing BPSO population particle fitness value and colony's maximum adaptation angle value: if judge fit1 (j) > g _best, then g _best=fit1 (j), gpopbest=pop1 (j), continue to perform step S10; Otherwise directly perform step S10;

Step S10: make j=j+1, if judge j > N, then performs step S11, otherwise returns execution step S7;

Step S11: make k=k+1, judges whether if k≤T, if meet, then returns and performs step S5, continues to evolve; If do not meet, then terminate to evolve, export colony optimal location gpopbest.

Further, in described step S3, the computation process of the particle fitness value of the sub-population of described BPSO and the ideal adaptation angle value of GA population is as follows:

$I_j\left(s\right)=\underset{j}{\mathrm{\&Pi;}}{s}_i$

In formula: I _js () is the SOT state of termination function of a jth terminal, j=1, and 2,3...D, D are total number of terminals, and the circuit of jth between terminal and power supply is called the lines upstream of this terminal, and the circuit between line end is called the downstream line of this terminal, s _ibe the sector status coding of the i-th section, total total D section circuit, Π is logical OR computing, then above formula is expressed as a jth terminal downstream line state when having at least one to be 1, I _js () is just 1, otherwise be just 0;

Above formula is substituted in following formula fitness function:

$\mathrm{fit}\left(n\right)=M-\left(\underset{j=1}{\overset{D}{\mathrm{\&Sigma;}}}\right|I_j-I_j\left(s\right)|+\mathrm{\&eta;}\underset{i}{\overset{D}{\mathrm{\&Sigma;}}}{s}_i)$

In formula: fit (n) represents the fitness value of the n-th individuality, n=1,2,3...N, M get the twice of total number of terminals, and namely M=2D, η are the arithmetic number between [0,1], are called weight coefficient.

Further, in described step S5, to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, thus the process forming the sub-population of BPSO of a new generation is:

Step SA51: make i=1, starts interative computation;

Step SA52: the particle rapidity v calculating the BPSO population at individual of a new generation according to previous generation particle optimal location popbest (i), colony optimal location gpopbest and particle position pop1 (i) _iwith particle position pop1 (i), circular is as follows:

$v_{\mathrm{id}}^{k+1}={\mathrm{\&omega;v}}_{\mathrm{id}}^k+c_1{\mathrm{\&xi;}}_1({\mathrm{popbest}}_d^k\left(i\right)-{\mathrm{pop}}_d^k\left(i\right))+c_2{\mathrm{\&xi;}}_2({\mathrm{gpopbest}}_d^k-{\mathrm{pop}}_d^k\left(i\right))$

$\left\{\begin{array}{cc}{\mathrm{pop}}_d^{k+1}\left(i\right)=1& r_{\mathrm{id}}^{k+1}<\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\\ {\mathrm{pop}}_d^{k+1}\left(i\right)=0& r_{\mathrm{id}}^{k+1}\&GreaterEqual;\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\end{array}\right.$

In formula: i=1,2,3...N, k represent that current iteration number of times, ω are inertia weight, c ₁and c ₂for Studying factors, ξ ₁, ξ ₂with be all the random number between interval [0,1], at the particle rapidity of d dimension and particle position when representing that i-th particle iteration kth is secondary respectively; the particle optimal location tieed up at d when representing i-th particle iteration kth time; the colony's optimal location tieed up at d when representing iteration kth time; Wherein function is as follows:

$\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)=\left\{\begin{array}{cc}0.98& v_{\mathrm{id}}^{k+1}>4\\ \frac{1}{1+e^{-v_{\mathrm{id}}^{k+1}}}& -4\&le;v_{\mathrm{id}}^{k+1}\&le;4\\ -0.98& v_{\mathrm{id}}^{k+1}<-4\end{array}\right.$

Step SA53: i-th particle of sub-for the BPSO of a new generation population is substituted in fitness value function and calculates particle fitness value pop1 (i);

Step SA54: make i=i+1, if judge i > N, then stops iteration, performs step S6; Otherwise return and perform step SA52.

Further, in described step S5, the process of the genetic manipulation of GA population is as follows:

Step SB51: select: select N*GGAP individuality from N number of particle, according to the size of each ideal adaptation angle value, the individuality that adaptive value is larger is larger by the probability selected, otherwise less.Body is provided by following formula by the probability selected one by one:

$F\left(i\right)=\frac{\mathrm{fit}\left(i\right)}{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{fit}\left(i\right)}$

In formula: fit (i) is i-th individual ideal adaptation angle value, F (i) be this individuality by the probability selected, i=1,2,3..., N;

Step SB52: intersect: it is right to be mixed at random by each individuality in GA population, to every a pair individuality, exchanges the chromosome dyad between them with crossover probability PX, produces offspring individual;

Step SB53: variation: Stochastic choice body one by one in GA population, changes with mutation probability PM the allele that the genic value on certain some chromogene seat is other for the individuality chosen;

Step SB54: heavily insert: N*GGAP offspring individual is inserted in N number of parent individuality by the sequence based on fitness size, replaces the most unconformable parent individual, thus forms the population of a new generation;

Step SB55: the population at individual of a new generation is substituted in fitness function respectively and calculates its ideal adaptation angle value.

For achieving the above object, the present invention adopts following technical scheme:

The present invention compared with prior art has following beneficial effect:

1, BPSOGA of the present invention utilizes the strategy that two Evolution of Population and message exchange are shared, BPSO and GA two sub-populations are first separately evolved alone, the individuality of two groupy phases with numbering is compared successively after a generation of having evolved, the parent selecting optimum individual to evolve as the next generation is individual, form information sharing, and take full advantage of the advantage of each algorithm;

2, the present invention jointly can supervise in evolution engineering, and when namely a kind of algorithm is tending towards being absorbed in local convergence wherein, another kind of algorithm helps it to jump out, and draws close to globally optimal solution, reduces the appearance of " immature oils " phenomenon while improving speed of convergence;

3, the present invention has certain fault-tolerance, and the distortion information for SOT state of termination coding still accurately can orient fault section.

Accompanying drawing explanation

Fig. 1 is algorithm flow chart of the present invention.

Fig. 2 is the distribution line schematic diagram containing Wind turbines of the embodiment of the present invention.

Fig. 3 is that section 3 of the present invention breaks down, when distortion does not occur the SOT state of termination coding uploaded, according to the ideal adaptation angle value distribution plan that BPSOGA algorithm draws.

Fig. 4 is that section 3 of the present invention breaks down, when distortion does not occur the SOT state of termination coding uploaded, according to the dimension state value figure that BPSOGA algorithm draws.

Fig. 5 is that section 3 of the present invention breaks down, when distortion occurs the SOT state of termination coding uploaded, according to the ideal adaptation angle value distribution plan that BPSOGA algorithm draws.

Fig. 6 is that section 3 of the present invention breaks down, when distortion occurs the SOT state of termination coding uploaded, according to the dimension state value figure that BPSOGA algorithm draws.

Fig. 7 is BPSO, GA, BPSOGA of the present invention tri-kinds of algorithm the convergence speed contrast tables.

Embodiment

Below in conjunction with drawings and Examples, the present invention will be further described.

Please refer to Fig. 1, the invention provides a kind of distribution line failure Section Location based on BPSO and GA, it is characterized in that comprising the following steps:

Step S1: obtain the SOT state of termination coding I treating section location distribution line _j, j=1,2,3..., D, D are total number of terminals, D=30 in the present embodiment, SOT state of termination coding I _j=[1 11000000000000 00 000000000000 0];

Step S2: the dimension separated using BPSOGA search volume is as the sum treating total number of terminals D and section circuit in the distribution line of section location, three is all 30, using the dimension state value of Space Solutions as sector status encoded radio, and initialization is carried out to parameter: the sub-population scale N of BPSO ₁=50, inertia weight ω=0.9, Studying factors c ₁and c ₂, c ₁=c ₂=1.5; GA population scale N ₂=50, crossover probability PX=0.7, mutation probability PM=0.2, heavily insertion factor GGAP=0.9, maximum iteration time T=100, get N=50;

The initialization of the sub-population of step S3:BPSO and GA population: particle position pop1 (i) and particle rapidity v in the sub-population of random initializtion BPSO _i, and calculate particle fitness value fit1 (i), particle optimal location popbest (i) and particle maximum adaptation angle value p _best(i), colony optimal location gpopbest and colony maximum adaptation angle value g _best; Random initializtion GA population at individual solution pop2 (i) and ideal adaptation angle value fit2 (i), i=1,2,3...50; Wherein individually in particle position pop1 (i) and GA in BPSO separates pop2 (i) and all uses binary coding, there is identical dimension, all characterize respective segments in distribution line sector status encode;

The computation process of the particle fitness value of the sub-population of described BPSO and the ideal adaptation angle value of GA population is as follows:

$I_j\left(s\right)=\underset{j}{\mathrm{\&Pi;}}{s}_i$

In formula: I _js () is the SOT state of termination function of a jth terminal, j=1,2,3...30, and total number of terminals is 30, and the circuit of jth between terminal and power supply is called the lines upstream of this terminal, and the circuit between line end is called the downstream line of this terminal, s _ibe the sector status coding of the i-th section, always have 30 section circuits, Π is logical OR computing, then above formula is expressed as a jth terminal downstream line state when having at least one to be 1, I _js () is just 1, otherwise be just 0;

Above formula is substituted in following formula fitness function:

$\mathrm{fit}\left(n\right)=M-\left(\underset{j=1}{\overset{D}{\mathrm{\&Sigma;}}}\right|I_j-I_j\left(s\right)|+\mathrm{\&eta;}\underset{i}{\overset{D}{\mathrm{\&Sigma;}}}{s}_i)$

In formula: fit (n) represents the fitness value of the n-th individuality, n=1,2,3...50, M get the twice of total number of terminals, and M=60, η are the arithmetic number between [0,1], are called weight coefficient, η=0.5.

Step S4: make k=1, carries out interative computation;

Step S5: to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, form the sub-population of BPSO of a new generation, and particle fitness value fit1 (i) of the sub-population of the BPSO upgrading described a new generation; Meanwhile, genetic manipulation is carried out to GA population, obtain the GA population of a new generation, and upgrade ideal adaptation angle value fit2 (i) of the GA population of described a new generation;

In described step S5, to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, thus the process forming the sub-population of BPSO of a new generation is:

Step SA51: make i=1, starts interative computation;

Step SA52: the particle rapidity v calculating the BPSO population at individual of a new generation according to previous generation particle optimal location popbest (i), colony optimal location gpopbest and particle position pop1 (i) _iwith particle position pop1 (i), circular is as follows:

$v_{\mathrm{id}}^{k+1}={\mathrm{\&omega;v}}_{\mathrm{id}}^k+c_1{\mathrm{\&xi;}}_1({\mathrm{popbest}}_d^k\left(i\right)-{\mathrm{pop}}_d^k\left(i\right))+c_2{\mathrm{\&xi;}}_2({\mathrm{gpopbest}}_d^k-{\mathrm{pop}}_d^k\left(i\right))$

$\left\{\begin{array}{cc}{\mathrm{pop}}_d^{k+1}\left(i\right)=1& r_{\mathrm{id}}^{k+1}<\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\\ {\mathrm{pop}}_d^{k+1}\left(i\right)=0& r_{\mathrm{id}}^{k+1}\&GreaterEqual;\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\end{array}\right.$

In formula: i=1,2,3...50, k represent current iteration number of times, and ω is inertia weight, ω=0.9, c ₁and c ₂for Studying factors, c ₁=c ₂=1.5, ξ ₁, ξ ₂with be all the random number between interval [0,1], at the particle rapidity of d dimension and particle position when representing that i-th particle iteration kth is secondary respectively; the particle optimal location tieed up at d when representing i-th particle iteration kth time; the colony's optimal location tieed up at d when representing iteration kth time; Wherein function is as follows:

$\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)=\left\{\begin{array}{cc}0.98& v_{\mathrm{id}}^{k+1}>4\\ \frac{1}{1+e^{-v_{\mathrm{id}}^{k+1}}}& -4\&le;v_{\mathrm{id}}^{k+1}\&le;4\\ -0.98& v_{\mathrm{id}}^{k+1}<-4\end{array}\right.$

Step SA53: i-th particle of sub-for the BPSO of a new generation population is substituted in fitness value function and calculates particle fitness value pop1 (i);

Step SA54: make i=i+1, if judge i > 50, then stops iteration, performs step S6; Otherwise return and perform step SA52.

In described step S5, the process of the genetic manipulation of GA population is as follows:

Step SB51: select: select 45 individualities from 50 particles, according to the size of each ideal adaptation angle value, the individuality that adaptive value is larger is larger by the probability selected, otherwise less.Body is provided by following formula by the probability selected one by one:

$F\left(i\right)=\frac{\mathrm{fit}\left(i\right)}{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{fit}\left(i\right)}$

In formula: fit (i) is i-th individual ideal adaptation angle value, F (i) be this individuality by the probability selected, i=1,2,3..., 50;

Step SB52: intersect: it is right to be mixed at random by each individuality in GA population, to every a pair individuality, exchanges the chromosome dyad between them with crossover probability PX=0.7, produces offspring individual;

Step SB53: variation: Stochastic choice body one by one in GA population, changes with mutation probability PM=0.2 the allele that the genic value on certain some chromogene seat is other for the individuality chosen;

Step SB54: heavily insert: 45 offspring individuals are inserted in 50 parent individualities by the sequence based on fitness size, replaces the most unconformable parent individual, thus forms the population of a new generation;

Step SB55: the population at individual of a new generation is substituted in fitness function respectively and calculates its ideal adaptation angle value.

Ideal adaptation angle value fit2 (i) of particle fitness value fit1 (i) of step S6: BPSO more of new generation sub-population and the GA population of corresponding a new generation, get parent particle that the large particle of fitness value or individuality evolve as two populations next generations and parent individual: make j=1, carry out interative computation;

Step S7: if judge fit1 (j) < fit2 (j), then fit1 (j)=fit2 (j), pop1 (j)=pop2 (j), otherwise fit2 (j)=fit1 (j), pop2 (j)=pop1 (j), continues to perform step S8;

Step S8: the renewal carrying out BPSO population particle maximum adaptation angle value: the size comparing BPSO population particle fitness value and particle maximum adaptation angle value: if fit1 (j) > is p _best(j), then p _bestj ()=fit1 (j), popbest (j)=pop1 (j), continues to perform step S9, otherwise proceed to execution step S10;

Step S9: the renewal carrying out BPSO population maximum adaptation angle value: the size comparing BPSO population particle fitness value and colony's maximum adaptation angle value: if judge fit1 (j) > g _best, then g _best=fit1 (j), gpopbest=pop1 (j), continue to perform step S10; Otherwise directly perform step S10;

Step S10: make j=j+1, if judge j > 50, then performs step S11, otherwise returns execution step S7;

Step S11: make k=k+1, if judge k≤100, if meet, then returns and performs step S5, continue to evolve; If do not meet, then terminate to evolve, export colony optimal location gpopbest.

Fig. 2 is the distribution line schematic diagram containing Wind turbines of the embodiment of the present invention, when section 3 breaks down, when there is not distortion in the SOT state of termination coding uploaded, the ideal adaptation angle value distribution plan drawn according to BPSOGA algorithm and dimension state value figure are respectively as shown in accompanying drawing 3 and Fig. 4, individual fitness value is 59.5 to the maximum, corresponding individuality is numbered 17, dimension state value according to known 17th the individual correspondence of dimension state value figure is [0 01 00000000000000000000000000 0], attainable region section 3 breaks down thus.

In an alternative embodiment of the invention, same section 3 breaks down, but when there is distortion for [1 1100000000000000000000000010 0] in the SOT state of termination uploaded coding, the ideal adaptation angle value distribution plan obtained according to BPSOGA algorithm and dimension state value figure are respectively as shown in accompanying drawing 5 and Fig. 6, and known the method still can determine fault section exactly.

In an alternative embodiment of the invention, BPSO, GA, BPSOGA tri-kinds of algorithms are utilized respectively same fault to be carried out continuously to the localization of fault of 30 times, the three kinds of algorithms obtained occur that " immature oils " number of times contrast table is as shown in table 1, and speed of convergence comparison diagram as shown in Figure 7.

There is " immature oils " number of times contrast table in table 1 three kinds of algorithms

Algorithm types	BPSO	GA	BPSOGA
				There is " immature oils " number of times	3	12	0

The foregoing is only preferred embodiment of the present invention, all equalizations done according to the present patent application the scope of the claims change and modify, and all should belong to covering scope of the present invention.

Claims (4)

1., based on a distribution line failure Section Location of BPSO and GA, it is characterized in that comprising the following steps:

Step S1: obtain the SOT state of termination coding I treating section location distribution line _j, j=1,2,3..., D, D are total number of terminals;

Step S2: the dimension separated using BPSOGA search volume, as the sum treating total number of terminals D and section circuit in the distribution line of section location, using the dimension state value of Space Solutions as sector status encoded radio, and carries out initialization to parameter: the sub-population scale N of BPSO ₁, inertia weight ω, Studying factors c ₁and c ₂; GA population scale N ₂, crossover probability PX, mutation probability PM, heavily insert factor GGAP, maximum iteration time T, get N ₁=N ₂=N;

The initialization of the sub-population of step S3:BPSO and GA population: particle position pop1 (i) and particle rapidity v in the sub-population of random initializtion BPSO _i, and calculate particle fitness value fit1 (i), particle optimal location popbest (i) and particle maximum adaptation angle value p _best(i), colony optimal location gpopbest and colony maximum adaptation angle value g _best; Random initializtion GA population at individual solution pop2 (i) and ideal adaptation angle value fit2 (i), i=1,2,3...N; Wherein individually in particle position pop1 (i) and GA in BPSO separates pop2 (i) and all uses binary coding, there is identical dimension, all characterize respective segments in distribution line sector status encode;

Step S4: make k=1, carries out interative computation;

Step S5: to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, form the sub-population of BPSO of a new generation, and particle fitness value fit1 (i) of the sub-population of the BPSO upgrading described a new generation; Meanwhile, genetic manipulation is carried out to GA population, obtain the GA population of a new generation, and upgrade ideal adaptation angle value fit2 (i) of the GA population of described a new generation;

Ideal adaptation angle value fit2 (i) of particle fitness value fit1 (i) of step S6: BPSO more of new generation sub-population and the GA population of corresponding a new generation, get parent particle that the large particle of fitness value or individuality evolve as two populations next generations and parent individual: make j=1, carry out interative computation;

Step S7: if judge fit1 (j) < fit2 (j), then fit1 (j)=fit2 (j), pop1 (j)=pop2 (j), otherwise fit2 (j)=fit1 (j), pop2 (j)=pop1 (j), continues to perform step S8;

Step S8: the renewal carrying out BPSO population particle maximum adaptation angle value: the size comparing BPSO population particle fitness value and particle maximum adaptation angle value: if fit1 (j) > is p _best(j), then p _bestj ()=fit1 (j), popbest (j)=pop1 (j), continues to perform step S9, otherwise proceed to execution step S10;

Step S9: the renewal carrying out BPSO population maximum adaptation angle value: the size comparing BPSO population particle fitness value and colony's maximum adaptation angle value: if judge fit1 (j) > g _best, then g _best=fit1 (j), gpopbest=pop1 (j), continue to perform step S10; Otherwise directly perform step S10;

Step S10: make j=j+1, if judge j > N, then performs step S11, otherwise returns execution step S7;

Step S11: make k=k+1, if judge k≤T, if meet, then returns and performs step S5, continue to evolve; If do not meet, then terminate to evolve, export colony optimal location gpopbest.

2. a kind of distribution line failure Section Location based on BPSO and GA according to claim 1, it is characterized in that: in described step S3, the computation process of the particle fitness value of the sub-population of described BPSO and the ideal adaptation angle value of GA population is as follows:

$I_j\left(s\right)=\underset{j}{\mathrm{\&Pi;}}{s}_i$

In formula: I _js () is the SOT state of termination function of a jth terminal, j=1, and 2,3...D, D are total number of terminals, and the circuit of jth between terminal and power supply is called the lines upstream of this terminal, and the circuit between line end is called the downstream line of this terminal, s _ibe the sector status coding of the i-th section, total total D section circuit, Π is logical OR computing, then above formula is expressed as a jth terminal downstream line state when having at least one to be 1, I _js () is just 1, otherwise be just 0;

Above formula is substituted in following formula fitness function:

$\mathrm{fit}\left(n\right)=M-\left(\underset{j=1}{\overset{D}{\mathrm{\&Sigma;}}}\right|I_j-I_j\left(s\right)|+\mathrm{\&eta;}\underset{i}{\overset{D}{\mathrm{\&Sigma;}}}{s}_i)$

In formula: fit (n) represents the fitness value of the n-th individuality, n=1,2,3...N, M get the twice of total number of terminals, and namely M=2D, η are the arithmetic number between [0,1], are called weight coefficient.

3. a kind of distribution line failure Section Location based on BPSO and GA according to claim 1, is characterized in that: in described step S5, to particle position pop1 (i) and the particle rapidity v of the sub-population of BPSO _iupgrade, thus the process forming the sub-population of BPSO of a new generation is:

Step SA51: make i=1, starts interative computation;

Step SA52: the particle rapidity v calculating the BPSO population at individual of a new generation according to previous generation particle optimal location popbest (i), colony optimal location gpopbest and particle position pop1 (i) _iwith particle position pop1 (i), circular is as follows:

$v_{\mathrm{id}}^{k+1}={\mathrm{\&omega;v}}_{\mathrm{id}}^k+c_1{\mathrm{\&xi;}}_1({\mathrm{popbest}}_d^k\left(i\right)-{\mathrm{pop}}_d^k\left(i\right))+c_2{\mathrm{\&xi;}}_2({\mathrm{gpopbest}}_d^k-{\mathrm{pop}}_d^k\left(i\right))$

$\left\{\begin{array}{cc}{\mathrm{pop}}_d^{k+1}\left(i\right)=1& r_{\mathrm{id}}^{k+1}<\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\\ {\mathrm{pop}}_d^{k+1}\left(i\right)=0& r_{\mathrm{id}}^{k+1}\&GreaterEqual;\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)\end{array}\right.$

In formula: i=1,2,3...N, k represent that current iteration number of times, ω are inertia weight, c ₁and c ₂for Studying factors, ξ ₁, ξ ₂with be all the random number between interval [0,1], at the particle rapidity of d dimension and particle position when representing that i-th particle iteration kth is secondary respectively; the particle optimal location tieed up at d when representing i-th particle iteration kth time; the colony's optimal location tieed up at d when representing iteration kth time; Wherein function is as follows:

$\mathrm{sigmoid}\left(v_{\mathrm{id}}^{k+1}\right)=\left\{\begin{array}{cc}0.98& v_{\mathrm{id}}^{k+1}>4\\ \frac{1}{1+e^{-v_{\mathrm{id}}^{k+1}}}& -4\&le;v_{\mathrm{id}}^{k+1}\&le;4\\ -0.98& v_{\mathrm{id}}^{k+1}<-4\end{array}\right.$

Step SA53: i-th particle of sub-for the BPSO of a new generation population is substituted in fitness value function and calculates particle fitness value pop1 (i);

Step SA54: make i=i+1, if judge i > N, then stops iteration, performs step S6; Otherwise return and perform step SA52.

4. a kind of distribution line failure Section Location based on BPSO and GA according to claim 1, is characterized in that: in described step S5, the process of the genetic manipulation of GA population is as follows:

Step SB51: select: select N*GGAP individuality from N number of particle, according to the size of each ideal adaptation angle value, the individuality that adaptive value is larger is larger by the probability selected, otherwise less, and body is provided by following formula by the probability selected one by one:

$F\left(i\right)=\frac{\mathrm{fit}\left(i\right)}{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{fit}\left(i\right)}$

In formula: fit (i) is i-th individual ideal adaptation angle value, F (i) be this individuality by the probability selected, i=1,2,3..., N;

Step SB52: intersect: it is right to be mixed at random by each individuality in GA population, to every a pair individuality, exchanges the chromosome dyad between them with crossover probability PX, produces offspring individual;

Step SB53: variation: Stochastic choice body one by one in GA population, changes with mutation probability PM the allele that the genic value on certain some chromogene seat is other for the individuality chosen;

Step SB54: heavily insert: N*GGAP offspring individual is inserted in N number of parent individuality by the sequence based on fitness size, replaces the most unconformable parent individual, thus forms the population of a new generation;

Step SB55: the population at individual of a new generation is substituted in fitness function respectively and calculates its ideal adaptation angle value.