CN104077496A - Intelligent pipeline arrangement optimization method and system based on differential evolution algorithm - Google Patents

Abstract

The invention discloses an intelligent pipeline arrangement optimization method and system based on a differential evolution algorithm. The method includes the steps of conducting mathematical modeling on a pipeline to be arranged and arrangement space so as to determine an arrangement object, the constraint condition and the evaluation criteria, coding the arrangement object through polar coordinates and posture vectors, conducting optimization solution on a pipeline arrangement optimization mathematical model according to the differential evolution algorithm, and conducting constraint condition verification and arrangement adjustment on the arrangement optimization scheme obtained through the solution so as to obtain a final arrangement scheme. By means of the intelligent pipeline arrangement optimization method and system based on the differential evolution algorithm, the design cycle can be greatly shortened, optimization performance can be enhanced, the purpose of arranging pipelines on a large scale within limited time can be achieved, and the method and the system have the advantages of being short in arrangement design time, high in optimization accuracy, capable of quantitatively evaluating the arrangement scheme and the like.

Description

Intelligent pipeline layout optimization method and system based on differential evolution algorithm

Technical field

The invention belongs to industrial automation technical field, more specifically, relate to a kind of intelligent pipeline layout optimization method and system based on differential evolution algorithm.

Background technology

Layout of beam line's optimal design refers in a given arrangement space, arranging of pipeline is carried out reasonably arranging to meet necessary constraint condition, thereby reach certain optimum index.Reasonably layout of beam line's design, is conducive to installation, operation and the maintenance of pipeline, and the pipeline causing in the time of can reducing maintenance repeats dismounting number of times, significant for increasing work efficiency.

Current conventional layout of beam line's method for designing is to carry out layout of beam line's design according to piping arrangement design drawing, based on experience and existing drawing data draw piping diagram, the layout of then carrying out pipeline according to piping diagram is installed.The method realizes fairly simple, but exist, the topological design time cycle is long, Optimal performance is not strong, be difficult to the shortcomings such as quantitative evaluation placement scheme, and when layout is larger, workload sharply increases, cause the situation that is difficult to solve, so the method is not suitable for solving of extensive layout of beam line optimization problem.

Summary of the invention

The object of the invention is to overcome the deficiencies in the prior art, a kind of intelligent pipeline layout optimization method based on differential evolution algorithm is provided, the method is set up suitable mathematical model to layout of beam line's problem, then adopts differential evolution algorithm to carry out layout optimization design.

As one aspect of the present invention, the invention provides a kind of intelligent pipeline layout optimization method based on differential evolution algorithm, comprise the steps:

Step 1: treat layout pipeline and arrangement space and carry out mathematical modeling, determine layout object, constraint condition and interpretational criteria;

Step 2: adopt polar coordinates and attitude vector to encode to layout object;

Step 3: adopt differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and carry out constraint condition check and layout adjustment to solving the layout optimization scheme obtaining, obtain rational placement scheme.

The objective function of layout of beam line's optimized mathematical model of preferably, wherein setting up in step 1 is:

$J={\∂}_1{J}_1^{\′}+{\∂}_2{J}_2^{\′}+{\∂}_3{J}_3^{\′}$

$J_1=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{n}_i$

$J_2=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{l}_i$

$J_3=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j=1}{\overset{2n}{\mathrm{\Σ}}}\frac{1}{{r_{\mathrm{ij}}}^2}$

Wherein, J represents target function value, represent specific gravity factor, J ₁, J ₂, J ₃be respectively minimize when maintenance pipeline repeat dismounting number of times, minimize total pipeline path and maximize the homogeneity that pipeline distributes, J ' ₁, J ' ₂, J ' ₃be respectively J ₁, J ₂, J ₃normalize to [0,1] interval value, n _irepresent the equipment number that i bar pipeline comprises, l _ithe path that represents i bar pipeline, r _ij ²represent the distance between any two points, k represents pipeline number, and n represents total mapping device quantity; And

The constraint condition of layout of beam line's optimized mathematical model comprises: arrangement space size constraint, the mapping device that needs must be within arrangement space scope; Every there is not mechanical interference in transmission pipeline; Between any two connection devices for the treatment of layout, should meet minimum spacing constraint.

Preferably, wherein step 3 comprises the following steps:

Step 31: following parameter is set: Population Size NP, maximum iteration time M, zoom factor F and the factor C that intersects _r;

Step 32: produce initial population in the solution space of layout optimization problem, and current iteration number of times m=0 is set, wherein, initial population is comprised of NP Chromosome G, i.e. { G ₁, G ₂..., G _nP; Initial solution is at minimum edge dividing value g _{i, j (min)}maximum boundary value g _{i, j (max)}between random produce uniformly; Maximum boundary value and minimum edge dividing value determine according to particular problem, and initial population produces at random by following formula:

g _i，j，0＝g _i，j(min)+rand(0，1)(g _i，j(max)-g _i，j(min))

Wherein, rand (0,1) is the random number of scope between [0,1];

Step 33: when algorithm iteration is to maximum iteration time, when m=M or fitness function meet following formula, go to step 38;

|f(m)-f(m-1)|＜σ

Wherein, m is current iteration number of times, and M is predefined iterations; F (m) be m for the fitness value of optimum individual, f (m-1) be m-1 for the fitness value of optimum individual, σ is a predefined small integer;

Step 34: carry out mutation operation, the difference vector between other two individualities of choosing at random in each target individual in current population and population is carried out to linear superposition, generate successively variation individual, form variation population;

Step 35: according to crossover probability C _rindividuality in current population and the individuality making a variation in population are carried out to interlace operation, produce successively intersection individual, form intersection population;

Step 36: select operation, the individuality in current population and the individuality that intersects in population are carried out to fitness value calculation, from current population with intersect and select population N bar chromosome that fitness value is large as population of future generation;

Step 37: using population of future generation as current population, current iteration number of times m adds 1, goes to step 33;

Step 38: from population, select the individuality of fitness function maximum, as optimum solution output, the corresponding placement scheme of body one by one, the placement scheme that optimum individual is corresponding is optimum solution, and flow process finishes.

Preferably, wherein, described fitness function value is calculated as follows:

f＝1/J

Wherein, f represents fitness function value, and J represents the target function value of layout of beam line's optimization problem.

Preferably, wherein the mutation operation described in step 33 is: from N-Generation population, choose at random sequence number and be respectively r ₁, r _2,, r ₃chromosome it is individual that difference vector each is individual and two other individual formation carries out linear superposition formation variation, produces according to the following formula the vector V that makes a variation _{i, N+1}:

$V_{i,N+1}=G_{r_{1,N}}+F(G_{r_{2,N}}-G_{r_{3,N}}),r_1\≠r_2\≠r_3\≠i$

Wherein, F ∈ [1, NP] is zoom factor; NP is Population Size; G _{i, N}={ g _{i, 1, N}, g _{i, 1, N}..., g _{i, n, N}; r ₁, r ₂, r ₃what represent is the random mutually different integer producing in [1, NP].

Preferably, wherein the interlace operation described in step 34 is: individual V will make a variation _{i, N+1}with individual G in the present age _{i, N}intersect, obtain intersecting individual U _{i, N+1}, computing formula is as follows:

Wherein, i=1,2 ..., NP, j=1,2 ..., W, the dimension that W is problem; Rand (0,1) is the random number of scope between [0,1]; C _rfor intersecting the factor; I _ifor from sequence [1,2 ... W] in a random integer of selecting, U _{i, N+1}={ u _{i, 1, N+1}, u _{i, 2, N+1}..., u _{i, n, N+1}, V _{i, N+1}={ v _{i, 1, N+1}, v _{i, 2, N+1}..., v _{i, n, N+1}.

Preferably, wherein the selection described in step 35 is operating as: will intersect individual U _{i, N+1}individual G with current population _{i, N}compare, select the outstanding person of fitness as filial generation, concrete formula is described below:

Wherein, f () represents individual fitness computing function.

As another aspect of the present invention, the present invention also provides a kind of installation method of pipeline, and the placement scheme of employing as above-mentioned any one intelligent pipeline layout optimization method optimization carries out connection.

As an also aspect of the present invention, the present invention also provides a kind of intelligent pipeline layout optimization system based on differential evolution algorithm, comprising:

The first module, is used for treating layout pipeline and arrangement space and carries out mathematical modeling, determines layout object, constraint condition and interpretational criteria;

The second module, is used for adopting polar coordinates and attitude vector to encode to layout object;

The 3rd module, is used for adopting differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and carries out constraint condition check and layout adjustment to solving the layout optimization scheme obtaining, and realizes the intelligent optimization placement scheme of pipeline.

With respect to traditional layout of beam line's method, depend on experimental knowledge, have that the design cycle is long, Optimal performance is strong, be difficult to use in shortcomings such as solving extensive layout of beam line problem, the intelligent pipeline layout optimization method based on differential evolution algorithm that the present invention proposes can greatly shorten the design cycle, strengthen Optimal performance, and can in finite time, solve extensive layout of beam line problem, have the topological design time short, optimize precision high, can quantitative evaluation placement scheme etc. advantage.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the intelligent pipeline layout optimization method based on differential evolution algorithm of the present invention.

Fig. 2 a-2c is respectively iterations of the present invention and is respectively 60,80,100 o'clock, the layout result of each pipeline.

Embodiment

Below in conjunction with accompanying drawing, embodiments of the invention are described in detail: the present embodiment is implemented take technical solution of the present invention under prerequisite, and in conjunction with detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

Fig. 1 shows the process flow diagram of the method for the invention, abstract by layout of beam line's optimization problem is carried out, and sets up mathematical model, and adopts differential evolution algorithm to solve, and obtains preferably layout of beam line's scheme, thereby realizes intelligent layout of beam line.Method of the present invention comprises the following steps:

The first step: layout of beam line's optimization problem is carried out to mathematical modeling, determine layout object, constraint condition and interpretational criteria.For layout of beam line's optimization problem, arranging of pipeline path is to take the connection device of pipeline enclosure to be guiding, therefore layout of beam line's optimization problem can be converted to layout is carried out in the position of connection device.The related layout object of layout of beam line's optimization problem refers to the connection device of all pipeline enclosures.The interpretational criteria of considering mainly contains: while minimizing maintenance, pipeline repeats the number of times of dismounting, minimizes total pipeline path, maximizes the homogeneity that pipeline distributes.According to the actual physics meaning of layout of beam line, need to meet some specific constraint conditions, be respectively: arrangement space size constraint, institute's mapping device that needs must be within arrangement space scope; There is not mechanical interference in every road transmission pipeline; Between any two connection devices for the treatment of layout, should meet minimum spacing constraint.

Second step: according to the characteristic of layout of beam line's problem, layout object is encoded.Coding to as if treat that mapping device, gene representation treat the pose attribute of mapping device, utilize gene the position and attitude of equipment is described, wherein ρ _i, θ _ithe polar coordinates of difference indication equipment i in layout plane coordinate system, the attitude angle of indication equipment i (being that true origin and equipment center point place ray and equipment are towards angle).Chromosome, by a plurality of genomic constitution, represents a placement scheme.Chromosome represents with G, and every chromosome is by a plurality of gene g _iform, the number of gene is to treat the quantity N of mapping device, and chromosome coding mode is G={g ₁, g ₂..., g _n.

The 3rd step: adopt differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, to solving the layout optimization scheme obtaining, carry out constraint condition check and layout adjustment, obtain meeting the placement scheme of constraint condition, then it is judged whether to meet end condition, if met, obtain preferably placement scheme.

The interpretational criteria of layout of beam line's optimization problem that the first step is mentioned is specific as follows:

In conjunction with the real background of pipeline layout optimization problem, adopt three qualities of evaluating factor pair placement scheme to evaluate, be respectively: while minimizing maintenance, pipeline repeats the number of times of dismounting, minimizes total pipeline path, maximize the homogeneity that pipeline distributes.

By minimize when maintenance the pipeline number of times that repeats dismounting to be converted to the number of components that every road pipeline closed space comprises minimum, calculate respectively the number of components that each road pipeline closed space comprises, obtain overall budget containing number of components, mathematic(al) representation is as follows:

$J_1=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{n}_i---\left(1\right)$

Wherein, n _irepresent the equipment number that i bar pipeline comprises, k represents pipeline number.

Total pipeline path is the shortest: calculate respectively every pipeline path and also sue for peace and obtain total path length, mathematical formulae is as follows:

$J_2=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{l}_i---\left(2\right)$

Wherein, l _ithe path that represents i bar pipeline, k represents pipeline number, the less expression of this expression formula path total length is less.

The homogeneity that catoptron distributes is maximum: calculate the homogeneity that in layout of beam line's plane, catoptron distributes.Carry out in the following way uniform distribution calculating:

$J_3=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j=1}{\overset{2n}{\mathrm{\Σ}}}\frac{1}{{r_{\mathrm{ij}}}^2}---\left(3\right)$

Wherein, r _ij ²represent the distance between any two points, n represents total mapping device quantity, and this expression formula represents the gravitation situation of Plane-point, and the less expression of expression formula distributes more even.

Finally, by J ₁, J ₂, J ₃value normalize to [0,1] interval, use respectively J ' ₁, J ' ₂, J ' ₃represent.Obtaining general objective function is:

$J={\∂}_1{J}_1^{\′}+{\∂}_2{J}_2^{\′}+{\∂}_3{J}_3^{\′}---\left(4\right)$

Wherein, represent specific gravity factor.For general objective functional value, the smaller the better.

The 3rd step is mentioned and is adopted differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and the key of differential evolution algorithm is to determine fitness function, mutation operator, crossover operator and selection operator.Specific design method is as follows:

Fitness function: according to the general objective function of layout of beam line's mathematical model of having set up, determine the fitness function of differential evolution algorithm, because general objective functional value is the smaller the better, and fitness function value is generally got and is the bigger the better, and therefore usings the inverse of general objective functional value as fitness function:

f＝1/J (5)

Wherein, f represents fitness function value, and J represents the target function value of layout of beam line's optimization problem, as shown in Equation 4.

Mutation operator: choose at random sequence number and be respectively r from N-Generation population ₁, r _2,, r ₃chromosome each target individual and other any two other individual difference vectors forming are carried out to linear superposition formation variation individuality, produce according to the following formula variation vector V _{i, N+1}:

$V_{i,N+1}=G_{r_{1,N}}+F(G_{r_{2,N}}-G_{r_{3,N}}),r_1\≠r_2\≠r_3\≠i---\left(6\right)$

Wherein, F ∈ [1, NP] is zoom factor; NP is Population Size; G _{i, N}={ g _{i, 1, N}, g _{i, 2, N}..., g _{i, n, N}; r ₁, r ₂, r ₃what represent is the random mutually different integer producing in [1, NP].

Crossover operator: individual V will make a variation _{i, N+1}with individual G in the present age _{i, N}intersect, obtain intersecting individual U _{i, N+1}.The computing formula of interlace operation is as follows:

Wherein, i=1,2 ..., NP; J=1,2 ..., W; W is the dimension of problem; Rand (0,1) is the random number of scope between [0,1]; C _rfor intersecting the factor; I _ifor from sequence [1,2 ... W] in a random integer of selecting; So, U _{i, N+1}={ u _{i, 1, N+1}, u _{i, 2, N+1}..., u _{i, n, N+1}, V _{i, N+1}={ v _{i, 1, N+1}, v _{i, 2, N+1}..., v _{i, n, N+1}.

Select operator: will intersect individual U _{i, N+1}individual G with current population _{i, N}compare, select the large person of fitness value as offspring individual.Concrete formula is described below:

Wherein, f () represents individual fitness computing function.Fitness function is definite according to the objective function of layout of beam line's problem, and the size of fitness can directly be reacted the quality of placement scheme.

After setting mutation operator, crossover operator and selection operator, just can adopt differential evolution algorithm to solve layout of beam line's optimized mathematical model.Specific algorithm step is as follows:

Step 1: parameters (each parameter need to be chosen by test of many times the parameter value of effect optimum): Population Size NP, maximum iteration time M, zoom factor F and the factor C that intersects _r;

Step 2: produce initial population in the solution space of layout optimization problem, and current iteration number of times m=0 is set.Initial population is comprised of NP Chromosome G, i.e. { G ₁, G ₂..., G _nP.Initial solution is at minimum edge dividing value g _{i, j (min)}maximum boundary value g _{i, j (max)}between random produce uniformly so that the initial solution producing can cover whole search volume.Maximum boundary value and minimum edge dividing value determine according to particular problem, and initial population produces at random by following formula:

g _i，j，0＝g _i，j(min)+rand(0，1)(g _i，j(max)-g _i，j(min)) (9)

Wherein, rand (0,1) is the random number of scope between [0,1].

Step 3: when algorithm iteration is to maximum iteration time, when m=M (wherein m is current iteration number of times, and M is predefined iterations) or fitness function meet following formula, go to step 8;

|f(m)-f(m-1)|＜σ (10)

In formula, f (m) is that m is for the fitness value of optimum individual; F (m-1) is that m-1 is for the fitness value of optimum individual; σ is a predefined small integer.

Step 4: carry out mutation operation, the difference vector between other two individualities of choosing at random in each target individual in current population and population is carried out to linear superposition, generate successively variation individual, form variation population;

Step 5: according to crossover probability C _rindividuality in current population and the individuality making a variation in population are carried out to interlace operation, produce successively intersection individual, form intersection population;

Step 6: select operation, the individuality in current population and the individuality that intersects in population are carried out to fitness value calculation, from current population with intersect and select population N bar chromosome that fitness value is large as population of future generation;

Step 7: using population of future generation as current population, current iteration number of times adds 1 (m=m+1), goes to step 3;

Step 8: from population, select the individuality of fitness function maximum, as optimum solution output, the corresponding placement scheme of body one by one, the placement scheme that optimum individual is corresponding is optimum solution.Whole flow process finishes.

Based on said method, take certain layout that installs pipes as example, pipeline number is 4, and number of devices is 20, and an end of all pipelines is connected on the spheroid that radius is 3.5m, and equipment size is 0.5m*0.5m, and workshop length is 25m, and width is 25m.The control parameter of the differential evolution algorithm arranging in experiment is as follows: crossover probability initial value is 0.9, and variation probability initial value is 0.1, and crossover operator initial value is 0.5, and Population Size is 200, and iterations is 100.When Fig. 2 has provided iterations and has been respectively 60 (Fig. 2 (a)), 80 (Fig. 2 (b)), 100 (Fig. 2 (c)), the layout result of each pipeline, target function value is respectively: 0.2562,0.2569 and 0.0087.

The present invention also provides a kind of intelligent pipeline layout optimization system based on differential evolution algorithm, comprising:

The first module, is used for treating layout pipeline and arrangement space and carries out mathematical modeling, determines layout object, constraint condition and interpretational criteria;

The second module, is used for adopting polar coordinates and attitude vector to encode to layout object;

The 3rd module, is used for adopting differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and carries out constraint condition check and layout adjustment to solving the layout optimization scheme obtaining, and realizes the intelligent optimization placement scheme of pipeline.

Known by the pipe design application of actual factory, the intelligent pipeline layout optimization method based on differential evolution algorithm that the present invention proposes can greatly shorten the design cycle, strengthen Optimal performance, and can in finite time, solve extensive layout of beam line problem, have the topological design time short, optimize precision high, can quantitative evaluation placement scheme etc. advantage.

Above-described specific embodiment; object of the present invention, technical scheme and beneficial effect are further described; be understood that; the foregoing is only specific embodiments of the invention; be not limited to the present invention; within the spirit and principles in the present invention all, any modification of making, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.

Claims (9)

1. the intelligent pipeline layout optimization method based on differential evolution algorithm, comprises the following steps:

Step 1: treat layout pipeline and arrangement space and carry out mathematical modeling, determine layout object, constraint condition and interpretational criteria;

Step 2: adopt polar coordinates and attitude vector to encode to layout object;

Step 3: adopt differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and carry out constraint condition check and layout adjustment to solving the layout optimization scheme obtaining, realize the intelligent optimization placement scheme of pipeline.

2. intelligent pipeline layout optimization method as claimed in claim 1, the objective function of layout of beam line's optimized mathematical model of wherein setting up in step 1 is:

$J={\∂}_1{J}_1^{\′}+{\∂}_2{J}_2^{\′}+{\∂}_3{J}_3^{\′}$

$J_1=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{n}_i$

$J_2=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{l}_i$

$J_3=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j=1}{\overset{2n}{\mathrm{\Σ}}}\frac{1}{{r_{\mathrm{ij}}}^2}$

Wherein, J represents target function value, represent specific gravity factor, J ₁, J ₂, J ₃be respectively minimize when maintenance pipeline repeat dismounting number of times, minimize total pipeline path and maximize the homogeneity that pipeline distributes, J ' ₁, J ' ₂, J ' ₃be respectively J ₁, J ₂, J ₃normalize to [0,1] interval value, n _irepresent the equipment number that i bar pipeline comprises, l _ithe path that represents i bar pipeline, r _ij ²represent the distance between any two points, k represents pipeline number, and n represents total mapping device quantity; And

The constraint condition of layout of beam line's optimized mathematical model comprises: arrangement space size constraint, the mapping device that needs must be within arrangement space scope; Every there is not mechanical interference in transmission pipeline; Between any two connection devices for the treatment of layout, should meet minimum spacing constraint.

3. intelligent pipeline layout optimization method as claimed in claim 1, wherein step 3 comprises the following steps:

Step 31: following parameter is set: Population Size NP, maximum iteration time M, zoom factor F and the factor C that intersects _r;

Step 32: produce initial population in the solution space of layout optimization problem, and current iteration number of times m=0 is set, wherein, initial population is comprised of NP Chromosome G, i.e. { G ₁, G ₂..., G _nP; Initial solution is at minimum edge dividing value g _{i, j (min)}maximum boundary value g _{i, j (min)}between random produce uniformly; Maximum boundary value and minimum edge dividing value determine according to particular problem, and initial population produces at random by following formula:

g _i，j，0＝g _i，j(min)+rand(0，1)(g _i，ju(max-g _i，j(min))

Wherein, rand (0,1) is the random number of scope between [0,1];

Step 33: when algorithm iteration is to maximum iteration time, when m=M or fitness function meet following formula, go to step 38;

|f(m)-f(m-1)|＜σ

Wherein, m is current iteration number of times, and M is predefined iterations; F (m) be m for the fitness value of optimum individual, f (m-1) be m-1 for the fitness value of optimum individual, σ is a predefined small integer;

Step 34: carry out mutation operation, the difference vector between other two individualities of choosing at random in each target individual in current population and population is carried out to linear superposition, generate successively variation individual, form variation population;

Step 35: according to crossover probability C _rindividuality in current population and the individuality making a variation in population are carried out to interlace operation, produce successively intersection individual, form intersection population;

Step 36: select operation, the individuality in current population and the individuality that intersects in population are carried out to fitness value calculation, from current population with intersect and select population N bar chromosome that fitness value is large as population of future generation;

Step 37: using population of future generation as current population, current iteration number of times m adds 1, goes to step 33:

Step 38: from population, select the individuality of fitness function maximum, as optimum solution output, the corresponding placement scheme of body one by one, the placement scheme that optimum individual is corresponding is optimum solution, and flow process finishes.

4. intelligent pipeline layout optimization method as claimed in claim 3, wherein, described fitness function value is calculated as follows:

f＝1/J

Wherein, f represents fitness function value, and J represents the target function value of layout of beam line's optimization problem.

5. intelligent pipeline layout optimization method as claimed in claim 3, wherein the mutation operation described in step 33 is: from N-Generation population, choose at random sequence number and be respectively r ₁, r _2,, r ₃chromosome it is individual that difference vector each is individual and two other individual formation carries out linear superposition formation variation, produces according to the following formula the vector V that makes a variation _{i, N+1}:

$V_{i,N+1}=G_{r_{1,N}}+F(G_{r_{2,N}}-G_{r_{3,N}}),r_1\≠r_2\≠r_3\≠i$

Wherein, F ∈ [1, NP] is zoom factor; NP is Population Size; G _{i, N}={ g _{i, 1, N}, g _{i, 2, N}..., g _{i, n, N}; r ₁, r ₂, r ₃what represent is the random mutually different integer producing in [1, NP].

6. intelligent pipeline layout optimization method as claimed in claim 3, wherein the interlace operation described in step 34 is: individual V will make a variation _{i, N+1}with individual G in the present age _{i, N}intersect, obtain intersecting individual U _{i, N+1}, computing formula is as follows:

Wherein, i=1,2 ..., NP, j=1,2 ..., W, the dimension that W is problem; Rand (0,1) is the random number of scope between [0,1]; C _rfor intersecting the factor; I _ifor from sequence [1,2 ... W] in a random integer of selecting, U _{i, N+1}={ u _{i, 1N+1}, u _{i, 2, N+1}, u _{i, n, N+1}, V _{i, N+1}={ v _{i, 1, N+1}, v _{i, 2, N+1}..., v _{i, n, N+1}.

7. intelligent pipeline layout optimization method as claimed in claim 3, wherein the selection described in step 35 is operating as: will intersect individual U _{i, N+1}individual G with current population _{i, N}compare, select the outstanding person of fitness as filial generation, concrete formula is described below:

Wherein, f () represents individual fitness computing function.

8. a pipe installation method, adopts the placement scheme of the intelligent pipeline layout optimization method optimization as described in claim 1-7 any one to carry out connection.

9. the intelligent pipeline layout optimization system based on differential evolution algorithm, comprising:

The first module, is used for treating layout pipeline and arrangement space and carries out mathematical modeling, determines layout object, constraint condition and interpretational criteria;

The second module, is used for adopting polar coordinates and attitude vector to encode to layout object;

The 3rd module, is used for adopting differential evolution algorithm that layout of beam line's optimized mathematical model is optimized and is solved, and carries out constraint condition check and layout adjustment to solving the layout optimization scheme obtaining, and realizes the intelligent optimization placement scheme of pipeline.