CN108491593A - A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm - Google Patents
A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm Download PDFInfo
- Publication number
- CN108491593A CN108491593A CN201810187977.7A CN201810187977A CN108491593A CN 108491593 A CN108491593 A CN 108491593A CN 201810187977 A CN201810187977 A CN 201810187977A CN 108491593 A CN108491593 A CN 108491593A
- Authority
- CN
- China
- Prior art keywords
- frog
- landscape
- fitness
- landscape pattern
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Structural Engineering (AREA)
- Civil Engineering (AREA)
- Architecture (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Processing Or Creating Images (AREA)
Abstract
An embodiment of the present invention provides a kind of landscape pattern optimizing methods, including:Any one landscape pattern optimizing scheme is expressed as a frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog by landscape pattern's raster data based on acquisition;The frog fitness of each frog in the frog population obtained is calculated, and is multiple sub-groups of preset quantity by frog population dividing according to frog fitness size;Update based on the algorithm that leapfrogs and mutation process carry out the study that leapfrogs to the frog in each sub-group, and repartition sub-group according to frog fitness size to all frogs after the study that leapfrogs, until reaching preset study number.A kind of landscape pattern optimizing method and system provided in an embodiment of the present invention, optimal landscape pattern optimizing scheme is solved by shuffled frog leaping algorithm, realizes the Synchronous fluorimetry of a variety of landscape types, the speed of service is very fast, model parameter is less, is suitable for the practical application of landscape pattern optimizing.
Description
Technical field
The present embodiments relate to Geographical Information Sciences technical field, it is based on mixing the calculation that leapfrogs more particularly, to one kind
The landscape pattern optimizing method and system of method.
Background technology
Landscape pattern generally refers to its spatial framework, the i.e. arrangement of the different landscape feature of size and shape spatially
And combination, including the type of landscape component units, number and spatial distribution and configuration, for example different types of patch can be in space
It is upper to be distributed in stochastic pattern, uniform type or accumulation type.It is the concrete embodiment of landscape heterogeneity, and is various ecological processes in difference
The result acted on scale.
Landscape pattern optimizing is particularly important to the Recycle mechanism of a variety of ecological processes in region, be a typical multiple target,
Non-linear, non-differentiability mixed questions need to consider previous Landscape Spatial Pattern Evolution, soil and soil during landscape pattern optimizing
The various problems such as loamy texture amount, underground water, landform.
But there is the master of defect and optimum results in itself certain design in the main model of existing landscape pattern optimizing
The property seen is stronger, and optimum results are often difficult to carry out in practice, affect the directive function to landscape pattern's construction.
Invention content
An embodiment of the present invention provides a kind of a kind of bases for overcoming the above problem or solving the above problems at least partly
In the landscape pattern optimizing method and system of shuffled frog leaping algorithm.
On the one hand the landscape pattern optimizing method based on shuffled frog leaping algorithm that an embodiment of the present invention provides a kind of, including:
Any one landscape pattern optimizing scheme is expressed as a blueness by S1, landscape pattern's raster data based on acquisition
The frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
S2, the frog fitness for calculating each frog in the frog population obtained, and will be green according to frog fitness size
Frog population dividing be preset quantity multiple sub-groups, the frog fitness by frog genetic fragment sum, frog landscape
Landscape suitability corresponding to concentration class and genetic fragment determines jointly;
S3, the update based on the algorithm that leapfrogs and mutation process carry out the study that leapfrogs to the frog in each sub-group, and right
All frogs after learning that leapfrog repartition sub-group according to frog fitness size, until reaching preset study number;
S4, it is by the landscape pattern optimizing scheme output represented by the highest frog of frog fitness in the frog population
Landscape pattern's grating image.
On the other hand the landscape pattern optimizing system based on shuffled frog leaping algorithm that an embodiment of the present invention provides a kind of, it is described
System includes:
Frog representation module is used for landscape pattern's raster data based on acquisition, by any one landscape pattern optimizing side
Case is expressed as a frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
Sub-group division module, the frog fitness for calculating each frog in the frog population obtained, and according to blueness
Frog fitness size by frog population dividing be preset quantity multiple sub-groups, the frog fitness by frog gene piece
Landscape suitability corresponding to section sum, frog landscape concentration class and genetic fragment determines jointly;
Leapfrog study module, for based on the algorithm that leapfrogs update and mutation process, to the frog in each sub-group into
The capable study that leapfrogs, and sub-group is repartitioned according to frog fitness size to all frogs after the study that leapfrogs, until reaching
Preset study number;
Prioritization scheme output module is used for the landscape represented by the highest frog of frog fitness in the frog population
The output of pattern prioritization scheme is landscape pattern's grating image.
According to the third aspect of the invention we, a kind of landscape pattern optimizing equipment based on shuffled frog leaping algorithm is provided, is wrapped
It includes:
Processor, memory, communication interface and bus;Wherein, the processor, memory, communication interface pass through described
Bus completes mutual communication;The communication interface is for the information between the test equipment and the communication equipment of display device
Transmission;The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program instruction
It is able to carry out a kind of landscape pattern optimizing method described above.
Fourth aspect of the invention embodiment provides a kind of computer program product, and the computer program product includes storage
Computer program in non-transient computer readable storage medium, the computer program include program instruction, when the journey
When sequence instruction is computer-executed, the computer is made to execute the above method.
Fifth aspect of the invention embodiment provides a kind of non-transient computer readable storage medium, the non-transient computer
Readable storage medium storing program for executing stores computer instruction, and the computer instruction makes the computer execute the above method.
A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm provided in an embodiment of the present invention, by mixed
The conjunction algorithm that leapfrogs solves optimal landscape pattern optimizing scheme, realizes the Synchronous fluorimetry of a variety of landscape types, and the speed of service is very fast,
Model parameter is less, is suitable for the practical application of Landscape optimization.
Description of the drawings
Fig. 1 is a kind of landscape pattern optimizing method flow diagram based on shuffled frog leaping algorithm provided in an embodiment of the present invention;
Fig. 2 is present status of land utilization schematic diagram provided in an embodiment of the present invention;
Fig. 3 is landscape pattern optimizing result schematic diagram provided in an embodiment of the present invention;
Fig. 4 is a kind of landscape pattern optimizing system construction drawing based on shuffled frog leaping algorithm provided in an embodiment of the present invention.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment is the present invention
A part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not having
The every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
Fig. 1 is a kind of landscape pattern optimizing method flow diagram based on shuffled frog leaping algorithm provided in an embodiment of the present invention,
As shown in Figure 1, the method includes:
Any one landscape pattern optimizing scheme is expressed as a blueness by S1, landscape pattern's raster data based on acquisition
The frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
S2, the frog fitness for calculating each frog in the frog population obtained, and will be green according to frog fitness size
Frog population dividing be preset quantity multiple sub-groups, the frog fitness by frog genetic fragment sum, frog landscape
Landscape suitability corresponding to concentration class and genetic fragment determines jointly;
S3, the update based on the algorithm that leapfrogs and mutation process carry out the study that leapfrogs to the frog in each sub-group, and right
All frogs after learning that leapfrog repartition sub-group according to frog fitness size, until reaching preset study number;
S4, it is by the landscape pattern optimizing scheme output represented by the highest frog of frog fitness in the frog population
Landscape pattern's grating image.
Specifically, in S1, landscape pattern's raster data is the grating image of current landscape pattern output, the landscape
Pattern prioritization scheme is the different landscape pattern corresponding to landscape pattern's different parameters, specifically, can adjust some grid
Landscape pattern's type in lattice, such as:Landscape pattern's type is adjusted to plough by waters.Further, due to the landscape
Pattern raster data is essentially image data, then the corresponding embodiment of the present invention is by the picture of landscape pattern's raster data
Member is expressed as a genetic fragment for frog.
It is understood that the thought used in the embodiment of the present invention for being substantially shuffled frog leaping algorithm is to landscape pattern
Raster data is handled.Shuffled frog leaping algorithm is a kind of completely new imitative biology intelligent optimization algorithm, has heuristic group
The characteristics of evolution, it has imitated the process that frog is looked for food, and the individual in population can learn to other individuals, according to the subgroup of division
Body carries out gene delivery, and optimal solution is obtained by global information exchange and local area deep-searching, reaches global search and part is believed
Cease the balance exchanged.The advantages of algorithm set particle cluster algorithm and genetic algorithm 2 kinds of Intelligent evolution algorithms, calculated performance is good
It is good, it can preferably avoid the problem that being absorbed in locally optimal solution.Shuffled frog leaping algorithm has just been used widely since proposition, in water
It succeeds application in the fields such as most optimum distribution of resources, Workshop Production task scheduling, the optimization of wind power plant Electrical Power System Dynamic.
In the prior art, since landscape pattern optimizing is often a typical multiple target, non-linear, non-differentiability mixes
Problem needs to consider the various problems such as previous Landscape Spatial Pattern Evolution, soil and soil quality, underground water, landform in optimization, existing
The scheme that technology provides tends not to effectively solve.
And shuffled frog leaping algorithm provided in an embodiment of the present invention is as a kind of novel intelligent optimization algorithm, it is such for solving
Problem has a clear superiority, and by computer modeling, can comparatively fast solve satisfactory landscape pattern optimizing scheme.
In S2, the frog fitness is by the genetic fragment sum of frog, frog landscape concentration class and genetic fragment institute
Corresponding landscape suitability determines that the fitness of the frog reflects the landscape pattern optimizing scheme representated by the frog jointly
To the adaptedness of landscape pattern, it is to be understood that the landscape pattern optimizing scheme representated by the higher frog of fitness is more
It is good.
Wherein, the pixel sum for landscape pattern's raster data that genetic fragment sum, that is, frog of the frog is included;
The frog landscape concentration class is the one of which spatial structure characteristic index of landscape, general to use formula RC=1-C/
CmaxIt is calculated, wherein RC is opposite concentration class index, and between value range is 0~1, C is complexity index method, CmaxIt is C
Maximum possible value can reflect the nonrandomness of different patch type or poly- in landscape by frog landscape concentration class index
Collection degree;Landscape suitability corresponding to the genetic fragment is suitability of the different landscape type in spatial dimension, at this
In the implementation process of inventive embodiments, by calculating landscape suitability, landscape pattern's suitability atlas can be generated.
It is understood that the embodiment of the present invention is obtained firstly the need of to frog population in step S2, the frog
Population is the set of the multiple frogs obtained, and acquisition modes preferably use the mode of variation to blueness in embodiments of the present invention
The frog is obtained, and mutation process is as follows:
Randomly select a certain number of genetic fragments of frog;
To selected a certain genetic fragment, its surrounding genes type is identified, calculate each gene type in current gene
Suitability at segment, if shared n kind gene types, suitability are expressed as P1,P2,P3…Pn;
Random number N is generated,
Judge current genetic mutation type according to the numerical value of N, if 0<N<P1, then variation is the 1st kind of gene type, if P1<N<
P2, then variation is the 2nd kind of gene type, and so on.
It is understood that using mutation operation, some or the multiple genetic fragments of frog is enabled to be changed,
To generate the new frog for being different from original frog, so multiple and different frogs can be generated using mutation operation, to constitute
Frog population.
In S3, the update of the algorithm that leapfrogs and mutation process refer to that frog carries out the study that leapfrogs in each sub-group
In the process, realize that the worst frog of sub-group learns to the best frog of sub-group, specifically, the update and change that pass through the algorithm that leapfrogs
Different process enables to the portion gene of the worst frog of sub-group to become the gene that exclusive or is updated to the best frog of sub-group.
It is understood that the minimum frog of frog fitness in the worst frog, that is, sub-group of sub-group, and sub-group is most
The highest frog of frog fitness in good frog, that is, sub-group.
Further, the learning process that leapfrogs is iterated through, the optimization of frog gene can be done step-by-step, improves each frog
Frog fitness, to selecting the highest frog of frog fitness from all frogs as the optimal frog of population.
In S4, output knot that landscape pattern optimizing scheme, that is, embodiment of the present invention corresponding to the optimal frog of population is provided
Fruit is translated into landscape pattern's grating image.
A kind of landscape pattern optimizing method provided in an embodiment of the present invention solves optimal landscape lattice by shuffled frog leaping algorithm
Office's prioritization scheme, realizes the Synchronous fluorimetry of a variety of landscape types, the speed of service is very fast, and model parameter is less, is suitable for landscape
The practical application of optimization.
On the basis of the above embodiments, step S2 is specifically included:
S21, for each frog, calculate the frog corresponding landscape pattern's suitability in all genetic fragments;
S22, it is based on preset landscape pattern's concentration class calculation formula, calculates the landscape pattern optimizing side that the frog represents
The landscape concentration class index of case;
Landscape corresponding to S23, genetic fragment sum, frog landscape concentration class and genetic fragment based on the frog
Suitability calculates the frog fitness;
S24, all frogs are ranked up according to frog fitness size in frog population, successively will according to clooating sequence
Frog is assigned in each sub-group of preset quantity.
Specifically, in S21, it is to be understood that genetic fragment included by every frog is multiple, and different gene
Landscape pattern's type of segment present position and reflection may be all different, then the embodiment of the present invention needs to calculate each blueness
Landscape pattern's suitability corresponding to each genetic fragment of the frog.
On calculating all genetic fragments after corresponding landscape pattern's suitability, according to corresponding in all genetic fragments
All landscape pattern's suitabilities, sum function can be utilized to calculate the landscape suitability of each frog, to calculate each
The landscape concentration class index of landscape pattern representated by frog, preset landscape pattern's concentration class calculation formula are:
Wherein, wherein CmaxIt is the maximum value of concentration class index, n is plaque type sum, P in landscapeijIt is plaque type i
The probability adjacent with j.
Specifically, the embodiment of the present invention is read the landscape suitability atlas of above-described embodiment acquisition by MATLAB softwares, from
And the parameter value calculated needed for landscape concentration class index is obtained according to landscape suitability atlas.
In S23, the scape corresponding to genetic fragment sum, frog landscape concentration class and genetic fragment based on the frog
Suitability is seen, calculates the frog fitness, calculation formula is as follows:
Wherein, F indicates frog fitness, WAFor frog landscape concentration class weight, WpFor landscape suitability weight, A is frog
Landscape concentration class, PijFor the landscape suitability corresponding to genetic fragment (i, j), D is frog genetic fragment sum.
It should be noted that in all embodiments of the invention, the formula about frog fitness calculates such as above formula institute
Show, meanwhile, the frog fitness calculation formula is also the object function in leapfroging learning process.
In S24, the embodiment of the present invention arranges all frogs in frog population according to frog fitness from big to small
Each frog is sequentially allocated in each sub-group by sequence according still further to clooating sequence.
Specifically, if preset sub-group quantity is M, every frog ranking fitness is divided by with M, by gained remainder
Identical frog is divided into the same sub-group.If remainder is 1 to be divided into the 1st sub-group, remainder is 0 to be divided into M
Frog population can thus be evenly dividing as several sub-groups by group.
On the basis of the above embodiments, step S21 includes:
Soil moisture, underground water buried depth, soil salt, soil nutrient, elevation, the gradient and rivers and lakes distance is chosen to make
For landscape pattern's suitability evaluation factor, and landscape pattern's suitability evaluation factor is converted to and landscape pattern's grid
The consistent raster data of lattice data distribution size of data;
Based on preset probit models, using the corresponding raster data of landscape pattern's suitability evaluation factor as solution
Variable is released, the corresponding binaryzation raster map layer of genetic fragment type calculates each genetic fragment type pair as explained variable
The landscape suitability answered.
It is understood that since the mutation process of shuffled frog leaping algorithm is typically random, however it is excellent in landscape pattern
In change, the mutation probability of each landscape types is not equivalent, and the embodiment of the present invention preferably uses probit models and comes to scape
Pattern suitability is seen to be calculated.
Probit models are a kind of generalized linear models, and explained variable Y is a two-valued variable (0,1 variable), thing
The probability that part occurs depends on explanatory variable.Probit models have good statistical property, it has also become most popular mould
One of type.The embodiment of the present invention introduce probit models, choose soil moisture, underground water buried depth, soil salt, soil nutrient,
Elevation, the gradient and rivers and lakes distance are independent variable, judge the suitability of landscape pattern at arbitrary spatial position, suitable by this
The mutation probability of suitable property control different landscape pattern, the Fast Convergent of implementation model.
It is understood that a variety of key features in optimizing as a result of pattern, such as:Soil moisture, underground water
Buried depth, soil salt, soil nutrient, elevation, the gradient and rivers and lakes distance are used as independent variable, are because becoming with landscape suitability
Amount structure probit models, then the variation direction that probability controls landscape in shuffled frog leaping algorithm is obtained by probit models,
Increase the practical significance of shuffled frog leaping algorithm.
Specifically, the embodiment of the present invention first by salt-soda soil area choose soil moisture, underground water buried depth, soil salt,
Soil nutrient, elevation, the gradient, with rivers and lakes distance be used as landscape pattern's suitability evaluation factor, while using space interpolation,
Grid conversion by each landscape pattern's suitability evaluation it is factor converting for landscape pattern's distributed data raster data of the same size,
So as to each suitability evaluation factor corresponding to raster data after converting as explanatory variable, while by each scape
Type-collection is seen at single raster map layer and binaryzation, as explained variable, it should be noted that exist in landscape pattern
A variety of landscape types, and the embodiment of the present invention illustrates a pixel of grid with the genetic fragment of frog, then different genes
Segment can also be divided into different genetic fragment types, thus by each genetic fragment type-collection at single raster map layer and two
Value recycles probit models, each corresponding landscape suitability of genetic fragment type is calculated, i.e., as dependent variable
Suitability of each landscape types in spatial dimension, the embodiment of the present invention preferably also form landscape pattern's suitability atlas,
It is operated for software.
On the basis of the above embodiments, the update based on the algorithm that leapfrogs described in step S3 and mutation process, to each
Frog in sub-group carries out the study that leapfrogs, and specifically includes:
For each sub-group, a random gene segment for the first frog is replaced with into the second frog same location
Genetic fragment, first frog are the frog that frog fitness is minimum in sub-group, and second frog is green in sub-group
The highest frog of frog fitness;
If the frog fitness of first frog does not improve after replacing, the gene of first frog is randomly selected
Segment, according to landscape suitability and landscape concentration class into row variation.
Specifically, the algorithm provided in an embodiment of the present invention that leapfrogs includes mainly two big operations:Update and variation, update essence
On be that the gene of the second frog same location is replaced with a random gene segment for the first frog for each sub-group
Segment, first frog are the frog that frog fitness is minimum in sub-group, and second frog is that frog is suitable in sub-group
The highest frog of response;
Such as:Frog A will be updated by the local gene of frog B, then process is as follows:
Random integers a, b, c, d are generated, wherein a, b is no more than the line number of frog genetic unit, and c, d are no more than genetic unit
Columns;
Using a, b as row coordinate, using c, d as row coordinate, rectangular extent is generated;
The genetic fragment that frog B in rectangular extent is generated in S2 is intercepted, frog A same areas are arrived in update, complete update behaviour
Make.
Mutation operation can be found in above-described embodiment, and details are not described herein for the embodiment of the present invention.
It is understood that by the update and variation of the execution that iterates, it can be by the sub-group in each sub-group
Worst frog learns to the best frog of sub-group, and after reaching preset study number, study stops, at this time entire frog kind
The highest frog of frog fitness of group is by the optimal landscape pattern optimizing scheme as the embodiment of the present invention.
On the basis of the above embodiments, if the frog fitness of first frog does not improve after the replacement,
The genetic fragment for randomly selecting first frog further includes according to landscape suitability and landscape concentration class into row variation:
The corresponding patch shape index of each landscape types of historical years is calculated with patch proximity index and patch to turn
Move probability;
Using the patch shape index, patch proximity index as dependent variable, the patch transition probability is used as from change
Amount carries out regression analysis, obtains the quantitative relation of the landscape indices and patch transition probability;
Quantitative relation based on the landscape indices Yu patch transition probability calculates turning for each patch of current year
Move probability;
Based on the transition probability of each patch of the current year, judge the variation for patch where the variation whether
Within a preset range, if described make a variation within a preset range, make a variation completion.
Wherein, the embodiment of the present invention can obtain the patch shape index and patch of landscape pattern's data of historical years first
Proximity index, to calculate the transfer area of each landscape patch of historical years, by shifting area and the patch gross area
The patch transition probability is calculated in ratio.
Further, the embodiment of the present invention is using the patch shape index, patch proximity index as dependent variable, patch
Transition probability carries out regression analysis as independent variable, obtains the quantitative relation of above-mentioned landscape indices and patch transition probability,
Again according to current year patch scale landscape indices, the transition probability of each patch of current year is calculated.
Based on the transition probability of each patch of the current year, whether within a preset range the variation is judged, if institute
State variation within a preset range, then make a variation completion.
It is understood that landscape indices are the high abstraction and quantificational expression to Regional Landscape Patterns, it is region
A variety of ecologies, the important characterization of economic process are one of important reference indexs that landscape pattern develops.By patch scale landscape lattice
Office's index and probit models couplings, the common evolution direction for determining research landscape pattern of area, make landscape pattern optimizing result more
Meet landscape rule, the practical significance of shuffled frog leaping algorithm optimum results can be improved.
Specifically, patch scale landscape indices and historical years each patch of the embodiment of the present invention according to historical years
Transition probability structure research area's patch scale landscape indices (patch shape index, patch proximity index) are shifted with patch
Quantitative relation between probability.Such as:If current year is a, the object of planning time is a+n, then calculates the patch ruler of a-n first
Landscape indices (patch shape index and patch proximity index) are spent, each plates of a-n is calculated later and shifts area, be based on
The area makees ratio with the patch gross area, obtains each patch transition probabilities of a-n.Referred to above-mentioned a-n patches scale landscape pattern
Number is dependent variable, and linear regression is carried out by independent variable of each patch transition probabilities of above-mentioned a-n, and the quantity obtained between the two is closed
System.
Further, the present invention calculates current year (a) patch scale landscape indices, is closed according to above-mentioned quantity
System, is calculated each patch transition probability of current year, this probability is multiplied with each plaque area, obtains each patch transfer face
Product, the foundation controlled using the area as landscape pattern's evolution quantity judge whether variation succeeds.
Specifically, when mutation operation is completed, it is more than evolution quantity to judge whether the variation will cause landscape pattern to be developed
Control, evolution quantity control if more than, then re-execute mutation operation, if the variation is less than the control of evolution quantity,
Complete mutation operation.
Preferably, it is 3 times of landscape pattern of research area number of types that setting of embodiment of the present invention evolution quantity, which controls, i.e., originally
The variation number of inventive embodiments setting is less than 3 times of landscape pattern of research area type class, such as:Study the landscape pattern in area
Type one shares 3 classes, respectively arable land, forest land and water body, then every time when variation is more than the control of evolution quantity will again into
Row variation, the embodiment of the present invention will be preferable to provide this number that makes a variation again and may not exceed 9 times.
The embodiment of the present invention constructs the quantitative relation between each patch transition probability and patch scale landscape indices, leads to
The quantity variation upper limit for crossing each landscape types in relationship control landscape pattern optimizing, avoids shuffled frog leaping algorithm and arbitrarily searches for
Caused by the excessive problem of initial solution range, improve model calculation efficiency.This way ensures optimum results and scape simultaneously
The consistency for seeing the natural evolvement process of pattern, reduces the enforcement difficulty of model optimization result.
The embodiment of the present invention is with Haixing County 2000 and landscape pattern's data in 2010, soil moisture, underground water buried depth, soil
Based on earth salinity, soil nutrient, elevation, the gradient and rivers and lakes range data, the embodiment of the present invention is emulated,
Wherein, Fig. 2 is present status of land utilization schematic diagram provided in an embodiment of the present invention, and Fig. 3 is landscape lattice provided in an embodiment of the present invention
Office's optimum results schematic diagram.
It is provided in an embodiment of the present invention to be as follows with reference to Fig. 2 and Fig. 3:
By the vector of arcgis turnstile lattice, space interpolation, raster data cut, grid resampling tool, to study area
Data are handled, the indexs such as Pixel size, grid line number and columns of unified each data.
Research area's soil moisture, underground water buried depth, soil salt, soil nutrient, elevation, slope are read using matlab softwares
Degree and rivers and lakes range data will study area's landscape pattern's data in 2010 as explained variable, profit as explanatory variable
Two-value regression analysis is carried out with probit models, generates frog landscape pattern of research area suitability atlas;
Area will be studied 2000, landscape pattern's data import fragstats softwares within 2010, calculate two times per one spot
The corresponding patch shape index of block and patch proximity index.By the raster symbol-base device of arcgis softwares, identification 2000 with
Each patch transfer part in 2010 determines each patch transfer area in 2000, and then obtain it by space attribute assignment operation
Transition probability.Regression analysis is carried out using matlab softwares, obtains the quantity of transition probability and patch scale landscape indices
Relationship.According to the quantitative relation and patch scale landscape indices in 2010, determines each patch transition probability in 2010 and turn
Area is moved, the foundation controlled in this, as landscape transition probability in optimization process;
Research area's landscape pattern's data in 2010 are read using MATLAB softwares, using random functions, are based on the landscape
Pattern data generate 1000 initial frogs, constitute initial frog population;
Landscape suitability atlas is read using MATLAB softwares, the landscape that each frog is calculated using sum function is suitable
Property, code is write according to landscape concentration class formula of index, calculates the landscape aggregation of landscape pattern representated by each frog
Spend index;
The fitness that every frog is judged according to object function is ranked up population frog according to fitness;
10 sub-groups are divided into according to frog ranking fitness, and by initial population, subgroup is judged by object function
The best frog of body, the worst frog of sub-group, the best frog of population;
Inside each sub-group, the random fragment of best frog in sub-group is chosen, replaces worst blueness inside sub-group
The corresponding position segment of the frog, if the fitness of worst frog does not improve after replacing, operation of morphing randomly selects worst
Genetic fragment inside frog, according to landscape suitability and landscape concentration class into row variation, and according to Landscape probability matrix
Whether landscape transfer face product is reasonable after judging variation, and variation is made only to occur in the reasonable scope;
Frog fitness is calculated according to object function, is resequenced to frog according to the fitness value, identifies population at this time
Best frog.Judge whether iterations reach 1000, if reaching, the algorithm operation that leapfrogs finishes, and population at this time is best
Frog is as landscape pattern's final optimization pass result.
It should be noted that parameter described in above-mentioned simulation process, which is only the embodiment of the present invention, preferably provides example, specifically
Parameter it is not limited in the embodiment of the present invention.
Fig. 4 is a kind of landscape pattern optimizing system construction drawing based on shuffled frog leaping algorithm provided in an embodiment of the present invention,
As shown in figure 4, the system comprises:Frog representation module 1, sub-group division module 2, leapfrog study module 3 and optimization side
Case output module 4, wherein:
Frog representation module 1 is used for landscape pattern's raster data based on acquisition, by any one landscape pattern optimizing side
Case is expressed as a frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
Sub-group division module 2 is used to calculate the frog fitness of each frog in the frog population obtained, and according to blueness
Frog fitness size by frog population dividing be preset quantity multiple sub-groups, the frog fitness by frog gene piece
Landscape suitability corresponding to section sum, frog landscape concentration class and genetic fragment determines jointly;
The study module 3 that leapfrogs for based on the algorithm that leapfrogs update and mutation process, to the frog in each sub-group into
The capable study that leapfrogs, and sub-group is repartitioned according to frog fitness size to all frogs after the study that leapfrogs, until reaching
Preset study number;
Prioritization scheme output module 4 is used for the landscape represented by the highest frog of frog fitness in the frog population
The output of pattern prioritization scheme is landscape pattern's grating image.
It is specific how by frog representation module 1, sub-group division module 2, leapfrog study module 3 and prioritization scheme
Output module 4 can be found in above-described embodiment to landscape pattern optimizing, and details are not described herein for the embodiment of the present invention.
A kind of landscape pattern optimizing system based on shuffled frog leaping algorithm provided in an embodiment of the present invention is leapfroged by mixing
Algorithm solves optimal landscape pattern optimizing scheme, realizes the Synchronous fluorimetry of a variety of landscape types, and the speed of service is very fast, model ginseng
Number is less, is suitable for the practical application of Landscape optimization.
The embodiment of the present invention provides a kind of landscape pattern optimizing system based on shuffled frog leaping algorithm, including:It is at least one
Processor;And at least one processor being connect with the processor communication, wherein:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
It enables to execute the method that above-mentioned each method embodiment is provided, such as including:S1, landscape pattern's raster data based on acquisition,
Any one landscape pattern optimizing scheme is expressed as a frog, and the pixel of landscape pattern's raster data is expressed as
The genetic fragment of frog;S2, the frog fitness for calculating each frog in the frog population obtained, and it is big according to frog fitness
It is small by multiple sub-groups that frog population dividing is preset quantity, the frog fitness by frog genetic fragment sum, green
Landscape suitability corresponding to frog landscape concentration class and genetic fragment determines jointly;S3, update and change based on the algorithm that leapfrogs
Different process carries out the study that leapfrogs to the frog in each sub-group, and is adapted to according to frog all frogs after the study that leapfrogs
Degree size repartitions sub-group, until reaching preset study number;S4, by frog fitness highest in the frog population
Frog represented by landscape pattern optimizing scheme output be landscape pattern's grating image.
The embodiment of the present invention discloses a kind of computer program product, and the computer program product is non-transient including being stored in
Computer program on computer readable storage medium, the computer program include program instruction, when described program instructs quilt
When computer executes, computer is able to carry out the method that above-mentioned each method embodiment is provided, such as including:S1, it is based on obtaining
Landscape pattern's raster data, any one landscape pattern optimizing scheme is expressed as a frog, and by the landscape pattern
The pixel of raster data is expressed as the genetic fragment of frog;S2, the frog for calculating each frog in the frog population obtained adapt to
Degree, and according to frog fitness size by frog population dividing be preset quantity multiple sub-groups, the frog fitness by
Landscape suitability corresponding to the genetic fragment sum of frog, frog landscape concentration class and genetic fragment determines jointly;S3, base
Update in the algorithm that leapfrogs and mutation process carry out the study that leapfrogs to the frog in each sub-group, and to the study that leapfrogs after
All frogs repartition sub-group according to frog fitness size, until reaching preset study number;S4, by the frog
Landscape pattern optimizing scheme output in population represented by the highest frog of frog fitness is landscape pattern's grating image.
The embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage
Medium storing computer instructs, and the computer instruction makes the computer execute the side that above-mentioned each method embodiment is provided
Method, such as including:Any one landscape pattern optimizing scheme is expressed as one by S1, landscape pattern's raster data based on acquisition
Frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;S2, the frog kind obtained is calculated
The frog fitness of each frog in group, and be multiple sons of preset quantity by frog population dividing according to frog fitness size
Group, the frog fitness is by the scape corresponding to the genetic fragment sum of frog, frog landscape concentration class and genetic fragment
Suitability is seen to determine jointly;S3, the update based on the algorithm that leapfrogs and mutation process, leapfrog to the frog in each sub-group
Study, and sub-group is repartitioned according to frog fitness size to all frogs after the study that leapfrogs, until reaching preset
Learn number;S4, the landscape pattern optimizing scheme represented by the highest frog of frog fitness in the frog population is exported
For landscape pattern's grating image.
One of ordinary skill in the art will appreciate that:Realize that all or part of step of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer read/write memory medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed;And storage medium above-mentioned includes:ROM, RAM, magnetic disc or light
The various media that can store program code such as disk.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be expressed in the form of software products in other words, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that:It still may be used
With technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features;
And these modifications or replacements, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of landscape pattern optimizing method based on shuffled frog leaping algorithm, which is characterized in that including:
Any one landscape pattern optimizing scheme is expressed as a frog by S1, landscape pattern's raster data based on acquisition, and
The pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
S2, the frog fitness for calculating each frog in the frog population obtained, and according to frog fitness size by frog kind
Group is divided into multiple sub-groups of preset quantity, and the frog fitness is assembled by genetic fragment sum, the frog landscape of frog
Landscape suitability corresponding to degree and genetic fragment determines jointly;
S3, the update based on the algorithm that leapfrogs and mutation process carry out the study that leapfrogs to the frog in each sub-group, and to leapfroging
All frogs after study repartition sub-group according to frog fitness size, until reaching preset study number;
S4, the landscape pattern optimizing scheme represented by the highest frog of frog fitness in the frog population is exported as landscape
Pattern grating image.
2. according to the method described in claim 1, it is characterized in that, step S2 is specifically included:
S21, for each frog, calculate the frog corresponding landscape pattern's suitability in all genetic fragments;
S22, it is based on preset landscape pattern's concentration class calculation formula, calculates the landscape pattern optimizing scheme that the frog represents
Landscape concentration class index;
Landscape corresponding to S23, genetic fragment sum, frog landscape concentration class and genetic fragment based on the frog is suitable
Property, calculate the frog fitness;
S24, all frogs are ranked up according to frog fitness size in frog population, according to clooating sequence successively by frog
It is assigned in each sub-group of preset quantity.
3. according to the method described in claim 2, it is characterized in that, step S21 includes:
It chooses soil moisture, underground water buried depth, soil salt, soil nutrient, elevation, the gradient and rivers and lakes distance and is used as scape
The pattern suitability evaluation factor is seen, and landscape pattern's suitability evaluation factor is converted to and landscape pattern's grid number
According to distributed data raster data of the same size;
Based on preset probit models, become the corresponding raster data of landscape pattern's suitability evaluation factor as explanation
Amount, the corresponding binaryzation raster map layer of each genetic fragment type calculate each genetic fragment type pair as explained variable
The landscape suitability answered.
4. according to the method described in claim 3, it is characterized in that, the update based on the algorithm that leapfrogs described in step S3 and variation
Process carries out the study that leapfrogs to the frog in each sub-group, specifically includes:
For each sub-group, a random gene segment for the first frog is replaced with to the gene of the second frog same location
Segment, first frog are the frog that frog fitness is minimum in sub-group, and second frog is that frog is suitable in sub-group
The highest frog of response;
If the frog fitness of first frog does not improve after replacing, the gene piece of first frog is randomly selected
Section, according to landscape suitability and landscape concentration class into row variation.
If 5. according to the method described in claim 4, it is characterized in that, it is described replace after first frog frog fitness
It does not improve, then randomly selects the genetic fragment of first frog, according to landscape suitability and landscape concentration class into row variation,
Further include:
It calculates the corresponding patch shape index of each landscape types of historical years and patch proximity index and patch transfer is general
Rate;
Using the patch shape index, patch proximity index as dependent variable, the patch transition probability as independent variable into
Row regression analysis obtains the quantitative relation of the landscape indices and patch transition probability;
Quantitative relation based on the landscape indices Yu patch transition probability, the transfer for calculating each patch of current year are general
Rate;
Based on the transition probability of each patch of the current year, judge the variation for patch where the variation whether pre-
If in range, if the variation is within a preset range, make a variation completion.
6. according to any methods of claim 2-5, which is characterized in that the frog fitness calculation formula is:
Wherein, F indicates frog fitness, WAFor frog landscape concentration class weight, WpFor landscape suitability weight, A is frog landscape
Concentration class, PijFor the landscape suitability corresponding to genetic fragment (i, j), D is frog genetic fragment sum.
7. a kind of landscape pattern optimizing system based on shuffled frog leaping algorithm, which is characterized in that the system comprises:
Frog representation module is used for landscape pattern's raster data based on acquisition, by any one landscape pattern optimizing scheme table
It is shown as a frog, and the pixel of landscape pattern's raster data is expressed as to the genetic fragment of frog;
Sub-group division module, the frog fitness for calculating each frog in the frog population obtained, and it is suitable according to frog
For response size by multiple sub-groups that frog population dividing is preset quantity, the frog fitness is total by the genetic fragment of frog
Landscape suitability corresponding to number, frog landscape concentration class and genetic fragment determines jointly;
Leapfrog study module, for based on the algorithm that leapfrogs update and mutation process, in each sub-group frog carry out the frog
Study is jumped, and sub-group is repartitioned according to frog fitness size to all frogs after the study that leapfrogs, until reaching default
Study number;
Prioritization scheme output module is used for the landscape pattern represented by the highest frog of frog fitness in the frog population
Prioritization scheme output is landscape pattern's grating image.
8. a kind of computer equipment, which is characterized in that including memory and processor, the processor and the memory pass through
Bus completes mutual communication;The memory is stored with the program instruction that can be executed by the processor, the processor
Described program instruction is called to be able to carry out the method as described in claim 1 to 6 is any.
9. a kind of computer program product, which is characterized in that the computer program product includes being stored in non-transient computer
Computer program on readable storage medium storing program for executing, the computer program include program instruction, when described program is instructed by computer
When execution, the computer is made to execute such as claim 1 to 6 any one of them method.
10. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Computer instruction is stored up, the computer instruction makes the computer execute such as claim 1 to 6 any one of them method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810187977.7A CN108491593A (en) | 2018-03-07 | 2018-03-07 | A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810187977.7A CN108491593A (en) | 2018-03-07 | 2018-03-07 | A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108491593A true CN108491593A (en) | 2018-09-04 |
Family
ID=63341860
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810187977.7A Pending CN108491593A (en) | 2018-03-07 | 2018-03-07 | A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108491593A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110222133A (en) * | 2019-05-29 | 2019-09-10 | 广州大学 | A kind of multiclass land use simulating and predicting method and device considering landscape index |
-
2018
- 2018-03-07 CN CN201810187977.7A patent/CN108491593A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110222133A (en) * | 2019-05-29 | 2019-09-10 | 广州大学 | A kind of multiclass land use simulating and predicting method and device considering landscape index |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | A multi-type ant colony optimization (MACO) method for optimal land use allocation in large areas | |
Liu et al. | Combining system dynamics and hybrid particle swarm optimization for land use allocation | |
Porta et al. | High performance genetic algorithm for land use planning | |
Stewart et al. | A multiobjective GIS-based land use planning algorithm | |
Demetriou et al. | An integrated planning and decision support system (IPDSS) for land consolidation: Theoretical framework and application of the land-redistribution modules | |
Taillandier et al. | A new BDI agent architecture based on the belief theory. Application to the modelling of cropping plan decision-making | |
Demetriou et al. | A spatial genetic algorithm for automating land partitioning | |
CN108805349A (en) | A kind of crop-planting spatial configuration optimal method based on improvement cuckoo algorithm | |
Florentino et al. | A multiple objective methodology for sugarcane harvest management with varying maturation periods | |
CN104700160A (en) | Vehicle route optimization method | |
CN104732091A (en) | Cellular automaton river bed evolution prediction method based on natural selection ant colony algorithm | |
CN103136707B (en) | A kind of Land_use change layered configuration method | |
CN102750732A (en) | Land resource utilization change dynamic prediction model based on GIS (Geographic Information System) and using method of dynamic prediction model | |
Dai et al. | Agent-based model of land system: Theory, application and modelling framework | |
CN110097529A (en) | A kind of farmland Grading unit division methods and system based on semantic rules | |
CN101419623A (en) | Geographical simulation optimizing system | |
CN109858180A (en) | A kind of area crops water consumption spatial framework optimum design method | |
Balbi et al. | Reviewing agent-based modelling of socio-ecosystems: a methodology for the analysis of climate change adaptation and sustainability | |
Bijandi et al. | Agricultural land partitioning model based on irrigation efficiency using a multi‐objective artificial bee colony algorithm | |
Chliaoutakis et al. | An agent-based model for simulating inter-settlement trade in past societies | |
Bijandi et al. | A novel approach for multi-stakeholder agricultural land reallocation using agent-based modeling: A case study in Iran | |
CN110222133A (en) | A kind of multiclass land use simulating and predicting method and device considering landscape index | |
CN108491593A (en) | A kind of landscape pattern optimizing method and system based on shuffled frog leaping algorithm | |
Demetriou et al. | LandParcelS: A module for automated land partitioning | |
Wright | ISIS: Toward an integrated spatial information system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |
|
RJ01 | Rejection of invention patent application after publication |