CN108491968A - Based on agricultural product quality and safety emergency resources scheduling model computational methods - Google Patents
Based on agricultural product quality and safety emergency resources scheduling model computational methods Download PDFInfo
- Publication number
- CN108491968A CN108491968A CN201810221376.3A CN201810221376A CN108491968A CN 108491968 A CN108491968 A CN 108491968A CN 201810221376 A CN201810221376 A CN 201810221376A CN 108491968 A CN108491968 A CN 108491968A
- Authority
- CN
- China
- Prior art keywords
- individual
- selecting
- population
- individuals
- sample
- 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
- 238000000205 computational method Methods 0.000 title abstract 2
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 33
- 108090000623 proteins and genes Proteins 0.000 claims description 29
- 239000000463 material Substances 0.000 claims description 18
- 238000005457 optimization Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000012821 model calculation Methods 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 2
- 230000002068 genetic effect Effects 0.000 abstract description 7
- 238000010353 genetic engineering Methods 0.000 abstract 2
- 239000011159 matrix material Substances 0.000 description 14
- 108700028369 Alleles Proteins 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 4
- 210000000349 chromosome Anatomy 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 2
- 238000002922 simulated annealing Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004043 dyeing Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0835—Relationships between shipper or supplier and carriers
- G06Q10/08355—Routing methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Animal Husbandry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Game Theory and Decision Science (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses based on agricultural product quality and safety emergency resources scheduling model computational methods, the genetic algorithm efficiency that this method mainly solves in current agricultural product quality and safety emergency resources scheduling Path selection is low, convergence rate is slow, the problem of the shortcomings of being easily absorbed in local optimum, the algorithm takes the method for dividing sub- population, individual in overlaping stages choose different demarcation carries out genetic manipulation, ensure population diversity, it avoids being absorbed in local optimum, simultaneously, individual in same division also carries out genetic manipulation, it is effectively kept the merit of individual, avoid that algorithm is absorbed in local optimum and convergence efficiency is excessively slow to be carried out at the same time, reach the balance adjusted between convergence rate and searching optimal solution.
Description
Technical Field
The invention relates to an agricultural product quality safety-based emergency resource scheduling method, in particular to a novel method for improving crossed genetic algorithm to be used for emergency resource scheduling path optimization by introducing clustering partitions, and the method is an optimal path for emergency material transportation in agricultural product quality safety commanding and scheduling realized by using methods such as computer technology, genetic algorithm, clustering and the like.
Background
With the rapid development of the agricultural product industry, the supervision of the agricultural products is asynchronous, the quality safety of the agricultural products frequently occurs, and the method has great influence on the daily life of people and the growth of the whole social economy. In order to reduce the loss and damage caused by the quality safety of agricultural products to the maximum extent, the research and application of the agricultural product quality safety emergency resource scheduling model are carried out, and whether the accident can be supported timely and efficiently is directly influenced by the quality of the model.
At present, emergency resource scheduling for agricultural product quality safety is mainly scheduling for emergency experts and scheduling for emergency materials, and the emergency material scheduling is an important research link for emergency command scheduling. The emergency resource scheduling of the agricultural products mainly means that after an agricultural product quality safety event occurs, an emergency department quickly and effectively supplies resources according to the severity and rescue state of the problem and the existing resources, so that the processing work of a problem area is met. The objective function of emergency resource scheduling generally utilizes various means of transport to enable the required amount of material to be transported to the accident site as quickly as possible, with a minimum total time. In recent years, many planning methods based on path preference have been proposed and have achieved certain effects. These methods can be broadly divided into: precise algorithms, such as those of mixed integer programming, discrete optimization; heuristic algorithms, such as Lagrangian relaxation algorithms, solution space reduction algorithms; meta-heuristic algorithms, such as GN algorithm, tabu search algorithm, simulated annealing algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, and artificial bee colony algorithm.
Although a large number of path planning algorithms are proposed, many algorithms also have the defects of needing certain prior knowledge, low efficiency, low convergence speed, being prone to falling into local optimization and the like, for example, the GN algorithm has the problems of low efficiency and high time complexity, and the simulated annealing algorithm SA has the problem of low convergence speed and the like. Therefore, how to plan the emergency resource scheduling route quickly, accurately and efficiently is a hot problem for researching agricultural product quality safety emergency resource scheduling.
Disclosure of Invention
The invention aims to solve the problems of slow convergence of heredity, easy falling into local optimization and the like in an agricultural product quality safety emergency resource scheduling model, and provides a calculation method based on the agricultural product quality safety emergency resource scheduling model.
The technical scheme adopted by the technical problem required to be solved by the invention is as follows: an emergency resource scheduling model calculation method based on agricultural product quality safety comprises the following steps:
step 1, coding an emergency resource scheduling model;
the method adopts a symbol coding mode and aims at four elements of an agricultural product quality safety emergency resource scheduling model: the material, the transport vehicle, the supply point and the demand point are coded, and the material is respectively represented as a1，a2，a3，…aqThe transport vehicle being denoted c1，c2，c3，…cpThe supply point is denoted as s1，s2，s3，…smThe demand point is denoted by d1，d2，d3，…dn(ii) a Wherein q represents the number of materials, p represents the number of transport vehicles, m represents the number of supply points, and n represents the number of demand points;
the method is characterized in that a transport vehicle finishes one transport task as a gene, the sequence of all tasks finished by the transport vehicle in the task is used as a gene segment, and the gene segments of all transport vehicles participating in the transport task are connected in parallel according to vehicle numbers to form different individuals.
Step 2, initializing the population, wherein the method comprises the following steps:
generating initial population individuals according to a rule of coding the emergency resource scheduling model in the step 1, and specifically comprising the following steps of:
1) in the registered column and stateFor selected transport vehicles from the set C of free transport vehicles, denoted CiTo the location s of the transport vehicleiAs a starting point, randomly selecting a task place as a destination in the demand point set D, and recording the task place as DiRandomly selecting one material in the material set A as a transportation material, and recording as aiConstitute a gene, denoted as xis，xisIndicating transport vehicles ciTransporting materials aiFrom home siTo a destination diA combination of (a);
2) all genes related to one transport vehicle complete the initialization work of gene segments in a planning cycle according to the execution sequence of transport tasks, and are marked as P (x)i)＝{xi1，xi2，xi3，…，xin1, 2, 3, …, and N, i represents the sequence number of the transportation task.
3) Combining gene segments of all transport vehicles in parallel to finish the population initialization work with the size of N, and recording the number as X ═ X1，x2，x3，…，xN}。
Step 3, selecting a population fitness function;
1) each initialized gene is brought into an objective function for calculation, and the calculation formula of the objective function is as follows:
where N is the number of initialization solution sets, N is the dimension of each gene, ykIs a target solution of an initial solution under the action of a neural network training sample, okIs the input of the initial solution under the action of the neural network training sample.
2) And after different objective function values are obtained, the function values are brought into the fitness function to obtain different fitness. The fitness function f (i) is calculated as:
step 4, dividing the population by K-means;
the basic idea of the K-Means algorithm is to divide a sample set into K sub-populations according to the magnitude of the distance between samples, i.e. the similarity between pattern vectors, for m samples in a given domain N. Individuals of the same population are connected as closely as possible, and the difference between different populations is as large as possible.
The clustering is specifically realized as follows:
1) selecting a K clustering center:
randomly selecting an individual as a clustering center and recording the individual as mu1。
Selecting new cluster center according to the maximum distance principle, and calculating all individual sample points to the cluster center mu1Selecting the point with the farthest distance as the second cluster center, and recording as mu2The distance dij calculation method is expressed as:
calculating pattern vectors for the remaining individualsCharacteristic vector ofAndthe minimum of the distances, denoted as di:
di＝min[di1，di2]，i＝1，2，…，N
μi＝max[di]
and analogizing in turn to obtain all K clustering centers.
2) Classification of remaining samples:
and classifying according to a minimum distance principle, and as a result, classifying the cluster centers to the class represented by the cluster center closest to the cluster centers. Partitioning into multiple disjoint subsets XI，X2，…，XkAnd the subset meets the requirement:
X1∪X2∪…∪Xk＝X
1≤i≠j≤k (1)
3) clustering represents:
clustered sample representation:
wherein,representing the cluster center, q representing the number of the cluster sequence, and a sub-population Xc(1. ltoreq. c. ltoreq.k) is expressed as:
the membership relationship satisfies that each sample has only one membership population, and each membership population is non-empty and is represented as:
in each class, calculating a sample mean value of the class, and taking the sample mean value as a new clustering center of the class;
4) if the category of a certain sample is changed in 3), turning to 3) to continue execution;
5) and returning the clustering core and each sample category, and terminating the clustering process.
Step 5, crossover operation
1) Selecting an individual x using rouletteiRecord the category label mu to which the individual belongsaFrom class μaTo select a best individual xj。
2) Selection and classification muaClass μ farthest away1Selecting the best individual x in the categoryl；
3) Individual xiWith the individual xjThe set of individuals generated by the crossover operation is X;
4) selecting an individual xiAnd xjIn (1), with the individual xmDistant individuals, say xjIndividual xjAnd xm(ii) an individual;
the individual set generated by the cross operation is Y;
5) the best two individuals in the set X and the set Y are selected as descendant individuals by a greedy algorithm.
Step 6, mutation operation;
when the individual adaptability is larger than the average adaptability, the variation probability P is selectedmThe value range is (0, 0 body 002), otherwise, the variation probability PmThe value ranges are (0 body 08, 0 body 1). Expressed as:
wherein F (i) represents the fitness of the gene i, N represents the sample gene scale, and γmRepresenting the ratio of the fitness of the individual to the average fitness.
Step 7, selecting operation;
the selection operator adopts a selection strategy preferred by a combined optimization evolutionary algorithm, and the selection not only keeps the optimal individual in each generation, but also accelerates the convergence speed of the algorithm.
Step 8, decoding;
and selecting the individual with the largest module degree value in the population individuals for decoding to obtain the optimal solution of the agricultural product quality safety emergency resource transportation path.
Compared with the prior art, the method mainly solves the problems of low efficiency, low convergence speed, easy falling into local optimum and the like of a genetic algorithm in the selection of the quality safety emergency resource scheduling path of the agricultural products at present.
Drawings
The invention is further described with reference to the following figures and detailed description:
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a flow chart of clustering sub-population partitioning operations in the method of the present invention;
FIG. 3 is a graph comparing the convergence of the method of the present invention with a standard genetic algorithm;
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
FIG. 1 is a flow chart of a calculation based on an agricultural product quality safety emergency resource scheduling model, and the method comprises the following steps:
step one, population coding and initialization.
And step two, calculating an individual fitness function F.
And step three, clustering and dividing the sub-populations, wherein a specific flow chart is shown in fig. 2.
And step four, performing cross operation.
And step five, mutation.
And step six, selecting.
And seventhly, decoding to obtain the optimal solution of the agricultural product quality safety emergency resource transportation path.
1. The method for discovering the complex network community based on spectral clustering improvement intersection is characterized by comprising the following steps: the method comprises the following steps of,
step 1, coding the network community mining problem, wherein the method comprises the following steps:
numbering nodes in the network, representing individuals in the population by adopting codes based on gene locus adjacency, also called chromosomes, wherein the codes are a graph-based representation method, the code length of the chromosomes is represented by the number of the nodes in the network, each gene in the dyeing corresponds to one node in the network, and each gene i takes any adjacent node j in the network as an allele based on the fact that the nodes and most adjacent nodes are in the same community;
step 2, initializing the population, wherein the method comprises the following steps:
generating initial population individuals according to the coding rule of the step 1, and specifically comprising the following steps
1) Initializing a chromosome encoding a length n, each gene having an allele of 0
2) Sequentially traversing each gene locus, and searching a set of adjacent nodes in the network
3) Randomly selecting one node in the adjacent node set as the allele of the gene locus
4) Repeating the steps 1), 2) and 3) to finish the initialization work of a specified number of individuals
Step 3, selecting fitness function
The method comprises the steps that a Newman network modularity function is provided, a complex network mining problem is converted into an optimization problem, the optimization target is the modularity function, the modularity function can depict the quality degree of a community dividing result and is a performance index for evaluating the quality of a community structure in a network, and in complex network community mining based on a genetic algorithm, each individual represents one division of the community in the network, so that the modularity function Q is adopted as a fitness function in the genetic algorithm;
the modularity function Q is expressed as follows:
wherein A ═ Auv)n×nAn adjacency matrix representing nodes in network G if there is an edge connection between nodes u and vThen A isuv1, otherwise Auv0; δ (r (u) and r (v)) are community identity functions, wherein r (u) represents the community in which u is located, r (v) represents the community in which v is located, and if r (u) r (v), the value of r (u) r (v) is 1, the nodes u and v are in the same community; otherwise, the value is 0, which indicates that the nodes u and v are not in the same community; k is a radical ofuDegree, k, representing node uvDegrees representing node v; e represents the total number of edges in the network G, defined as
Step 4, dividing population individuals by spectral clustering
The basic idea of the spectral clustering algorithm is derived from a spectrogram theory, the clustering problem is regarded as a graph segmentation problem, the essence of the graph segmentation problem is that a feature vector of a Laplace matrix is utilized for clustering, and the graph segmentation problem is a pairing clustering algorithm;
the specific implementation of spectral clustering is as follows:
1) calculating a similarity matrix S
Firstly, constructing an undirected weighted graph, regarding individuals in a population as nodes in a network, regarding similarity among the individuals as edge weights in the network, and setting the population size as m
The similarity between individuals in the population is represented by normalized mutual information, and if A and B are two individuals in the population, the similarity S (A and B) between A and B is calculated as follows:
wherein C is a scrambling matrix whose elements CijRepresenting the number of common nodes owned by community i in individual A and community j in individual B; cAAnd CBThe number of communities in the division represented by the individual A and the individual B respectively; ci.Is the sum of the ith row element in matrix C, C.jIs the sum of the j column elements in matrix C, N isThe total number of nodes in the complex network; if A and B are identical, S (A, B) is 1, and if A and B are different, S (A, B) is 0, and the larger S (A, B) is, the more similar the two individuals A and B are; s is an adjacent matrix with m rows and m columns of element values as the similarity between individuals;
2) degree of calculation matrix D
Taking the sum of each row of elements in the matrix S as the element on the diagonal of the matrix D, and taking the elements at other positions as 0 to obtain D
3) Calculating the Laplace matrix L ═ D-W
4) Sorting the eigenvalues from small to large to obtain the first k eigenvalues and corresponding eigenvectors, and aligning the eigenvalues to the corresponding eigenvectors
Matrix M of M rows and k columns arranged to form a matrix
5) Regarding each row of the matrix M as a vector of a k-dimensional space, and clustering by using a k-means method; clustering
The category to which each row belongs in the result is the category to which the corresponding individual in the population belongs;
step 5, crossover operation
1) Selecting an individual p using roulette1Recording the category label i of the individual, and selecting one from the category i
Best individual p2；
2) Selecting the class j with the farthest distance from the class i, and selecting the best individual p in the class j3
3) Individual p1With the individual p2The set of individuals resulting from the crossover operation is X
4) Selection of individuals p1And p2In, with the individual p3Distant individuals, assumed to be p2Individual p2With the individual p3
The individual set generated by the crossover operation is Y
5) Selecting the best two individuals in the set X and the set Y as filial individuals through a greedy algorithm;
step 6, mutation operation
Selecting a node which enables the local modularity to be increased from adjacent nodes of a variant gene position as an allele during variation;
the local modularity is defined as follows:
wherein M islEdge representing the ratio of the sum of the total number of edges within the community to the sum of the number of edges connecting the community to the rest of the networkinRepresenting the number of connected edges within a community, edgeoutRepresenting the sum of the number of connecting edges of the community and other parts of the network;
step 7, selection operation
The selection operator adopts a mu + lambda selection strategy preferred by a combinatorial optimization evolutionary algorithm, the selection not only keeps the optimal individual in each generation, but also accelerates the convergence speed of the algorithm;
step 8, decoding
And selecting the individual with the largest modular degree value in the population individuals for decoding to obtain the optimal solution of the complex network community division.
Claims (1)
1. An emergency resource scheduling model calculation method based on agricultural product quality safety is characterized by comprising the following steps: the method comprises the following steps:
step 1, coding an emergency resource scheduling model;
the method adopts a symbol coding mode and aims at four elements of an agricultural product quality safety emergency resource scheduling model: the material, the transport vehicle, the supply point and the demand point are coded, and the material is respectively represented as a1，a2，a3，…aqThe transport vehicle being denoted c1，c2，c3，…cpThe supply point is denoted as s1，s2，s3，…smThe demand point is denoted by d1，d2，d3，…dn(ii) a Wherein q represents the number of materials, p represents the number of transport vehicles, m represents the number of supply points, and n represents the number of demand points;
the method comprises the following steps of taking a transportation task completed by a transportation vehicle as a gene, taking the sequence of all tasks completed by the transportation vehicle in the task as a gene segment, and connecting the gene segments of all the transportation vehicles participating in the transportation task in parallel according to vehicle numbers to form different individuals;
step 2, initializing the population, wherein the method comprises the following steps:
generating initial population individuals according to a rule of coding the emergency resource scheduling model in the step 1, and specifically comprising the following steps of:
1) the transport vehicle selected in the set C of transport vehicles registered in the column and having an idle state, denoted CiTo the location s of the transport vehicleiAs a starting point, randomly selecting a task place as a destination in the demand point set D, and recording the task place as DiRandomly selecting one material in the material set A as a transportation material, and recording as aiConstitute a gene, denoted as xis，xisIndicating transport vehicles ciTransporting materials aiFrom home siTo a destination diA combination of (a);
2) all genes related to one transport vehicle complete the initialization work of gene segments in a planning cycle according to the execution sequence of transport tasks, and are marked as P (x)i)＝{xi1，xi2，xi3，…，xin1, 2, 3, …, and N, i represents the execution sequence number of the transportation tasks;
3) combining gene segments of all transport vehicles in parallel to finish the population initialization work with the size of N, and recording the number as X ═ X1，x2，x3，…，xN}；
Step 3, selecting a population fitness function;
1) each initialized gene is brought into an objective function for calculation, and the calculation formula of the objective function is as follows:
where N is the number of initialization solution sets, N is the dimension of each gene, ykIs a target solution of an initial solution under the action of a neural network training sample, okIs the input of an initial solution under the action of a neural network training sample;
2) after different objective function values are obtained, the function values are brought into the fitness function to obtain different fitness; the fitness function f (i) is calculated as:
step 4, dividing the population by K-means;
the basic idea of the K-Means algorithm is that for m samples in a given domain N, a sample set is divided into K sub-populations according to the distance between the samples, namely the similarity between mode vectors; individuals of the same population are connected together as closely as possible, and the difference of different populations is as large as possible;
the clustering is specifically realized as follows:
1) selecting a K clustering center:
randomly selecting an individual as a clustering center and recording the individual as mu1；
Selecting new cluster center according to the maximum distance principle, and calculating all individual sample points to the cluster center mu1Selecting the point with the farthest distance as the second cluster center, and recording as mu2The distance dij calculation method is expressed as:
calculating pattern vectors for the remaining individualsCharacteristic vector ofAndthe minimum of the distances, denoted as di:
di＝min[di1，di2]，i＝1，2，…，N
μi＝max[di]
and so on to obtain all K clustering centers;
2) classification of remaining samples:
classifying according to a minimum distance principle, and classifying the cluster centers into the classes represented by the cluster centers closest to the cluster centers as a result; partitioning into multiple disjoint subsets XI，X2，…，XkAnd the subset meets the requirement:
X1∪X2∪…∪Xk＝X
1≤i≠j≤k (1)
3) clustering represents:
clustered sample representation:
wherein,representing the cluster center, q representing the number of the cluster sequence, and a sub-population Xc(1. ltoreq. c. ltoreq.k) is expressed as:
the membership relationship satisfies that each sample has only one membership population, and each membership population is non-empty and is represented as:
μci∈Eh
in each class, calculating a sample mean value of the class, and taking the sample mean value as a new clustering center of the class;
4) if the category of a certain sample is changed in 3), turning to 3) to continue execution;
5) returning the clustering core and each sample category, and terminating the clustering process;
step 5, crossover operation
1) Selecting an individual x using rouletteiRecord the category label mu to which the individual belongsaFrom class μaTo select a best individual xj；
2) Selection and classification muaClass μ farthest away1Selecting the best individual x in the categoryl；
3) Individual xiWith the individual xjThe set of individuals generated by the crossover operation is X;
4) selecting an individual xiAnd xjIn (1), with the individual xmDistant individuals, say xjIndividual xjAnd xm(ii) an individual;
the individual set generated by the cross operation is Y;
5) selecting the best two individuals in the set X and the set Y as filial individuals through a greedy algorithm;
step 6, mutation operation;
when the individual adaptability is larger than the average adaptability, the variation probability P is selectedmThe value range is (0, 0.002), otherwise the variation probability PmThe value range is (0.08, 0.1); expressed as:
wherein F (i) represents the fitness of the gene i, N represents the sample gene scale, and γmRepresenting the ratio of the fitness of the individual to the average fitness;
step 7, selecting operation;
the selection operator adopts a selection strategy preferred by a combinatorial optimization evolutionary algorithm, and the selection not only keeps the optimal individual in each generation, but also accelerates the convergence speed of the algorithm;
step 8, decoding;
and selecting the individual with the largest module degree value in the population individuals for decoding to obtain the optimal solution of the agricultural product quality safety emergency resource transportation path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810221376.3A CN108491968A (en) | 2018-03-17 | 2018-03-17 | Based on agricultural product quality and safety emergency resources scheduling model computational methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810221376.3A CN108491968A (en) | 2018-03-17 | 2018-03-17 | Based on agricultural product quality and safety emergency resources scheduling model computational methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108491968A true CN108491968A (en) | 2018-09-04 |
Family
ID=63339782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810221376.3A Pending CN108491968A (en) | 2018-03-17 | 2018-03-17 | Based on agricultural product quality and safety emergency resources scheduling model computational methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108491968A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948918A (en) * | 2019-03-08 | 2019-06-28 | 北京交通大学 | The synthesis distribution method of many storings moneys of possession emergency |
| CN110501983A (en) * | 2019-07-31 | 2019-11-26 | 农业农村部南京农业机械化研究所 | Expert control system and control method based on batch seed-coating machine |
| CN111260252A (en) * | 2020-02-18 | 2020-06-09 | 广东电网有限责任公司 | Power communication network field operation and maintenance work order scheduling method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070005522A1 (en) * | 2005-06-06 | 2007-01-04 | Wren William E | Resource assignment optimization using direct encoding and genetic algorithms |
| US20130308570A1 (en) * | 2012-05-17 | 2013-11-21 | Beijing University Of Posts And Telecommunications | Method for joint optimization of schedule and resource allocation based on the genetic algorithm |
| CN104504229A (en) * | 2014-09-19 | 2015-04-08 | 杭州电子科技大学 | Intelligent bus scheduling method based on hybrid heuristic algorithm |
| CN105303450A (en) * | 2015-11-09 | 2016-02-03 | 北京工业大学 | Complex network community discovery method based on spectral clustering improved intersection |
| CN107609630A (en) * | 2017-08-02 | 2018-01-19 | 广东建设职业技术学院 | A kind of depth confidence network parameter optimization method and system based on artificial bee colony |
-
2018
- 2018-03-17 CN CN201810221376.3A patent/CN108491968A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070005522A1 (en) * | 2005-06-06 | 2007-01-04 | Wren William E | Resource assignment optimization using direct encoding and genetic algorithms |
| US20130308570A1 (en) * | 2012-05-17 | 2013-11-21 | Beijing University Of Posts And Telecommunications | Method for joint optimization of schedule and resource allocation based on the genetic algorithm |
| CN104504229A (en) * | 2014-09-19 | 2015-04-08 | 杭州电子科技大学 | Intelligent bus scheduling method based on hybrid heuristic algorithm |
| CN105303450A (en) * | 2015-11-09 | 2016-02-03 | 北京工业大学 | Complex network community discovery method based on spectral clustering improved intersection |
| CN107609630A (en) * | 2017-08-02 | 2018-01-19 | 广东建设职业技术学院 | A kind of depth confidence network parameter optimization method and system based on artificial bee colony |
Non-Patent Citations (2)
| Title |
|---|
| 周艳聪等: "基于改进遗传算法的物流配送路径优化研究", 《计算机工程与科学》 * |
| 喻德旷等: "多受灾点应急救援车辆调度的优化遗传算法", 《计算机系统应用》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948918A (en) * | 2019-03-08 | 2019-06-28 | 北京交通大学 | The synthesis distribution method of many storings moneys of possession emergency |
| CN109948918B (en) * | 2019-03-08 | 2021-04-20 | 北京交通大学 | Comprehensive distribution method for local emergency mass storage materials |
| CN110501983A (en) * | 2019-07-31 | 2019-11-26 | 农业农村部南京农业机械化研究所 | Expert control system and control method based on batch seed-coating machine |
| CN110501983B (en) * | 2019-07-31 | 2021-03-26 | 农业农村部南京农业机械化研究所 | Expert control system and control method based on batch-type coating machine |
| CN111260252A (en) * | 2020-02-18 | 2020-06-09 | 广东电网有限责任公司 | Power communication network field operation and maintenance work order scheduling method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zeebaree et al. | Combination of K-means clustering with Genetic Algorithm: A review | |
| Mukhopadhyay et al. | An interactive approach to multiobjective clustering of gene expression patterns | |
| CN108491968A (en) | Based on agricultural product quality and safety emergency resources scheduling model computational methods | |
| CN110442143B (en) | Unmanned aerial vehicle situation data clustering method based on combined multi-target pigeon swarm optimization | |
| Zareizadeh et al. | A robust gene clustering algorithm based on clonal selection in multiobjective optimization framework | |
| Sree et al. | Identification of protein coding regions in genomic DNA using unsupervised FMACA based pattern classifier | |
| CN106991442A (en) | The self-adaptive kernel k means method and systems of shuffled frog leaping algorithm | |
| CN112783210B (en) | Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system | |
| CN110991518A (en) | Two-stage feature selection method and system based on evolution multitask | |
| Sarker et al. | Evolutionary computation: A gentle introduction | |
| Ebrahimpour et al. | Distributed feature selection: A hesitant fuzzy correlation concept for microarray high-dimensional datasets | |
| Alkafaween et al. | Improving initial population for genetic algorithm using the multi linear regression based technique (MLRBT) | |
| Han et al. | An efficient genetic algorithm for optimization problems with time-consuming fitness evaluation | |
| CN114546609A (en) | DNN inference task batch scheduling method facing heterogeneous cluster | |
| Shi et al. | Multi-objective multi-label classification | |
| Vimaladevi et al. | A microarray gene expression data classification using hybrid back propagation neural network | |
| Mukhopadhyay et al. | Multiobjective approach to categorical data clustering | |
| CN115293430A (en) | Unmanned node cooperation method and system based on cooperative coevolution algorithm | |
| Gong et al. | Solving multiobjective clustering using an immune-inspired algorithm | |
| Akey Sungheetha | An efficient clustering-classification method in an information gain NRGA-KNN algorithm for feature selection of micro array data | |
| Mondal et al. | MOSCFRA: a multi-objective genetic approach for simultaneous clustering and gene ranking | |
| CN113792494A (en) | Multi-target flexible job shop scheduling method based on migrating bird group algorithm and cross fusion | |
| Boulif | A new cut-based genetic algorithm for graph partitioning applied to cell formation | |
| CN108090514B (en) | Infrared image identification method based on two-stage density clustering | |
| Maulik et al. | Genetic algorithms and multiobjective optimization |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |