Workshop scheduling method and system based on improved multi-objective genetic algorithm

Abstract

The present invention provides a workshop scheduling method and system based on an improved multi-objective genetic algorithm, and relates to the technical field of task scheduling. The workshop scheduling method based on the improved multi-objective genetic algorithm of the present invention obtains the task data of the workshop scheduling, sets the input parameters according to the task data; obtains the pre-built workshop fuzzy scheduling model; obtains the optimal scheduling plan for the workshop scheduling based on the input parameters, the workshop fuzzy scheduling model, and the genetic algorithm with dynamically changing crossover probability and mutation probability. The adaptive crossover probability and mutation probability designed by the present invention enable the population to dynamically adjust the global search and local search capabilities, avoid the population from falling into premature maturity, and help improve the quality of the solution.

Description

Technical Field

[0001] The invention relates to the technical field of task scheduling, and in particular to a workshop scheduling method and system based on an improved multi-objective genetic algorithm. Background Art

[0002] The workshop scheduling problem is a resource allocation problem that satisfies the requirements of task configuration and sequence constraints. It is one of the most difficult combinatorial optimization problems and is an NP-hard problem. Since its solution space is massive and the massive feasible solutions correspond to massive calculations, general solutions cannot effectively obtain better feasible solutions.

[0003] In the prior art, a feasible solution for workshop scheduling is generally solved by a genetic algorithm, and the process is as follows: S1, population initialization; S2, fast non-dominated sorting of the initial population; S3, selection, crossover, and mutation operations are performed on the initial population to generate a first-generation offspring population; S4, the parent population and the offspring population are merged, and fast non-dominated sorting is performed, and the crowding degree of the individuals in each non-dominated layer is calculated, and suitable individuals are selected according to the non-dominated relationship and the crowding degree of the individuals to form a new parent population; S5, selection, crossover, and mutation operations are performed on the parent population to generate a new generation of offspring population; S6, whether the maximum number of iterations is met is determined, if so, the iteration process is terminated and the result is output, otherwise it returns to S4.

[0004] However, in existing methods, the crossover probability and mutation probability are fixed during the iteration process, which makes the population prone to stagnation and fall into a local optimal solution, resulting in low quality of feasible solutions. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the shortcomings of the prior art, the present invention provides a workshop scheduling method and system based on an improved multi-objective genetic algorithm, which solves the technical problem that the crossover probability and mutation probability in the existing method are fixed during the iteration process, making the population easily fall into a stagnant state and fall into a local optimal solution.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] In a first aspect, the present invention provides a shop scheduling method based on an improved multi-objective genetic algorithm, comprising:

[0010] S1. Obtain task data for workshop scheduling and set input parameters according to the task data;

[0011] S2, obtaining a pre-built workshop fuzzy scheduling model;

[0012] S3. Based on the input parameters, the workshop fuzzy scheduling model, and the genetic algorithm with dynamically changing crossover probability and mutation probability, an optimal scheduling solution for workshop scheduling is obtained.

[0013] Preferably, the pre-built workshop fuzzy scheduling model includes an objective function and constraints.

[0014] The objective function includes:

[0015]

[0016]

[0017] in,

[0018] The objective function (1) represents minimizing the maximum fuzzy completion time;

[0019] The objective function (2) represents minimizing the total fuzzy processing cost;

[0020] is the maximum fuzzy completion time; is the total processing cost of the workpiece.

[0021] Preferably, the constraints include:

[0022]

[0023]

[0024]

[0025]

[0026] x ij ∈{0,1},i=1,...,m,j=1,...,n (7)

[0027] in,

[0028] Constraint (3) is used to calculate the fuzzy completion time on each machine;

[0029] Constraint (4) is used to calculate the total fuzzy processing cost, which is equal to the cumulative sum of all machine processing costs;

[0030] Constraint (5) represents the maximum fuzzy completion time of the system, which is equal to the maximum completion time of all machines;

[0031] Constraint (6) indicates that each workpiece can only and must be assigned to any one machine for processing;

[0032] Constraint (7) indicates that if job j is assigned to machine i for processing, then x ij =1, otherwise 0;

[0033] m is the number of machines; n is the number of workpieces; is the fuzzy completion time of job j, is the fuzzy completion time of machine i; v i is the processing speed of machine i; β i is the unit processing cost of machine i.

[0034] Preferably, the genetic algorithm with dynamically changing crossover probability and mutation probability includes:

[0035] S301, encoding the workshop work plan by integer coding;

[0036] S302, randomly generate NP parent chromosomes, each of which has a length of n and corresponds to a corresponding workpiece. The value on the jth dimension represents the processing machine to which the workpiece is assigned. Determine the current number of iterations T and the maximum number of iterations T max and the individual initial crossover probability p3, where n is the number of workpieces, 1≤j≤n;

[0037] S303: Fuzzy processing time for workpiece j The processing time after defuzzification is Calculate the target value of each chromosome on the target function (1) and the target function (2);

[0038] S304, sorting the two target values ​​from small to large, performing specific search and update on the parent chromosome, and performing fast non-dominated sorting on the newly generated chromosome;

[0039] S305, determining the optimal classification number k of chromosomes by using Canopy algorithm * , calculate the Hamming distance between chromosomes, cluster the populations with high similarity through the K-means clustering algorithm, and generate multiple cluster groups;

[0040] S306, generate random numbers r1∈(0,1), r3∈(0,1), if r1<p1, then randomly generate two integers A1∈(0,NP), A2∈(0,NP), A1≠A2 in the same cluster group, and perform same-group crossover on these two chromosomes; if r1>p1, then randomly generate two integers A3∈(0,NP), A4∈(0,NP), A3≠A4 in different groups, and perform different-group crossover on these two chromosomes to generate new chromosomes; during the crossover process, select chromosomes that satisfy r3≤p3;

[0041] p1=1-e -kx

[0042] Where x represents the number of iterations; k represents the conventional coefficient, k>0;

[0043] S307, for each chromosome obtained in step S306, a random number r2 is generated, and if r2≤p2 is satisfied, a mutation operation is performed on the particle;

[0044] p2=n * / n max

[0045] Among them, n * Indicates the number of currently feasible solutions; n max represents the maximum disposable number;

[0046] S308. Determine whether T≥Tmax. If not, set T=T+1 and go to S304. If satisfied, output all solutions of the non-dominated frontier and the corresponding scheduling plans to obtain the optimal scheduling plan for workshop scheduling.

[0047] Preferably, the specific search includes:

[0048] In the parent chromosome, sorting is performed for the two targets respectively, and the first k1 chromosomes are selected, and the more common information appearing in the k1 chromosomes is retained to improve the k2 chromosomes that are arranged later; if there is no common information or the process has been executed, the differential mutation operation is used to update the chromosome information.

[0049] Preferably, the S306 further includes:

[0050] If the crossover probability of all individuals is the same, then design the corresponding crossover probability p3 for each individual, generate a random number r3, and if r3≤p3, perform a mutation operation on the chromosome;

[0051]

[0052] Among them, p3 is the initial crossover probability of each individual, t and t maxThey represent the number of generations that an individual has survived and the total number of generations that the population has experienced.

[0053] Preferably, the S304 further includes:

[0054] During the search process, a taboo table is constructed to store all chromosome types that have appeared in recent rounds of iterative operations and the N1 crossover and mutation operations that have been performed on the same chromosome recently. This avoids returning to the original solution until the taboo is lifted after N2 rounds; N1 and N2 are artificially defined values.

[0055] In a second aspect, the present invention provides a shop scheduling system based on an improved multi-objective genetic algorithm, comprising:

[0056] A data acquisition module is used to obtain task data for workshop scheduling and set input parameters according to the task data;

[0057] A model acquisition module is used to acquire a pre-built shop fuzzy scheduling model;

[0058] The scheme acquisition module is used to acquire the optimal scheduling scheme for workshop scheduling based on the input parameters, the workshop fuzzy scheduling model, and a genetic algorithm with dynamically changing crossover probability and mutation probability.

[0059] In a third aspect, the present invention provides a computer-readable storage medium storing a computer program for workshop scheduling based on an improved multi-objective genetic algorithm, wherein the computer program enables a computer to execute the workshop scheduling method based on the improved multi-objective genetic algorithm as described above.

[0060] In a third aspect, the present invention provides an electronic device, comprising:

[0061] one or more processors;

[0062] Memory; and

[0063] One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs including instructions for executing the shop scheduling method based on the improved multi-objective genetic algorithm as described above.

[0064] (III) Beneficial effects

[0065] The present invention provides a workshop scheduling method and system based on an improved multi-objective genetic algorithm. Compared with the prior art, it has the following beneficial effects:

[0066] The present invention discloses a workshop scheduling method based on an improved multi-objective genetic algorithm, which obtains task data of workshop scheduling, sets input parameters according to the task data, obtains a pre-built workshop fuzzy scheduling model, and obtains the optimal scheduling solution for workshop scheduling based on the input parameters, the workshop fuzzy scheduling model, and a genetic algorithm with dynamically changing crossover probability and mutation probability. The adaptive crossover probability and mutation probability designed by the present invention enable the population to dynamically adjust the global search and local search capabilities, avoid the population from falling into premature maturity, and help improve the quality of the solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 The block diagram is a shop scheduling method based on an improved multi-objective genetic algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] The embodiment of the present application provides a workshop scheduling method and system based on an improved multi-objective genetic algorithm, which solves the technical problem that the existing workshop scheduling method is prone to fall into a local optimal solution during the solution process, thereby avoiding the population from falling into premature maturity and improving the quality of the solution.

[0071] The technical solution in the embodiment of the present application is to solve the above technical problems, and the overall idea is as follows:

[0072] The workshop scheduling problem is a multi-objective scheduling problem. In the prior art, such multi-objective scheduling problems are generally solved by traditional genetic algorithms. However, the prior art has the following problems: 1. In the workshop scheduling process, it is usually considered that the workpiece has a clear processing time. But in the real world, the processing time of the workpiece is not always known. 2. When solving the multi-objective workshop scheduling problem, the traditional genetic algorithm usually sets the chromosome crossover probability and mutation probability as fixed values, which easily leads to the population being stagnant. 3. When performing the crossover operation, the traditional genetic algorithm often randomly selects two parent chromosomes from the population for crossover to obtain new offspring chromosomes. This method does not fully consider the differences between individuals, resulting in poor quality of the offspring chromosomes. 4. When performing the mutation operation, the traditional genetic algorithm often only randomly transforms certain gene fragments, but does not make full use of the excellent gene fragments of elite individuals. 5. When performing the mutation operation, the traditional genetic algorithm does not make full use of the relevant historical information, resulting in repeated poor solutions for some mutations. To solve the above problems, an embodiment of the present invention provides a workshop scheduling method and system based on an improved multi-objective genetic algorithm, which improves the multi-objective genetic algorithm for solving the multi-objective workshop fuzzy scheduling problem and improves the efficiency of the solution process and the quality of the solution.

[0073] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0074] The embodiment of the present invention provides a workshop scheduling method based on an improved multi-objective genetic algorithm, such as Figure 1 As shown, including:

[0075] S1. Obtain task data for workshop scheduling and set input parameters according to the task data;

[0076] S2, obtaining a pre-built workshop fuzzy scheduling model;

[0077] S3. Based on the input parameters, the workshop fuzzy scheduling model, and the genetic algorithm with dynamically changing crossover probability and mutation probability, an optimal scheduling solution for workshop scheduling is obtained.

[0078] The adaptive crossover probability and mutation probability designed in the embodiment of the present invention enable the population to dynamically adjust the global search and local search capabilities, avoid the population from falling into premature maturity, and help improve the quality of the solution.

[0079] The following is a detailed description of each step:

[0080] In step S1, the task data of the workshop scheduling is obtained, and the input parameters are set according to the task data. The specific implementation process is as follows:

[0081] The task data of workshop scheduling includes the number of machines, the number of workpieces, the processing speed of each machine, the unit processing cost of each machine, etc.

[0082] The input parameters include:

[0083] Number of machines m; Number of workpieces n; Fuzzy completion time of workpiece j Fuzzy completion time of machine i Processing speed v of machine i i ; Maximum fuzzy completion time Unit processing cost of machine i β i ;Total processing cost of workpiece

[0084] In step S2, a pre-built workshop fuzzy scheduling model is obtained. The specific implementation process is as follows:

[0085] The shop floor fuzzy scheduling model includes objective function and constraints.

[0086] The objective function includes:

[0087]

[0088]

[0089] in:

[0090] The objective function (1) represents minimizing the maximum fuzzy completion time;

[0091] The objective function (2) represents minimizing the total fuzzy processing cost.

[0092] The constraints include:

[0093]

[0094]

[0095]

[0096]

[0097] x ij ∈{0,1},i=1,...,m,j=1,...,n (7)

[0098] Constraint (3) is used to calculate the fuzzy completion time on each machine;

[0099] Constraint (4) is used to calculate the total fuzzy processing cost, which is equal to the cumulative sum of all machine processing costs;

[0100] Constraint (5) represents the maximum fuzzy completion time of the system, which is equal to the maximum completion time of all machines;

[0101] Constraint (6) indicates that each workpiece can only and must be assigned to any one machine for processing;

[0102] Constraint (7) indicates that if job j is assigned to machine i for processing, then x ij =1, otherwise 0.

[0103] In step S3, based on the input parameters, the shop fuzzy scheduling model and the improved genetic algorithm, the optimal scheduling solution for shop scheduling is obtained. The specific implementation process is as follows:

[0104] S301. Encode the workshop work plan using integer coding, that is, each gene on the chromosome represents a workpiece, and the value of each gene bit is limited to [1, m], representing the processing machine to which the corresponding workpiece is assigned. For example, for the scheduling problem of 4 workpieces and 2 machines, the chromosome (2, 1, 1, 2) means that workpieces 1 and 4 are assigned to machine 2 for processing, and workpieces 2 and 3 are assigned to machine 1 for processing.

[0105] S302, randomly generate NP parent chromosomes (schemes), each of which has a length (dimension) of n and corresponds to a corresponding workpiece. The value on the jth (1≤j≤n) dimension represents the processing machine to which the workpiece is assigned, and determine the current number of iterations T and the maximum number of iterations T max And individual initial crossover probability p3.

[0106] S303: Fuzzy processing time for workpiece j The processing time after defuzzification is Calculate the completion time and cost of each chromosome The corresponding target value.

[0107] S304, sort the two target values ​​from small to large, perform specific search and update on the parent chromosome, and perform fast non-dominated sorting on the newly generated chromosome. Specifically, it includes:

[0108] The specific search includes: sorting the two targets in the parent chromosomes, selecting the first k1 chromosomes, retaining the common information (excellent information fragments) that appear in the k1 chromosomes, and using them to improve the k2 chromosomes that are ranked later. If these information fragments do not exist or have been operated, then the differential mutation operation is used to update the chromosome information.

[0109] At the same time, during the search process, a short-term loop memory table - taboo table is constructed to store all chromosome types that have appeared in recent rounds of iterative operations and the N1 crossover and mutation operations that have been performed on the same chromosome recently. This movement is a neighborhood change of the solution. The movement in the taboo table is called a taboo move. In order to avoid returning to the original solution (possibly by accepting an inferior solution), thereby avoiding falling into a loop. The taboo is not lifted until after N2 rounds. The introduction of the taboo table avoids repeated calculations of the same solution and speeds up the search speed of the algorithm. Among them, N1 and N2 are artificially defined values.

[0110] S305, determining the optimal classification number k of chromosomes by using Canopy algorithm * , calculate the Hamming distance between chromosomes, cluster the populations with high similarity through the K-means clustering algorithm, and generate multiple cluster groups.

[0111] S306. Generate random numbers r1∈(0,1) and r3∈(0,1). If r1<p1, then randomly generate two integers A1∈(0,NP) and A2∈(0,NP), A1≠A2 in the same clustering group, and perform same-group crossover on the two chromosomes. If r1>p1, then randomly generate two integers A3∈(0,NP) and A4∈(0,NP), A3≠A4 in different groups, and perform different-group crossover on the two chromosomes. Note that the selected chromosome must satisfy r3≤p3, otherwise reselect the chromosome to generate a new chromosome.

[0112] It should be noted that in the embodiment of the present invention, the genetic algorithm tends to use heterogeneous crossover to adjust the structure of the solution on a large scale in the early stage of execution, in order to obtain a better solution through global search, and heterogeneous crossover should be performed with a greater probability to find the global optimal solution; in the later stage of the algorithm execution, the quality of the solution is relatively high, and the same-group crossover can effectively improve the quality of the solution without destroying the structure of the solution. To this end, an adaptive crossover operator (related to the number of algorithm iterations) is designed to determine whether to perform same-group or heterogeneous crossover. The operator is expressed as follows:

[0113] p1=1-e -kx

[0114] Where x represents the number of iterations; k represents the conventional coefficient, k>0. Generate a random number r1 between 0 and 1. If r1<p1, perform same-group crossover; otherwise, perform different-group crossover.

[0115] If the crossover probability of all individuals is the same, the differences between individuals will be ignored, resulting in the loss of information about individuals with better quality. Therefore, a corresponding crossover probability is designed for each individual. During the algorithm evolution process, if an individual has not been eliminated after many generations, it indicates that its quality is good. The number of generations that the individual exists is called the survival length.

[0116]

[0117] Among them, p3 is the initial crossover probability of each individual, t and t max They represent the number of generations that an individual has survived and the total number of generations that the population has experienced.

[0118] Generate a random number r3. If r3≤p3, perform mutation on the chromosome.

[0119] S307: For each chromosome obtained in step S306, a random number r2 is generated. If r2≤p2 is satisfied, a mutation operation is performed on the particle. Specifically, the operation includes:

[0120] In order to retain the better solution (less dominated number) as much as possible, the mutation rate p2 is set according to the domination and dominated situation of the feasible solution, and a random number r2 is generated. If r2≤p2, the chromosome is mutated. p2 is expressed as follows:

[0121] p2=n * / n max

[0122] Among them, n * Indicates the number of currently feasible solutions; n max Represents the maximum controllable number.

[0123] S308. Determine whether T≥Tmax. If not, set T=T+1 and go to S304. If satisfied, output all solutions of the non-dominated frontier (Pareto solution set) and the corresponding scheduling plan to obtain the optimal scheduling plan for workshop scheduling.

[0124] The embodiment of the present invention further provides a shop scheduling system based on an improved multi-objective genetic algorithm, comprising:

[0125] A data acquisition module is used to obtain task data for workshop scheduling and set input parameters according to the task data;

[0126] A model acquisition module is used to acquire a pre-built shop fuzzy scheduling model;

[0127] The scheme acquisition module is used to acquire the optimal scheduling scheme for workshop scheduling based on the input parameters, the workshop fuzzy scheduling model, and a genetic algorithm with dynamically changing crossover probability and mutation probability.

[0128] It can be understood that the workshop scheduling system based on the improved multi-objective genetic algorithm provided in the embodiment of the present invention corresponds to the above-mentioned workshop scheduling method based on the improved multi-objective genetic algorithm. The explanations, examples, beneficial effects and other parts of the relevant contents can refer to the corresponding contents in the workshop scheduling method based on the improved multi-objective genetic algorithm, and will not be repeated here.

[0129] An embodiment of the present invention also provides a computer-readable storage medium storing a computer program for shop scheduling based on an improved multi-objective genetic algorithm, wherein the computer program enables a computer to execute the shop scheduling method based on the improved multi-objective genetic algorithm as described above.

[0130] An embodiment of the present invention further provides an electronic device, including:

[0131] one or more processors;

[0132] Memory; and

[0133] One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs including instructions for executing the shop scheduling method based on the improved multi-objective genetic algorithm as described above.

[0134] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0135] 1. The adaptive crossover probability and mutation probability designed in the embodiment of the present invention enable the population to dynamically adjust the global search and local search capabilities, avoid the population from falling into premature maturity, and help improve the quality of the solution.

[0136] 2. The embodiment of the present invention performs fuzzy processing on the situation of uncertain workpiece processing time that is prevalent in workshop scheduling, and establishes a mathematical model so that the problem can better reflect the actual production environment.

[0137] 3. The clustering method designed in the embodiment of the present invention is helpful to classify the population and improve the effect of the crossover operation.

[0138] 4. The specific search mechanism designed in the embodiment of the present invention helps to make full use of the excellent gene fragments of elite individuals in the population and improve the quality of the variation solution.

[0139] 5. The taboo search list designed in the embodiment of the present invention can avoid repeated generation of solutions with poor quality and repeated calculations, thereby improving the execution efficiency of the algorithm.

[0140] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0141] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A shop scheduling method based on an improved multi-objective genetic algorithm, characterized in that: Including: S1. Obtain the task data of the workshop scheduling, and set the input parameters according to the task data; S2. Obtain the pre-constructed workshop fuzzy scheduling model; S3. Based on the input parameters, the workshop fuzzy scheduling model, and the genetic algorithm with dynamically changing crossover probability and mutation probability, obtain the optimal scheduling plan for the workshop scheduling; Wherein, the pre-constructed workshop fuzzy scheduling model includes an objective function and constraint conditions, The objective function includes: Wherein, Objective function (1) represents minimizing the maximum fuzzy completion time; Objective function (2) represents minimizing the total fuzzy processing cost; is the maximum fuzzy completion time; is the total processing cost of the workpiece; The genetic algorithm with dynamically changing crossover probability and mutation probability includes: S301. Encode the workshop work plan using integer encoding; S302, randomly generate NP parent chromosomes, each of which has a length of n and corresponds to a corresponding workpiece. The value on the jth dimension represents the processing machine to which the workpiece is assigned. Determine the current number of iterations T and the maximum number of iterations T max and the individual initial crossover probability p3, where n is the number of workpieces, 1≤j≤n; S303: Fuzzy processing time for workpiece j The processing time after defuzzification is Calculate the target value of each chromosome on the target function (1) and the target function (2); S304. Sort the two objective values from small to large respectively, perform specific search and update on the parental chromosomes, and perform fast non-dominated sorting on the newly generated chromosomes; wherein, the specific search includes: Among the parental chromosomes, sort the two objectives respectively, select the top k1 chromosomes, retain the more common information that appears in these k1 chromosomes, and use it to improve the k2 chromosomes with lower rankings; if there is no common information or this process has been executed, use differential mutation operation to update the chromosome information; S305, determining the optimal classification number k of chromosomes by using Canopy algorithm * , calculate the Hamming distance between chromosomes, cluster the populations with high similarity through the K-means clustering algorithm, and generate multiple cluster groups; S306. Generate random numbers r1∈(0,1), r3∈(0,1). If r1 < p1, randomly generate two integers A1∈(0,NP), A2∈(0,NP) in the same clustering group, A1≠A2, and perform in-group crossover on these two chromosomes; if r1 > p1, randomly generate two integers A3∈(0,NP), A4∈(0,NP) in different groups, A3≠A4, and perform out-group crossover on these two chromosomes to generate new chromosomes; in the crossover process, select chromosomes that satisfy r3≤p3; p1=1-e -kx Wherein, x represents the number of iterations; k represents a conventional coefficient, k > 0; S307. For each chromosome obtained in step S306, generate a random number r2. If it satisfies r2≤p2, perform mutation operation on this particle; p2=n * / n max Among them, n * Indicates the number of currently feasible solutions; n max represents the maximum disposable number; S308. Judge whether T≥Tmax. If not satisfied, let T = T + 1, and go to S304; if satisfied, output all solutions of the non-dominated front and the corresponding scheduling plan to obtain the optimal scheduling plan for the workshop scheduling; Wherein, S306 further includes: If the crossover probabilities of all individuals are the same, design corresponding crossover probability p3 for each individual, generate a random number r3. If r3≤p3, perform mutation operation on this chromosome; Among them, p3 is the initial crossover probability of each individual, t and t max They represent the number of generations that an individual has survived and the total number of generations that the population has experienced.

2. The shop scheduling method based on the improved multi-objective genetic algorithm according to claim 1, characterized in that: The constraint conditions include: x ij ∈{0,1},i=1,…,m,j=1,…,n (7) Wherein, Constraint (3) is used to calculate the fuzzy completion time on each machine; Constraint (4) is used to calculate the total fuzzy processing cost, and this total fuzzy processing cost is equal to the sum of the processing costs of all machines; Constraint (5) represents the system maximum fuzzy completion time, and this maximum fuzzy completion time is equal to the maximum completion time among all machines; Constraint (6) represents that each workpiece can and must be assigned to any one machine for processing; Constraint (7) indicates that if job j is assigned to machine i for processing, then x ij =1, otherwise 0; m is the number of machines; n is the number of workpieces; is the fuzzy processing time of workpiece j, is the fuzzy completion time of machine i; v i is the processing speed of machine i; β i is the unit processing cost of machine i.

3. The shop scheduling method based on the improved multi-objective genetic algorithm according to claim 1, characterized in that: S304 further includes: During the search process, a tabu list is constructed to store all the chromosome types that have appeared in the recent several rounds of iterative operations and the N1 times of crossover and mutation operations performed on the same chromosome most recently, to avoid returning to the original solution until the tabu is lifted after N2 rounds; N1 and N2 are artificially defined values.

4. A shop scheduling system based on an improved multi-objective genetic algorithm, characterized in that: Including: A data acquisition module, used to acquire the task data of the workshop scheduling and set the input parameters according to the task data; A model acquisition module, used to acquire the pre-constructed workshop fuzzy scheduling model; A solution acquisition module, used to obtain the optimal scheduling solution of the workshop scheduling based on the input parameters, the workshop fuzzy scheduling model, and the genetic algorithm with dynamically changing crossover probability and mutation probability; Among them, the pre-constructed workshop fuzzy scheduling model includes an objective function and constraint conditions, The objective function includes: Among them, Objective function (1) represents minimizing the maximum fuzzy completion time; Objective function (2) represents minimizing the total fuzzy processing cost; is the maximum fuzzy completion time; is the total processing cost of the workpiece; The genetic algorithm with dynamically changing crossover probability and mutation probability includes: S301. Encode the workshop work plan using integer encoding; S302, randomly generate NP parent chromosomes, each of which has a length of n and corresponds to a corresponding workpiece. The value on the jth dimension represents the processing machine to which the workpiece is assigned. Determine the current number of iterations T and the maximum number of iterations T max and the individual initial crossover probability p3, where n is the number of workpieces, 1≤j≤n; S303: Fuzzy processing time for workpiece j The processing time after defuzzification is Calculate the target value of each chromosome on the target function (1) and the target function (2); S304. Sort the two objective values from small to large respectively, perform specific search and update on the parent chromosomes, and perform fast non-dominated sorting on the newly generated chromosomes; among them, the specific search includes: Among the parent chromosomes, sort for the two objectives respectively, select the top k1 chromosomes, retain the more common information that appears in these k1 chromosomes, and use it to improve the k2 chromosomes with a lower ranking; if there is no common information or this process has been executed, use differential mutation operation to update the chromosome information; S305, determining the optimal classification number k of chromosomes by using Canopy algorithm * , calculate the Hamming distance between chromosomes, cluster the populations with high similarity through the K-means clustering algorithm, and generate multiple cluster groups; S306. Generate random numbers r1 ∈ (0, 1), r3 ∈ (0, 1). If r1 < p1, randomly generate two integers A1 ∈ (0, NP), A2 ∈ (0, NP) in the same clustering group, A1 ≠ A2, and perform in-group crossover on these two chromosomes; if r1 > p1, randomly generate two integers A3 ∈ (0, NP), A4 ∈ (0, NP) in different groups, A3 ≠ A4, and perform out-group crossover on these two chromosomes to generate new chromosomes; during the crossover process, select chromosomes that satisfy r3 ≤ p3; p1=1-e -kx Among them, x represents the number of iterations; k represents a conventional coefficient, k > 0; S307. For each chromosome obtained in step S306, generate a random number r2. If it is satisfied that r2 ≤ p2, perform mutation operation on this particle; p2=n * / n max Among them, n * Indicates the number of currently feasible solutions; n max represents the maximum disposable number; S308. Judge whether T ≥ Tmax. If not satisfied, let T = T + 1, and go to S304; if satisfied, output all the solutions of the non-dominated front and the corresponding scheduling solutions to obtain the optimal scheduling solution of the workshop scheduling; Among them, S306 also includes: If the crossover probabilities of all individuals are the same, design corresponding crossover probability p3 for each individual, generate a random number r3, and if r3 ≤ p3, perform mutation operation on this chromosome; Among them, p3 is the initial crossover probability of each individual, t and t max They represent the number of generations that an individual has survived and the total number of generations that the population has experienced.

5. A computer-readable storage medium, characterized in that: It stores a computer program for shop scheduling based on an improved multi-objective genetic algorithm, wherein the computer program enables a computer to execute the shop scheduling method based on an improved multi-objective genetic algorithm as claimed in any one of claims 1 to 3.

6. An electronic device, characterized in that: include: one or more processors; Memory; as well as One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include programs for executing the shop scheduling method based on the improved multi-objective genetic algorithm as described in any one of claims 1 to 3.

Images