CN111353604A - Flexible assembly multi-objective dynamic optimization method - Google Patents
Flexible assembly multi-objective dynamic optimization method Download PDFInfo
- Publication number
- CN111353604A CN111353604A CN201811584219.5A CN201811584219A CN111353604A CN 111353604 A CN111353604 A CN 111353604A CN 201811584219 A CN201811584219 A CN 201811584219A CN 111353604 A CN111353604 A CN 111353604A
- Authority
- CN
- China
- Prior art keywords
- population
- assembly line
- individuals
- assembly
- evolution
- 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.)
- Granted
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 39
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 29
- 230000002068 genetic effect Effects 0.000 claims abstract description 12
- 238000000638 solvent extraction Methods 0.000 claims abstract description 8
- 230000035772 mutation Effects 0.000 claims description 20
- 238000004519 manufacturing process Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 14
- 238000009826 distribution Methods 0.000 claims description 5
- 206010064571 Gene mutation Diseases 0.000 claims description 2
- 230000003203 everyday effect Effects 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims 1
- 239000000047 product Substances 0.000 description 22
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 238000004026 adhesive bonding Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 230000003139 buffering effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000004593 Epoxy Substances 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 239000011265 semifinished product Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
- H01F41/00—Apparatus or processes specially adapted for manufacturing or assembling magnets, inductances or transformers; Apparatus or processes specially adapted for manufacturing materials characterised by their magnetic properties
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides a flexible assembly multi-objective dynamic optimization method, which comprises the following steps: constructing a flexible assembly line multi-target dynamic optimization model: the model comprises the definition of a flexible assembly line, the precondition of multi-objective optimization of the assembly line, the parameter definition of a multi-objective dynamic optimization function, the multi-objective optimization function and the constraint condition of function parameters; designing a genetic algorithm based on population iterative partitioning: through double-layer gene coding, a fitness function taking product competitiveness as an index is designed, an elite selection strategy is adopted for a selection operator, staged crossing is designed for a crossing operator, spaced individuals are selected for random crossing at the early stage of evolution, excellent individual crossing is selected at the later stage of evolution, and different variation probabilities are set for the variation operators at different evolution stages, so that global optimization and rapid convergence of an algorithm are realized together; and solving the optimal result of the flexible assembly line planning according to the demand target by utilizing the designed genetic algorithm based on population iterative division.
Description
Technical Field
The invention relates to an assembly line optimization design technology, in particular to a flexible assembly multi-objective dynamic optimization method.
Background
In recent years, with the progress and popularization of information technologies such as industrial robots, machine vision, internet of things and the like, flexible assembly systems oriented to various and variable mass production modes are increasingly applied. Meanwhile, the development of personalized demands drives the subdivision degree of industrial products to be further enhanced, and the pursuit of high cost performance and high quality consistency of products and the demand of market dynamics on quick response of assembly production processes all require that multi-objective optimization problems such as product assembly quality, assembly line productivity, assembly line manufacturing cost and reusability of the flexible assembly system must be considered at the beginning of design.
With the changes of an assembly line, a real environment and the like, the flexible assembly line needs to coordinate different workstations to complete the assembly of parts and also needs to deal with the influences of environmental factors, human demand changes and the like. For example, in some assembly lines with workstation revisiting, the assembly of the initial assembly body is completed, the detection workstation is transferred, the detection is unqualified, and the revising workstation is required to adjust, so that the assembly beat is not a fixed value. Even sometimes, a certain workstation of the assembly line breaks down to cause the change of the assembly rhythm, and the assembly capacity of the assembly line fluctuates and the cost bearing capacity of enterprises changes and other factors along with the time, so the multi-objective optimization of the assembly line is a dynamic optimization process.
Disclosure of Invention
The invention aims to provide a flexible assembly multi-objective dynamic optimization method, which comprises the following steps:
constructing a flexible assembly line multi-target dynamic optimization model: the model comprises the definition of a flexible assembly line of the transformer, the precondition of multi-objective optimization of the assembly line, the parameter definition of a multi-objective dynamic optimization function, the multi-objective optimization function and the constraint condition of function parameters;
designing a genetic algorithm based on population iterative partitioning: through double-layer gene coding, a fitness function taking product competitiveness as an index is designed, an elite selection strategy is adopted for a selection operator, staged crossing is designed for a crossing operator, spaced individuals are selected for random crossing at the early stage of evolution, excellent individual crossing is selected at the later stage of evolution, and different variation probabilities are set for the variation operators at different evolution stages, so that global optimization and rapid convergence of an algorithm are realized together;
and solving the optimal result of the flexible assembly line planning according to the demand target by utilizing the designed genetic algorithm based on population iterative division.
Compared with the prior art, the invention has the following advantages: (1) the method is used for constructing a flexible assembly line multi-target dynamic optimization model and designing genetic algorithm solution based on population iterative partitioning, is suitable for multi-target optimization design of equipment, and has universality. The application of the method can solve the problems possibly occurring in the running process of the equipment in advance, save the enterprise cost and improve the quality of the equipment; (2) a dynamic optimization method based on working condition change is designed, and the method can be used for minimizing the influence on equipment production and improving the flexibility of the equipment while performing resource integration on the conventional assembly line. (3) The invention enhances the flexibility of the assembly equipment, improves the stability of the assembly quality of products and the response speed of enterprises to market changes, reduces the equipment cost and maximizes the utilization rate of the assembly line.
The invention is further described below with reference to the accompanying drawings.
Drawings
FIG. 1 is a flexible assembly line layout.
Fig. 2 is an algorithm work flow diagram.
FIG. 3 is a gene string coding diagram.
FIG. 4 is a gene string decoding diagram.
Detailed Description
In connection with fig. 1, the present invention optimizes the design of a flexible assembly line. Taking a flexible assembly line of a transformer as an example, an assembly workstation with five functions is designed, and the assembly workstation mainly comprises five workstations, namely an assembly workstation, a correction and carrying workstation, a gluing workstation, an auxiliary debugging workstation and a pressurization and locking test workstation. The assembly workstation mainly completes initial assembly of a product assembly and comprises seven working procedures of iron core pushing, center column placing, coil placing, side column sleeve coil, insulating paper feeding and clamping, epoxy plate mounting and iron core press-fitting; the correcting and carrying work station mainly corrects and positions the initial assembly body, turns over the semi-finished product and separates and places the top iron core; the gluing workstation is responsible for gluing the single side of the split top iron core, namely gluing the single side of the amorphous iron core; the auxiliary debugging workstation is responsible for gluing the surfaces of the three columns in the other part after the splitting and placing air gap paper; and the pressurization locking test workstation is responsible for the pressurization locking of the whole assembly after gluing and the measurement and adjustment of inductance.
After the first work station finishes the initial assembly of the product assembly, the initial assembly is conveyed to the second work station through the conveyor, the second work station corrects and positions the initial assembly through the positioning fixture, the top iron core is disassembled and is distributed to the third work station, the third work station finishes the gluing work of the top iron core, the rest parts are conveyed to the fourth work station, the fourth work station performs three-column surface gluing on the assembly and places air gap paper, after the fourth work station finishes, the third work station can retreat to the third work station by means of intelligent control of the conveyor, the top iron core which is coated with glue is installed by the third work station, finally the assembly integrally reaches the fifth work station, the whole product assembly is pressurized, tested and adjusted, and the blanking is locked after the test is finished.
With reference to fig. 2, a flexible assembly line multi-objective dynamic optimization method includes constructing a flexible assembly line multi-objective dynamic optimization model, designing a genetic algorithm based on population iterative partitioning, and designing a dynamic optimization method based on working condition changes.
Constructing a flexible assembly line multi-target dynamic optimization model: the model takes the arrangement of workstations, the setting of a buffer area and the assembly speed as variables, and simultaneously considers the constraint of a work flow, the constraint of equipment cost and the constraint of the assembly speed. The method comprises the steps of defining a flexible assembly line, the precondition of multi-objective optimization of the assembly line, the parameter definition of a multi-objective dynamic optimization function, the multi-objective optimization function and the constraint condition of function parameters;
the method for the flexible assembly line multi-target dynamic optimization model comprises the following specific steps:
step S101, defining a flexible assembly line, and determining the station of the assembly line and the type of a product to be assembled;
step S102, determining the precondition of product assembly, including the distribution of the process on the workstation, the assembly limit speed of the process, the single value of the product, the manufacturing cost of each workstation, the upper limit of the input cost of the assembly line, the time of the assembly line put into production every day, and the demand condition of the market for the assembly production product;
step S103, defining parameters of a multi-objective optimization function, taking transformer assembly as an example, and determining decision variables of the model in the parameters, wherein the table 1 is a multi-objective dynamic optimization model parameter definition table of a flexible assembly line of the transformer;
TABLE 1
Step S104, writing a multi-objective optimization function of the flexible assembly line, including an assembly line productivity maximization function, an assembly line manufacturing cost minimization function and a product assembly quality optimization function
α＝60Tdmax{n1S1,n2S2,...,njSj} (1)
Ck＝min(sum(njcj)) (2)
Step S105, writing a parameter constraint equation in the multi-objective optimization function in a column, including
Ai＝[a1a2... ai](4)
Sj＜AiAm(6)
Vj-1＝NjSj(9)
n1S1＝n2S2＝n3S3＝...＝njSj(10)
Ck＜Cm(11)
In steps 104 and 105, formula (1) represents that the assembly line capacity is maximized, formula (2) represents that the assembly line manufacturing cost is minimized, formula (3) represents an assembly quality optimal equation of a product, formula (4) represents the assembly speed of each process, formula (5) represents the distribution of each process on the workstation, formula (6) represents the assembly speed constraint of each workstation, formula (7) represents the number arrangement of each workstation, and formula (8) represents the manufacturing cost of each workstation; equation (9) represents the amount of buffering in the buffer, equation (10) represents assembly flow constraints, the total assembly capacity of each workstation in the assembly line is equal or similar, and equation (11) represents assembly line manufacturing cost constraints.
After a multi-objective optimization model of the flexible assembly line is obtained, solving the model according to a designed genetic algorithm based on population iterative partitioning: through double-layer gene coding, a fitness function taking product competitiveness as an index is designed, an elite selection strategy is adopted for a selection operator, staged crossing is designed for a crossing operator, spaced individuals are selected for random crossing at the early stage of evolution, excellent individual crossing is selected at the later stage of evolution, and different variation probabilities are set for the variation operators at different evolution stages, so that global optimization and rapid convergence of an algorithm are realized together.
Designing a genetic algorithm based on population iterative partitioning, wherein the algorithm flow is as follows:
step S301, carrying out gene string double-layer coding on the decision variable;
step S302, initializing population and setting evolution algebra TmaxInitially, T is 0;
step S303, calculating the fitness of individuals in the population;
step S304, selecting excellent individuals from the population in step S303 by using an elite selection strategy;
step S305, performing cross operation on the individuals obtained after the operator operation is selected in the step S304, and performing gene block single-point cross on the individuals in the population through population iteration times and fitness value division;
step S306, carrying out mutation operation on the individuals obtained after the operation of the cross operator in the step S305, and carrying out gene mutation on the individuals in the population by setting the mutation probability of iterative division of the population to obtain brand-new individuals and populations;
step S307, adding 1 to the algebra T of the evolution iteration, that is, T ═ T + 1;
step S308, determining T and TmaxIf T is<TmaxSkipping to step S303, otherwise skipping to the next step;
and step S309, finishing the algorithm, and decoding the obtained individual gene string by taking the transformer model as an example.
In step S301, taking a transformer model as an example, fig. 3 is a decision variable coding diagram of the transformer model, and a new coding mode, namely, a two-layer coding, is adopted for the gene string coding, wherein the first layer represents the assembly speed of the workstation and is called an assembly speed code, the second layer represents the number distribution code of the workstation and is called a number code, the assembly capacity coding length is J × X, J represents the number of the types of the workstation and is 2XNumber code length is J × Y, J represents the number of workstation types, 2YThe representation can contain the smallest binary number of a single maximum workstation.
FIG. 4 is a diagram of a transformer model decision variable decoding, which can be used to obtain the assembly capability and number of the first and second workstations according to the above gene string codes, such as the decoding of FIG. 2.
In step 302, the assembling speed and the number of the workstations are set, and it is required to satisfy that the assembling beats of the workstations are similar or equal, and the constraint conditions of the cost and the assembling speed are satisfied. The initialization population consists of M individuals, which are the codes for a set of possible solutions. The idea of generating valid individual gene codes is to first randomly generate a solution that satisfies the constraints and then encode the feasible solution. The method comprises the following specific steps:
in step 3021, the individual counter count is set to 1.
Step 3022, randomly generating a feasible solution:
workstation assembly capacity and number distribution aggregation:
SN＝{s1n1,s2n2,s3n3,...,sjnj}
sjis a matrix SjElement (1) of
Determining each s based on job balance and cost constraints, and workstation assembly capacity limitsiniA range value of (d);
at each siniWithin the range of values of (a), to which a numerical value is arbitrarily assigned;
when i is j, finishing taking the SN value;
step 3023, encoding the feasible solution generated in step 2.
In step 3024, count +1, if count is less than or equal to M, go to step 3202, otherwise go to step 3205.
Step 3025, population initialization is complete.
In step S303, aiming at the characteristics of multi-objective dynamic optimization of the assembly line, a new concept of product competitiveness is proposed by integrating the optimization objective function, where the product competitiveness is (product daily output × product assembly quality)/product assembly cost, that is, product competitiveness
By introducing novel product competitiveness definition and unifying dimensions, the balance among various indexes of multi-objective dynamic optimization is realized. Therefore, the fitness function is obtained:
in the formula, a, b and Td、CjAre all known quantities.
In step 303, the selection operator adopts an optimized elite selection strategy to ensure that population convergence obtains an optimal solution of the multi-objective optimization problem. And if the optimal fitness value of the next generation population is smaller than the optimal fitness value of the current population, directly copying all individuals of which the fitness values are larger than the optimal fitness value of the next generation in the current population into the next generation population. In the early stage, the individual diversity of the population is ensured, and the copied individuals randomly replace the individuals in the next generation of the population. When the evolution enters the later stage, in order to accelerate the convergence speed of the optimal solution, the copied individuals replace the worst individuals in the next generation population. Population of iteration size M for a given t generation
P(t)＝{AN1(t),AN2(t),AN3(T),...,ANM(t)}
The selection operator improvement operation steps are as follows:
step 3031, calculating the most basic fitness value of the t generation population
fmax(t)＝max{f(AN1(t)),f(AN2(t)),f(AN3(t)),...,f(ANM(t))} (13)
Step 3032, calculating the optimal fitness value of the t +1 generation population
fmax(t+1)＝max{f(AN1(t+1)),f(AN2(t+1)),f(AN3(t+1)),...,f(ANM(t+1))} (14)
Step 3033, if fmax(t)＞fmax(t +1) replication of the corresponding individuals in P (t)
{ANk'(t)}＝{ANk(t)f(ANk(t))＞fmax(t+1),ANk(t)∈P(t)} (15)
k＝1,2,...,K
Step 3034, judging the iteration algebra t and the maximum iteration algebra G, if so
By ANk' (t) random substitution to ANj(t+1)，ANk(t +1) ∈ P (t +1), j-1, 2, 3.., K, otherwise ANk' (t) replaces the next worst plurality of individuals.
In step 305, the core role of the genetic algorithm isAnd a crossover operator, wherein for gene crossover, the invention provides a population gene block single-point crossover divided based on population iteration times and fitness values. For individuals in population p (t), fitness values f (k) (1, 2, 3.., M) are calculated. In the early stage of population iteration
In the parent population, two individuals are extracted at intervals, individual gene strings are randomly separated according to the gene blocks and are sequentially crossed, and the diversity of the individuals in the early stage of evolution is ensured. And later in population iteration
And comparing the fitness values of the population individuals, and preferentially selecting the parent individuals when crossing the individuals with the fitness values larger than the average fitness value so as to ensure the rapid convergence of the later evolution.
In step 306, mutation operation mutates the gene, so as to avoid the algorithm solution from falling into local optimum and assist other operators to find out global optimum solution. Binary coding is used, the mutation operator being dependent on the probability of mutation Pc2The individual genes resulting from the crossover operation are inverted. Randomly generating a number between 0 and 1 for the gene at the locus, and judging the random number and Pc2If the random number is larger than the mutation probability, the gene on the corresponding locus is negated, then the constraint condition is judged, and the individual is updated. The mutation operation is carried out early in the whole evolution process, and the mutation probability P is taken to ensure that the individuals are prevented from falling into local optimumc2(ii) a Preventing mutation operator from damaging optimal individual at last stage of evolution, and taking mutation probability 2Pc2/3。
Designing a dynamic optimization method based on working condition change: after the optimal result of the initial state of the assembly line is obtained, the change of the assembly beat caused by the revisit of the workstation and the possible several special working conditions is considered, the change quantity of the assembly beat of the workstation caused by the revisit of the workstation or the change of the working conditions is input, and after the buffering of the buffer area is averaged, the dynamic optimization of the assembly line under the condition that the assembly quality and the output of the assembly line are not greatly influenced is realized.
Claims (6)
1. A flexible assembly multi-objective dynamic optimization method is characterized by comprising the following steps:
constructing a flexible assembly line multi-target dynamic optimization model: the model comprises the definition of a flexible assembly line, the precondition of multi-objective optimization of the assembly line, the parameter definition of a multi-objective dynamic optimization function, the multi-objective optimization function and the constraint condition of function parameters;
designing a genetic algorithm based on population iterative partitioning: through double-layer gene coding, a fitness function taking product competitiveness as an index is designed, an elite selection strategy is adopted for a selection operator, staged crossing is designed for a crossing operator, spaced individuals are selected for random crossing at the early stage of evolution, excellent individual crossing is selected at the later stage of evolution, and different variation probabilities are set for the variation operators at different evolution stages, so that global optimization and rapid convergence of an algorithm are realized together;
and solving the optimal result of the flexible assembly line planning according to the demand target by utilizing the designed genetic algorithm based on population iterative division.
2. The method as claimed in claim 1, wherein the flexible assembly line multi-objective dynamic optimization model comprises the following specific steps:
step S101, determining the stations of an assembly line and the types of products of the same series and different models which can be assembled according to the process requirements of the flexible assembly line of the transformer for assembling the products;
step S102, determining the precondition of multi-objective optimization of the assembly line, including the distribution of the working procedure on the working station, the assembly limit speed of the working procedure, the single value of the product, the manufacturing cost of each working station, the upper limit of the input cost of the assembly line, the time of the assembly line put into production every day, and the demand condition of the market for the assembly production product;
step S103, determining decision variables including the assembly speed of each workstation, the number of each workstation and the buffer amount setting of each buffer area;
step S104, determining a multi-objective optimization function of the flexible assembly line, including
(1) Assembly line capacity maximization function
α＝60Tdmax{n1S1,n2S2,...,njSj}
Wherein, TdFor assembly line work hours per day, njIs the number of workstations j, SjReal-time assembly speed for workstation 1 to workstation j;
(2) lowest cost function of assembly line manufacturing
Ck＝min(sum(njcj))
Wherein, cjThe manufacturing cost of the jth workstation;
(3) best function of product assembly quality
Wherein a and b are constants
Step S105, determining constraint conditions including
(1) Constraint equation of assembly speed of each process
Ai＝[a1a2... ai]
Wherein, aiIs the ith procedure;
(2) constraint equations for process assignments on workstations
Wherein the content of the first and second substances,
(3) assembly speed constraint equation for each workstation
Sj＜AiAm
SjTo comprise s1,s2,...,sjA matrix of (a);
(4) equipment cost constraint equation
Ck＜Cm
CmThe highest investment cost of the whole assembly line is achieved;
(5) assembly flow constraint equation for flexible assembly
n1s1＝n2s2＝n3s3＝...＝njsj。
3. The method of claim 1, wherein the process of designing the genetic algorithm based on population iterative partitioning is as follows:
step S201, carrying out gene string double-layer coding on decision variables;
step S202, initializing population and setting evolution algebra TmaxInitially, T is 0;
step S203, calculating the fitness of individuals in the population
Wherein N isjThe number of the jth workstation;
step S204, utilizing an elite selection strategy, and directly copying all individuals of which the fitness values are greater than the next-generation optimal fitness value in the population of the step S203 into the next-generation population if the next-generation population optimal fitness value is less than the current population optimal fitness value;
step S205, performing cross operation on the individuals obtained after the operator operation is selected in the step S204, and performing gene block single-point cross on the individuals in the population through population iteration times and fitness value division;
step S206, carrying out mutation operation on the individuals obtained after the operation of the cross operator in the step S305, and carrying out gene mutation on the individuals in the population by setting the mutation probability of iterative division of the population to obtain brand-new individuals and the population;
step S207, adding 1 to the algebra T of the evolution iteration, that is, T ═ T + 1;
step S208, determining T and TmaxIf T is<TmaxSkipping to step S203, otherwise skipping to step S209;
and step S209, finishing the algorithm and decoding the obtained individual gene string.
4. The method of claim 3, wherein T ≦ 2T for step 205 early in population iterationmaxExtracting two individuals at intervals in the parent population, randomly separating individual gene strings according to gene blocks and sequentially crossing the individual gene strings; at the late stage of population iteration, i.e. T>2TmaxAnd/3, comparing the fitness values of the individuals in the population, and preferentially selecting the parent individuals when crossing for the individuals with the fitness values larger than the average fitness value.
5. The method according to claim 3, wherein in step 206, mutation operation is performed on the individuals obtained after the crossover operator operation in step S205, and the mutation operator depends on the mutation probability Pc2Inverting the individual gene generated by the cross operation, specifically:
randomly generating a number between 0 and 1 for the gene at the locus, and judging the random number and Pc2If the random number is larger than the mutation probability, the gene on the corresponding locus is negated, then the constraint condition is judged, and the individual is updated.
6. The method of claim 5, wherein the mutation operation is performed early in the evolution process, and the mutation probability P is takenc2(ii) a Preventing mutation operator from damaging optimal individual at last stage of evolution, and taking mutation probability 2Pc2/3；
And when the early stage of the evolution process is T < Tmax2/3, otherwise, the evolution process is in a late stage, wherein T is the iteration number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811584219.5A CN111353604B (en) | 2018-12-24 | 2018-12-24 | Flexible assembly multi-objective dynamic optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811584219.5A CN111353604B (en) | 2018-12-24 | 2018-12-24 | Flexible assembly multi-objective dynamic optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111353604A true CN111353604A (en) | 2020-06-30 |
| CN111353604B CN111353604B (en) | 2022-08-16 |
Family
ID=71195541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811584219.5A Active CN111353604B (en) | 2018-12-24 | 2018-12-24 | Flexible assembly multi-objective dynamic optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111353604B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112003281A (en) * | 2020-08-26 | 2020-11-27 | 广东电网有限责任公司广州供电局 | Optimal configuration method, device and equipment for dynamic voltage restorer |
| CN112091950A (en) * | 2020-08-21 | 2020-12-18 | 华南理工大学 | Robot kinematic parameter identification method based on hybrid genetic simulated annealing algorithm |
| CN112109083A (en) * | 2020-08-21 | 2020-12-22 | 华南理工大学 | Robot kinematic parameter identification method based on genetic tabu search algorithm |
| CN113609759A (en) * | 2021-07-19 | 2021-11-05 | 清华大学深圳国际研究生院 | Reconfigurable flexible assembly line layout method and device |
| CN116300763A (en) * | 2023-03-31 | 2023-06-23 | 华中科技大学 | Mixed flow shop mathematical heuristic scheduling method and system considering machine configuration |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573820A (en) * | 2014-12-31 | 2015-04-29 | 中国地质大学（武汉） | Genetic algorithm for solving project optimization problem under constraint condition |
| CN107844835A (en) * | 2017-11-03 | 2018-03-27 | 南京理工大学 | Multiple-objection optimization improved adaptive GA-IAGA based on changeable weight M TOPSIS multiple attribute decision making (MADM)s |
-
2018
- 2018-12-24 CN CN201811584219.5A patent/CN111353604B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573820A (en) * | 2014-12-31 | 2015-04-29 | 中国地质大学（武汉） | Genetic algorithm for solving project optimization problem under constraint condition |
| CN107844835A (en) * | 2017-11-03 | 2018-03-27 | 南京理工大学 | Multiple-objection optimization improved adaptive GA-IAGA based on changeable weight M TOPSIS multiple attribute decision making (MADM)s |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112091950A (en) * | 2020-08-21 | 2020-12-18 | 华南理工大学 | Robot kinematic parameter identification method based on hybrid genetic simulated annealing algorithm |
| CN112109083A (en) * | 2020-08-21 | 2020-12-22 | 华南理工大学 | Robot kinematic parameter identification method based on genetic tabu search algorithm |
| CN112003281A (en) * | 2020-08-26 | 2020-11-27 | 广东电网有限责任公司广州供电局 | Optimal configuration method, device and equipment for dynamic voltage restorer |
| CN113609759A (en) * | 2021-07-19 | 2021-11-05 | 清华大学深圳国际研究生院 | Reconfigurable flexible assembly line layout method and device |
| CN113609759B (en) * | 2021-07-19 | 2023-07-25 | 清华大学深圳国际研究生院 | Reconfigurable flexible assembly line layout method and device |
| CN116300763A (en) * | 2023-03-31 | 2023-06-23 | 华中科技大学 | Mixed flow shop mathematical heuristic scheduling method and system considering machine configuration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111353604B (en) | 2022-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353604B (en) | Flexible assembly multi-objective dynamic optimization method | |
| CN107256441B (en) | Power distribution network planning construction scheme design method based on non-dominated sorting genetic algorithm | |
| CN110598941A (en) | Bionic strategy-based dual-target scheduling method for particle swarm optimization manufacturing system | |
| CN109063892A (en) | Industry watt-hour meter prediction technique based on BP-LSSVM combination optimization model | |
| CN110570018B (en) | Planning and designing method for customized production workshop | |
| CN113052537B (en) | Logistics vehicle low-carbon route planning method based on heuristic particle swarm optimization | |
| CN116512250B (en) | Disassembling line balancing method based on man-machine cooperation | |
| CN105373845A (en) | Hybrid intelligent scheduling optimization method of manufacturing enterprise workshop | |
| CN112605988A (en) | Random multi-product robot disassembly line balance control method | |
| CN110597213A (en) | Production scheduling method and system for distributed hybrid flow shop | |
| CN113779885B (en) | Tolerance optimization method based on genetic algorithm | |
| CN115689070A (en) | Energy prediction method for optimizing BP neural network model based on imperial butterfly algorithm | |
| CN103778486A (en) | Power distribution network load predication method | |
| CN113281993B (en) | Greedy K-mean self-organizing neural network multi-robot path planning method | |
| CN114491839A (en) | Intelligent optimization method and system for machining parameters of numerical control electric spark machine tool | |
| CN110851247A (en) | Cost optimization scheduling method for constrained cloud workflow | |
| CN113408951A (en) | Optimal flexible scheduling method and system based on dynamic information accumulated lion group | |
| CN112633662A (en) | Encoding and decoding method for flexible job shop scheduling under limited transportation condition | |
| CN107092745A (en) | A kind of window lifting plate forming technology optimization method based on variance analysis | |
| CN112101612A (en) | Prediction method of town gas usage, storage medium and terminal equipment | |
| CN113779803A (en) | Machining and assembling interaction method based on dynamic tolerance | |
| CN110956320B (en) | Address selecting and distributing method of sustainable supply chain based on dynamic relaxation intelligent algorithm | |
| CN115145301A (en) | Multi-unmanned aerial vehicle task allocation method based on improved particle swarm optimization | |
| CN114648247A (en) | Remanufacturing decision-making method integrating process planning and scheduling | |
| CN114187975A (en) | Method and system for optimizing and automatically updating data model of diesel hydrogenation device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information |
Inventor after: He Boxia Inventor after: Liu Zhi Inventor after: Cai Feng Inventor after: Kong Lingcong Inventor after: Gu Shichen Inventor before: Liu Zhi Inventor before: Cai Feng Inventor before: Kong Lingcong Inventor before: Gu Shichen Inventor before: He Fuxia |
|
| CB03 | Change of inventor or designer information |