A method and device for optimizing configuration of MOEA-based extrusion production line beat on-demand control resources

Abstract

The present application relates to the technical field of extrusion production line beat optimization, more particularly to a kind of MOEA-based extrusion production line beat on-demand control resource optimization configuration method and device, comprising the following steps: S1.gene coding;S2.population initialization;S3.construct objective function: the optimization target of objective function includes balance loss rate, station loss index, smoothing index and production waste amount;S4.get fitness function by objective function and solve, if the current fitness function value obtained by solving is optimal solution, then output current population, otherwise execute step S5;S5.gene crossover and mutation, generate new population, then return to step S4;Wherein, the crossover probability and mutation probability of generating new population are calculated according to population and fitness function.This application can solve the resource optimization configuration problem of extrusion production line, can optimize the resources of production line according to actual production order demand, improve the balance rate and production efficiency of production line.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of extrusion production line beat optimization, and more particularly to a MOEA-based resource optimization configuration method and device for on-demand control of extrusion production line beats. BACKGROUND

[0002] With the rapid development of digital technology, through the use of various types of sensors, production data can be collected in real time, and through data analysis, autonomous perception, transmission and diagnosis of problems can be achieved, and full automation production can be completed.

[0003] Profile extrusion belongs to discrete manufacturing, and whether the resource configuration of the production line is reasonable will directly affect the balance and operation efficiency of the production line. The production line layout has different processes, and the processes have a sequence, and the processes are continuous. The production line also has multiple stations, and a station can be used to process one or more processes. For a production line, in actual production, there are different processing times for different batches of raw materials in the same processing process. With the increase in batches, production according to batch order often requires a longer production period.

[0004] For traditional extrusion production lines, the degree of automation is often not high, and the production workshop cannot automatically optimize the production scheduling of the batches to be processed before performing the production task. Manual production scheduling requires a lot of time and effort, which leads to the inability to effectively improve production efficiency. In addition, the resource configuration of the traditional extrusion production line is not perfect, and when a device in the production line fails and stops working, the station and process distribution in the production line can only be adjusted manually, which is time-consuming and labor-intensive, and the balance of the load of each station cannot be effectively guaranteed, which is not conducive to improving production efficiency. SUMMARY

[0005] The present application aims to overcome the shortcomings of the prior art and provide a MOEA-based resource optimization configuration method and device for on-demand control of extrusion production line beats, which can solve the problem of resource optimization configuration of the extrusion production line. According to the actual production order demand, the resources of the production line are optimized to improve the balance rate and production efficiency of the production line.

[0006] To solve the above technical problems, the technical solution adopted by the present application is:

[0007] A MOEA-based resource optimization configuration method for on-demand control of extrusion production line beats is provided, comprising the following steps:

[0008] S1. Gene coding: assuming that the production line has J processes and K stations, define the gene string after coding the process as the process gene string, and the gene string after coding the station as the station gene string, and then perform step S2;

[0009] S2. Population initialization: randomly allocate J processes to K stations to generate an initial population, and then perform step S3;

[0010] S3. Constructing the objective function f: the optimization objective of the objective function f includes balancing loss rate, station loss index, smoothing index, and production waste amount;

[0011] S4. Obtain the fitness function F by the objective function f and solve it, if the current fitness function value obtained by solving is the optimal solution, output the current population, otherwise perform step S5;

[0012] S5. Gene crossover and mutation: the process gene string in the current population is crossed in a sequential crossover manner, and the gene bits of the process gene string are randomly selected for mutation, the process gene string after crossover and mutation satisfies the process precedence constraint relationship, a new population is generated, and then step S4 is returned; wherein the crossover probability p c and the mutation probability p m are generated according to the initial population or the current population obtained in step S2, and the fitness function F obtained in step S4.

[0013] Further, in step S3, the objective function f is:

[0014] f(w1,w2,w3,w4)=w1×η loss_efficiency +w2×e Loss_work +w3×SI+w4×loss

[0015]

[0016] wherein, η loss_efficiency represents the balancing loss rate, e Loss_work represents the station loss index, SI represents the smoothing index, and loss represents the production waste amount, w1, w2, w3, w4 all represent weight coefficients.

[0017] Further, the calculation formula of the balancing loss rate is:

[0018]

[0019] wherein, LB represents the production line balancing rate, CT represents the production cycle, T k represents the production time of each station, K represents the number of stations allocated by the production line, and N represents the number of production batch tasks.

[0020] Further, the calculation formula of the station loss index is:

[0021] Further, the calculation formula of the station loss index is:

[0022]

[0023] wherein K the ory represents the theoretical minimum number of stations, K represents the number of stations allocated to the production line, CT represents the production cycle time, and T k represents the production time of each station.

[0024] Further, the calculation formula of the smoothness index is as follows:

[0025]

[0026] wherein N represents the number of production batch tasks, K represents the number of stations allocated to the production line, CT represents the production cycle time, and T k represents the production time of each station.

[0027] Further, the calculation formula of the production cycle time CT is as follows:

[0028]

[0029] T k ≤ CT (k = 1, 2, …, K),

[0030] wherein K represents the number of stations in the production line, represents the effective working time of the i-th production batch at each station, and mission represents the order quantity of customer demand.

[0031] Further, the calculation formula of the production waste quantity is as follows:

[0032]

[0033]

[0034]

[0035] wherein n ij represents the function corresponding to the j-th demand length under the i-th sawing mode, x i represents the number of raw materials corresponding to the i-th sawing mode, a j represents the actual demand quantity of the demand length, L0 represents the original length of the profile, and y represents the total raw material; wherein, when the production waste quantity is a constant value, w4 = 0 is set.

[0036] Further, the step S4 specifically comprises the following steps:

[0037] S41. Calculate the fitness function F:

[0038]

[0039] In the formula, f represents the objective function, C represents a constant greater than 0, and C ≥ maxf;

[0040] S42. Determine whether the fitness function value obtained in the current solution is the optimal solution: If the change range of the fitness function value obtained in the current solution is less than the specified change threshold range, it is the optimal solution and the current population is output; otherwise, execute step S5.

[0041] Further, in step S5, the crossover probability p c and the mutation probability p m are calculated by the following formulas respectively:

[0042]

[0043]

[0044] In the formula, F max and F min represent the maximum and minimum fitness values in the current population respectively; F avg represents the average fitness value of the expected population, and the value range of F avg is [0.85, 0.95]; k1, k2, k3, k4, k5, k6, k7, k8 all represent constants from 0 to 1, and satisfy k3 < k4 and k7 < k8; F c represents the larger fitness value of the two parents when they are about to perform the crossover operation, and F m represents the larger fitness value of the two parents when they are about to perform the mutation operation.

[0045] The present invention also includes a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above method are implemented.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] The present invention is a method and device for optimizing the resource allocation of the extrusion production line rhythm on demand based on MOEA, which can solve the problem of optimizing the resource allocation of the extrusion production line, optimize the resources of the production line according to the actual production order requirements, and control the production rhythm on demand; the present invention selects the balance loss rate, station loss index, smoothness index and production waste volume of the production line as the optimization objectives, and obtains the optimal corresponding arrangement strategy of processes and stations through an improved genetic algorithm, which can help the staff to be more convenient and efficient during the production operation of the extrusion production line and improve the production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1This is a flowchart of a method for optimizing resource allocation based on MOEA for on-demand control of cycle time in an extrusion production line, according to the present invention. Detailed Implementation

[0049] The present invention will be further described below with reference to specific embodiments. The accompanying drawings are for illustrative purposes only, representing schematic diagrams rather than actual physical objects, and should not be construed as limiting the scope of this patent. To better illustrate the embodiments of the present invention, some components in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0050] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0051] Example 1

[0052] like Figure 1 The figure shows a first embodiment of the resource optimization allocation method for on-demand control of extrusion production line cycle time based on MOEA of the present invention, which includes the following steps:

[0053] S1. Gene Encoding: Suppose there are J processes and K workstations in the production line. Define the gene string after process encoding as the process gene string and the gene string after workstation encoding as the workstation gene string. Then execute step S2. It should be noted that the encoding order of the gene strings follows the standard process of the production batch as the starting point.

[0054] S2. Population initialization: One workstation can correspond to one or more processes. J processes are randomly assigned to K workstations to generate an initial population, and then step S3 is executed.

[0055] S3. Construct the objective function f: The optimization objectives of the objective function f include: balance loss rate, station loss index, smoothing index, and production waste amount;

[0056] S4. Obtain the fitness function F through the objective function f and solve it. If the current fitness function value is the optimal solution, output the current population; otherwise, proceed to step S5.

[0057] S5. Gene Crossover and Mutation: Perform sequential crossover on the process gene strings in the current population, and randomly select gene positions within the process gene strings for mutation. The resulting process gene strings satisfy the sequential constraints of the processes, generating a new population. Then return to step S4; where the crossover probability p of generating the new population is... c And the mutation probability p m The fitness function F is calculated based on the initial population or current population obtained in step S2 and the fitness function F obtained in step S4.

[0058] This invention includes a resource optimization allocation method for extrusion production line cycle time control based on MOEA, which can solve the resource optimization allocation problem of extrusion production line. According to the actual production order demand, the resources of the production line are optimized and the production cycle time is controlled on demand. This invention selects the production line balance loss rate, station loss index, smoothing index and production waste amount as optimization targets, and obtains the optimal process and station correspondence strategy through an improved genetic algorithm, which can help workers to perform production operations on the extrusion production line more conveniently and efficiently, and improve production efficiency.

[0059] Example 2

[0060] This embodiment is similar to Embodiment 1, except that in step S3, considering optimization objectives such as balance loss rate, station loss index, smoothing index, and production waste amount, the objective function f is defined as follows:

[0061] f(w1,w2,w3,w4)=w1×η loss_efficiency +w2×e Loss_work +w3×SI+w4×loss

[0062]

[0063] In the formula, η loss_efficiency e represents the equilibrium loss rate. Loss_work denoted by f(x), SI represents the smoothing index, loss represents the amount of production waste, and w1, w2, w3, and w4 represent weighting coefficients. It should be noted that the objective function f is constrained by the sequence of processes on the profile extrusion production line.

[0064] Specifically, the formula for calculating the equilibrium loss rate is:

[0065]

[0066] In the formula, LB represents the production line balance rate, CT represents the production cycle time, and T represents the production line balancing rate. kThis represents the production time for each workstation, K represents the number of workstations allocated to the production line, and N represents the number of production batches. It should be noted that the production line balance rate and the production line balance loss rate are indicators used to measure the efficiency and continuity of a profile extrusion production line, respectively. The higher the production line balance rate LB, the lower the balance loss rate η. loss_efficiency The smaller the value, the more balanced the work distribution at each station on the profile extrusion production line, and the better the continuity of the production line.

[0067] Specifically, the formula for calculating the workstation loss index is as follows:

[0068]

[0069]

[0070] In the formula, K theory K represents the theoretical minimum number of workstations, CT represents the number of workstations allocated to the production line, and T represents the production cycle time. k This represents the production time for each workstation. It should be noted that the workstation loss index e... Loss_work The station loss index e can be used to evaluate the number of operations and the degree of deviation from the objective in an actual profile extrusion production line. Loss_work The smaller the value, the higher the production efficiency of the profile extrusion production line.

[0071] Specifically, the formula for calculating the smoothing exponent is:

[0072]

[0073] In the formula, N represents the number of production batches, K represents the number of workstations allocated to the production line, CT represents the production cycle time, and T... k This represents the production time for each workstation. It should be noted that the smoothing index SI is used to measure the deviation of the operation time of each workstation in the profile extrusion production line from the production cycle time in each batch, and to evaluate the dispersion of the time distribution of each workstation. If the smoothing index SI value is smaller, it means that the difference in operation time between each workstation on the profile extrusion production line is smaller, which indicates that the load of the profile extrusion production line for that batch is more balanced.

[0074] In the above formula, the formula for calculating the production cycle CT is:

[0075]

[0076] T k ≤CT(k=1,2,...,K),

[0077] In the formula, K represents the number of workstations in the production line, and T k iT represents the effective working time of the i-th production batch at each workstation, mission represents the order quantity required by the customer, and T k This represents the production time for each workstation. It's important to note that production cycle time refers to the time required to complete one product within a profile extrusion production line environment; that is, the time interval between completing two identical products. Production cycle time is a crucial indicator for evaluating production line efficiency and reflects production capacity. The operating time of each workstation must not exceed the production cycle time. Generally, the maximum production time of all workstations on the production line is selected as the boundary condition for the production cycle time.

[0078] Specifically, the formula for calculating the amount of production waste is as follows:

[0079]

[0080]

[0081]

[0082] In the formula, n ij Let x represent the function corresponding to the j-th required length under the i-th sawing method. i a represents the number of raw material pieces corresponding to the i-th sawing method. j Let L0 represent the actual required length of the profile, and y represent the total raw material quantity. Assuming that the amount of production waste is minimized and remains constant (i.e., w4 = 0), the objective function f can be transformed into:

[0083] f(w1,w2,w3)=w1×η loss_efficiency +w2×e Loss_work +w3×SI

[0084]

[0085] In the formula, η loss_efficiency e represents the equilibrium loss rate. Loss_work SI represents the workstation loss index, SI represents the smoothing index, loss represents the amount of production waste, and w1, w2, and w3 all represent weighting coefficients.

[0086] In this embodiment, step S4 specifically includes the following steps:

[0087] S41. Calculate the fitness function F:

[0088]

[0089] In the formula, f represents the objective function, C represents a constant greater than 0, and C≥maxf ensures that the fitness function value is within [0,1].

[0090] S42. Determine whether the fitness function value obtained in the current solution is the optimal solution: If the change range of the fitness function value obtained in the current solution is less than the specified change threshold range, it is the optimal solution and the current population is output; otherwise, step S5 is executed. It should be noted that the selection of the fitness function satisfies that the smaller the objective function value, the larger the fitness, so the selected optimal population is also the population with the largest fitness, and the solution to the actual problem is output. In step S42, specifically, when the change of the fitness function values calculated in the recent three times is less than 0.1%, the fitness value is the optimal solution at this time; it should be noted that this change range can be selected according to the actual scenario requirements and is not limited to 0.1%.

[0091] In this embodiment, in step S5, the crossover probability p c and the mutation probability p m are calculated as follows:

[0092]

[0093]

[0094] In the formula, F max and F min respectively represent the maximum and minimum fitness values in the current population; F avg represents the average fitness value of the expected population, and the value range of F avg is [0.85, 0.95]; k1, k2, k3, k4, k5, k6, k7, k8 all represent constants from 0 to 1, and satisfy k3 < k4 and k7 < k8; F c represents the larger fitness value of the two parents when they are about to perform the crossover operation, and F m represents the larger fitness value of the two parents when they are about to perform the mutation operation. It should be noted that the crossover probability p c and the mutation probability p m will affect the algorithm solution effect. If the parameters of the two are adjusted improperly, it is possible to directly ignore the individuals with high fitness, and the crossover probability p c and the mutation probability p m in this invention can automatically change with the fitness, and excellent individuals can be retained with a higher probability, better approaching the global optimal solution.

[0095] It should be further noted that when F c is closer to F max , will be closer to 0, and at this time the decay change of the exponential function is more gentle. When the fitness is close to F max , within a range, the crossover probability p c will be relatively low, and the crossover probability p cThe changes are not drastic, and more individuals with higher fitness are more likely to survive; the same applies to mutations: when F... m The closer to F max hour, The closer the fitness value is to 0, the more gradual the decay of the exponential function becomes. max When, within a certain range, the mutation probability p m Both will be low, and the mutation probability p m The changes were not drastic, and individuals with higher fitness were more likely to survive.

[0096] In this embodiment, step S5 specifically includes the following steps:

[0097] S51. Crossover: Randomly select the start and end positions in the parent process gene strings of any two batches, so that the two parent process gene strings are divided into three regions. Then, crossover the genetic information in the middle region to the same position in the offspring. For other genetic information in the parent, fill in the genetic information of the offspring in the order of the parent. If there is a duplicate, skip it.

[0098] S52. Mutation: Randomly select a mutation point in the process gene string in the current population, and insert the genetic information of the mutation point into a certain position with the order of the processes as a constraint. The genetic information before and after the mutation point is filled in sequentially.

[0099] It should be noted that when individuals become homogeneous, the population's fitness is already high. If the crossover and mutation probabilities are high, these superior individuals may be eliminated in the next iteration. In this invention, when the overall fitness of the population is high, the exponential decay property is utilized to reduce the crossover and mutation probabilities. Since the set average fitness is a constant, the function value can evolve in the preset direction to find better individuals.

[0100] Example 3

[0101] The present invention also includes a computer device comprising a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in Embodiment 1 or 2.

[0102] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing resource allocation based on MOEA for on-demand control of cycle time in an extrusion production line, characterized in that, Includes the following steps: S1. Gene Encoding: Suppose there are J processes and K workstations in the production line. Define the gene string after process encoding as process gene string and the gene string after workstation encoding as workstation gene string. Then execute step S2. S2. Population initialization: Randomly assign J processes to K workstations to generate an initial population, and then execute step S3; S3. Construct the objective function f: The optimization objectives of the objective function f include: the balance loss rate, the workstation loss index, the smoothing index, and the amount of production waste; wherein, the objective function f is: f(w1,w2,w3,w4)=w1×η loss_efficiency +w2×e Loss_work +w3×SI+w4×loss W T =[w1,w2,w3,w4], In the formula, η loss_efficiency e represents the equilibrium loss rate. Loss_work SI represents the workstation loss index, SI represents the smoothing index, loss represents the amount of production waste, and w1, w2, w3, and w4 all represent weighting coefficients. The formula for calculating the amount of production waste is: In the formula, n ij Let x represent the function corresponding to the j-th required length under the i-th sawing method. i a represents the number of raw material pieces corresponding to the i-th sawing method. j The actual required quantity of the required length is represented by L0, where L0 represents the original length of the profile and y represents the total amount of raw materials. S4. Obtain the fitness function F through the objective function f and solve it. If the current fitness function value is the optimal solution, output the current population; otherwise, proceed to step S5. S5. Gene Crossover and Mutation: The process gene strings in the current population are crossovered sequentially, and gene positions in the process gene strings are randomly selected for mutation. The crossover and mutated process gene strings satisfy the sequential constraints of the processes, generating a new population. Then, return to step S4; wherein, the crossover probability p of generating the new population is... c And the mutation probability p m The fitness function F is calculated based on the initial population or current population obtained in step S2 and the fitness function F obtained in step S4.

2. The method for optimizing resource allocation based on MOEA for on-demand control of extrusion production line cycle time as described in claim 1, characterized in that, The formula for calculating the equilibrium loss rate is: In the formula, LB represents the production line balance rate, CT represents the production cycle time, and T represents the production line balancing rate. k K represents the production time for each workstation, K represents the number of workstations allocated to the production line, and N represents the number of production batches.

3. The method for optimizing resource allocation for on-demand cycle time control of an extrusion production line based on MOEA as described in claim 1, characterized in that, The formula for calculating the workstation loss index is as follows: In the formula, K theory K represents the theoretical minimum number of workstations, CT represents the number of workstations allocated to the production line, and T represents the production cycle time. k This indicates the production time for each workstation.

4. The method for optimizing resource allocation based on MOEA for on-demand control of extrusion production line cycle time as described in claim 1, characterized in that, The formula for calculating the smoothing index is as follows: In the formula, N represents the number of production batches, K represents the number of workstations allocated to the production line, CT represents the production cycle time, and T... k This indicates the production time for each workstation.

5. The method for optimizing resource allocation based on MOEA for on-demand control of extrusion production line cycle time according to any one of claims 2 to 4, characterized in that, The formula for calculating the production cycle CT is: T k ≤CT(k=1,2,...,K), In the formula, K represents the number of workstations in the production line. This represents the effective working time of the i-th production batch at each workstation, and mission represents the order quantity required by the customer.

6. The method for optimizing resource allocation for on-demand cycle time control of an extrusion production line based on MOEA according to claim 1, characterized in that, Assuming the amount of production waste is a constant, let w4 = 0.

7. The method for optimizing resource allocation for on-demand cycle time control of an extrusion production line based on MOEA according to claim 1, characterized in that, Step S4 specifically includes the following steps: S41. Calculate the fitness function F: In the formula, f represents the objective function, C represents a constant greater than 0, and C≥maxf; S42. Determine if the fitness function value obtained by the current solution is the optimal solution: If the range of change of the fitness function value obtained by the current solution is less than the specified change threshold range, then it is the optimal solution and the current population is output; otherwise, proceed to step S5.

8. The method for optimizing resource allocation for on-demand cycle time control of an extrusion production line based on MOEA according to claim 1, characterized in that, In step S5, the crossover probability p c And the mutation probability p m The calculation formulas are as follows: In the formula, F max F min These represent the maximum and minimum fitness values ​​in the current population, respectively. F avg represents the average fitness value of the expected population, and F avg ranges from [0.85, 0.95]; k1, k2, k3, k4, k5, k6, k7, k8 all represent constants from 0 to 1, and satisfy k3 < k4 and k7 < k8; F c represents the larger fitness value of the two parents when they are about to perform the crossover operation, and F m represents the larger fitness value of the two parents when they are about to perform the mutation operation.

9. A computer device comprising a memory and a processor, said memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images