Productivity optimization method and system, storage medium and terminal
By establishing a capacity maximization model and combining methods of particle swarm, simulated annealing and genetic algorithms, the problems of inefficiency and processing cycle extension caused by different equipment configurations in the prior art are solved, and scientific optimization of production capacity and system automation are achieved.
Patent Information
- Application Number
- CN202311310964.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-10
- Publication Date
- 2025-04-11
AI Technical Summary
The prior art has problems of inefficiency and prolonged processing cycles when arranging the processing content of equipment of different configurations, and lacks system support and theoretical basis.
By establishing a capacity maximization model, using a combination of particle swarm algorithm, simulated annealing algorithm and genetic algorithm, initial populations are randomly generated, and the speed and position of particles are updated through the iterative process to screen out the global optimal particles to determine the parameters for maximum capacity.
It has achieved the ability to find a better combination of production capacity, provided scientific methods to achieve capacity optimization, avoid equipment efficiency losses and processing cycle extension, and improve system automation and accuracy.
Smart Images
Figure CN119808821A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of semiconductor manufacturing technology, and in particular to a production capacity optimization method and system, a storage medium and a terminal. Background Art
[0002] The semiconductor production process is complex and the plant construction cycle is long. The configuration of equipment will vary greatly with the expansion of production capacity, equipment upgrades, equipment delivery time and market demand. This results in different processing efficiency for the same process when processed on equipment with different configurations.
[0003] However, the prior art still has many problems in arranging the processing content of equipment with different configurations. Summary of the invention
[0004] The technical problem solved by the present invention is to provide a production capacity optimization method and system, a storage medium and a terminal, so as to provide a scientific method for achieving production capacity optimization.
[0005] To solve the above problems, the present invention provides a capacity optimization method, comprising: establishing a capacity maximization model based on the efficiency and equipment configuration of processing different products, the capacity maximization model including an objective function and constraints; randomly generating N initial populations according to the constraints, each of the initial populations comprising M particles, each of the particles being a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2; performing S iterations on the N initial populations to obtain N populations, wherein S is an integer greater than or equal to 2; screening out a global optimal particle from the N populations, and determining all solutions corresponding to the global optimal particle as parameters for capacity maximization.
[0006] Optionally, each iteration process of the initial population includes: using a simulated annealing algorithm to calculate the fitness value of the initial population; updating the speed and position of each particle in the initial population, running a genetic algorithm to randomly screen each particle, and performing crossover and mutation on the screened particles.
[0007] Optionally, the method of running a genetic algorithm to randomly screen each particle and performing crossover and mutation on the screened particles includes: randomly selecting some particles from the screened particles and placing them in a crossover pool, and performing crossover processing on the selected particles to obtain the next generation of particles after crossover changes; selecting the mutated next generation of particles from the next generation of particles after crossover changes, and forming a new population together with the particles that have not entered the crossover pool.
[0008] Optionally, the formula for updating the velocity and position of each particle in the initial population is:
[0009] Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k));
[0010] Xi(k+1)=Xi(k)+Vi(k+1);
[0011] Among them, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i in the kth iteration can find; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
[0012] Optionally, the objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
[0013] Optionally, the constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the range of variation of the equipment configuration.
[0014] Correspondingly, the technical solution of the present invention also provides a capacity optimization system, including: a modeling module, used to establish a capacity maximization model based on the efficiency and equipment configuration of processing different products, and the capacity maximization model includes an objective function and constraints; a particle swarm algorithm module, used to randomly generate N initial populations according to the constraints, each of the initial populations contains M particles, and each of the particles is a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2; an iterative algorithm module, which performs S iterations on the N initial populations to obtain N populations, wherein S is an integer greater than or equal to 2; an output module, which screens the global optimal particles from the N populations, and determines all solutions corresponding to the global optimal particles as parameters for capacity maximization.
[0015] Optionally, the iterative algorithm module includes: a simulated annealing algorithm module, which is used to calculate the fitness value of the initial population using a simulated annealing algorithm; a genetic algorithm module, which is used to update the speed and position of each particle in the initial population, run the genetic algorithm to randomly screen each particle, and crossover and mutate the screened particles; and a judgment module, which is used to judge whether the current number of iterations reaches a preset number of iterations S.
[0016] Optionally, the genetic algorithm module includes: a screening module, used to randomly screen each particle; a crossover processing module, used to randomly select some particles from the screened particles and put them into the crossover pool, and perform crossover processing on the selected particles to obtain the next generation of particles after crossover changes; a recombination module, used to select the mutated next generation of particles from the next generation of particles after crossover changes, and form a new population together with the particles that have not entered the crossover pool.
[0017] Optionally, the formula for updating the velocity and position of each particle in the initial population is:
[0018] Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k));
[0019] Xi(k+1)=Xi(k)+Vi(k+1);
[0020] Among them, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i in the kth iteration can find; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
[0021] Optionally, the objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
[0022] Optionally, the constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the range of variation of the equipment configuration.
[0023] Correspondingly, the technical solution of the present invention further provides a storage medium on which computer instructions are stored, characterized in that when the computer instructions are executed, the steps of the method described in any of the above technical solutions are executed.
[0024] Correspondingly, the technical solution of the present invention also provides a terminal, including a memory and a processor, wherein the memory stores computer instructions that can be run on the processor, and is characterized in that when the processor runs the computer instructions, the steps of the method described in any of the above technical solutions are executed.
[0025] Compared with the prior art, the technical solution of the present invention has the following advantages:
[0026] In the capacity optimization method of the technical solution of the present invention, N populations are obtained by performing S iterations on the N initial populations respectively; the global optimal particle is screened from the N populations, and all solutions corresponding to the global optimal particle are determined as parameters for maximizing capacity. The particle swarm method can be used to find a better combination of capacities, providing a scientific method for achieving capacity optimization.
[0027] Furthermore, each iteration of the initial population includes: using a simulated annealing algorithm to calculate the fitness value of the initial population; updating the speed and position of each particle in the initial population, running a genetic algorithm to randomly screen each particle, and performing crossover and mutation on the screened particles. By combining the genetic algorithm and the simulated annealing algorithm with the particle swarm algorithm, the shortcomings of the particle swarm itself, namely, "early optimization" and local optimization, can be effectively avoided.
[0028] Furthermore, the method of running a genetic algorithm to randomly screen each particle and performing crossover and mutation on the screened particles includes: randomly selecting some particles from the screened particles and placing them in a crossover pool; performing crossover processing on the selected particles to obtain the next generation of particles after crossover changes; selecting the mutated next generation of particles from the next generation of particles after crossover changes, and forming a new population together with the particles that have not entered the crossover pool, which can effectively improve the particles' ability to search for the global optimality.
[0029] The capacity optimization system of the technical solution of the present invention includes: a particle swarm algorithm module, which is used to randomly generate N initial populations according to the constraint conditions, each of the initial populations contains M particles, and each of the particles is a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2; an iterative algorithm module, which performs S iterations on the N initial populations to obtain N populations; an output module, which selects the global optimal particles from the N populations, and determines all solutions corresponding to the global optimal particles as parameters for capacity maximization. The particle swarm method can be used to find a better combination of capacity, providing a scientific method for achieving capacity optimization.
[0030] Furthermore, the iterative algorithm module includes: a simulated annealing algorithm module, which is used to calculate the fitness value of the initial population using the simulated annealing algorithm; a genetic algorithm module, which is used to update the speed and position of each particle in the initial population, run the genetic algorithm to randomly screen each particle, and cross and mutate the screened particles; and a judgment module, which is used to judge whether the current number of iterations reaches the preset number of iterations S. By combining the genetic algorithm and the simulated annealing algorithm with the particle swarm algorithm, the shortcomings of the particle swarm itself, such as "early optimization" and local optimization, can be effectively avoided.
[0031] Furthermore, the genetic algorithm module includes: a screening module for randomly screening each particle; a crossover processing module for randomly selecting some particles from the screened particles and placing them in a crossover pool, and performing crossover processing on the selected particles to obtain the next generation of particles after crossover changes; a recombination module for selecting the next generation of particles after mutation from the next generation of particles after crossover changes, and forming a new population together with the particles that have not entered the crossover pool, which can effectively improve the particles' ability to search for the global optimum. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 is a schematic flow chart of a method for optimizing production capacity in an embodiment of the present invention;
[0033] Figure 2 It is a structural schematic diagram of a production capacity optimization system in an embodiment of the present invention. DETAILED DESCRIPTION
[0034] As described in the background art, the prior art still has many problems in arranging the processing contents of equipment with different configurations, which will be explained in detail below.
[0035] The current dispatch method is mainly based on product arrival time and priority. The allocation of processing equipment mainly considers whether the equipment is idle and the principle of habit, which depends largely on the experience and judgment of the dispatcher and the idleness of the equipment. This method lacks system support and theoretical basis. On the one hand, it will lead to the loss of equipment efficiency. On the other hand, it will also extend the product processing cycle, reduce customer satisfaction, and may also cause additional asset expenditures when evaluating new capacity combination plans.
[0036] On this basis, the present invention provides a capacity optimization method and system, a storage medium and a terminal, wherein N populations are obtained by performing S iterations on the N initial populations, where S is an integer greater than or equal to 2; the global optimal particle is screened from the N populations, and all solutions corresponding to the global optimal particle are determined as parameters for maximizing capacity. The particle swarm method can be used to find a better combination of capacity, providing a scientific method for achieving capacity optimization.
[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0038] Figure 1 : is a flow chart of a method for optimizing production capacity in an embodiment of the present invention, comprising:
[0039] Step S101, establishing a capacity maximization model based on the efficiency and equipment configuration of processing different products, wherein the capacity maximization model includes an objective function and constraint conditions;
[0040] Step S102, randomly generating N initial populations according to the constraint conditions, each of the initial populations includes M particles, each of the particles is a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2;
[0041] Step S103, performing S iterations on the N initial populations to obtain N populations, where S is an integer greater than or equal to 2;
[0042] Step S104: a global optimal particle is screened from the N populations, and all solutions corresponding to the global optimal particle are determined as parameters for maximizing production capacity.
[0043] The following is a detailed description of each step of the capacity optimization method.
[0044] Step S101: establishing a capacity maximization model based on the efficiency of processing different products and equipment configuration, wherein the capacity maximization model includes an objective function and constraint conditions.
[0045] In this embodiment, the objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
[0046] In this embodiment, the constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the variation range of the equipment configuration.
[0047] In this embodiment, a specific example data is used for illustration, and please refer to Table 1 to Table 5 for details.
[0048] Please refer to Table 1, which shows the equipment configuration for processing different products, as follows:
[0049] equipment Recipe1 Recipe2 Recipe3 Recipe4 Recipe5 E1 1 1 0 1 1 E2 1 0 0 0 0 E3 0 0 1 1 0 E4 1 1 0 1 1 E5 0 1 0 1 1
[0050] Table 1
[0051] Among them, E1~E5 in Table 1 represent the equipment names, Recipe represents the process (here refers to the product), and whether the equipment can run the corresponding recipe constitutes an X1 matrix, 1 represents that it can run, and 0 represents that it cannot run.
[0052] Please refer to Table 2, which shows the efficiency of processing different products, as follows:
[0053]
[0054]
[0055] Table 2
[0056] Among them, E1~E5 in Table 2 represent the equipment names, Recipe represents the process, and the WPH (piece hourly output) corresponding to the equipment running Recipe constitutes the V1 matrix.
[0057] Step S102: randomly generating N initial populations according to the constraint conditions, each of the initial populations comprising M particles, each of the particles constituting a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2.
[0058] In this embodiment, the essence of the initial population is a solution set of the objective function, and the essence of each particle in the initial population is a solution of the objective function.
[0059] In this embodiment, any of the initial populations includes M particles, that is, includes M solutions. The N value and the M value can be set based on actual application needs, for example, the N value can be set to 5 and the M value can be set to 50.
[0060] It should be noted that, in this embodiment, when any particle of any of the initial populations is generated, the solution formed by the particle must satisfy the constraint conditions.
[0061] Step S103: performing S iterations on the N initial populations to obtain N populations, where S is an integer greater than or equal to 2.
[0062] In this embodiment, each iteration process of the initial population includes: using a simulated annealing algorithm to calculate the fitness value of the initial population; updating the speed and position of each particle in the initial population, running a genetic algorithm to randomly screen each particle, and performing crossover and mutation on the screened particles.
[0063] In this embodiment, by combining the genetic algorithm and the simulated annealing algorithm with the particle swarm algorithm, the shortcomings of the particle swarm itself, such as "early optimization" and local optimization, can be effectively avoided.
[0064] In this embodiment, the number of iterations S is taken as 40 as an example.
[0065] In this embodiment, the method of running a genetic algorithm to randomly screen each particle and performing crossover and mutation on the screened particles includes: randomly selecting some particles from the screened particles and placing them in a crossover pool; performing crossover processing on the selected particles to obtain the next generation of particles after crossover changes; selecting the mutated next generation of particles from the next generation of particles after crossover changes, and forming a new population together with the particles that have not entered the crossover pool, which can effectively improve the particles' ability to search for the global optimality.
[0066] In this embodiment, the formula for updating the speed and position of each particle in the initial population is:
[0067] Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k));
[0068] Xi(k+1)=Xi(k)+Vi(k+1);
[0069] Among them, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i in the kth iteration can find; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
[0070] Please refer to Table 3 to set pbest and gbest. In Table 3, it is assumed that pbest and gbest are as follows:
[0071] equipment Recipe1 Recipe2 Recipe3 Recipe4 Recipe5 E1 0 0 0 0 1 E2 1 0 0 0 0 E3 0 0 1 0 0 E4 0 1 0 1 0 E5 0 0 0 0 0
[0072] Table 3
[0073] According to Tables 1 to 3, V2 = ωV1 + c1r1 (pbest-X1) + c2r2 (gbest-X1), where ω is the inertia weight, ranging from 0.9 to 1.2, and is taken as 1 here; c1 and c2 represent the individual local learning factor and the group global learning factor, respectively, and are taken as 1 here; r1 and r2 are random numbers in the range of [0,1], and are both taken as 1 here.
[0074] By taking the above values, the formula changes to: V t+1 =V t +(pbest-X t )+(gbest-X t ), the corresponding V2 matrix is shown in Table 4, as follows:
[0075] equipment Recipe1 Recipe2 Recipe3 Recipe4 Recipe5 E1 5 12 0 18.5 9.1 E2 7.5 0 0 0 0 E3 0 0 13.1 18.6 0 E4 5.4 14.5 0 22.1 6.5 E5 0 13 0 19.2 7
[0076] Table 4
[0077] The corresponding X2 matrix is shown in Table 5, as follows:
[0078] equipment Recipe1 Recipe2 Recipe3 Recipe4 Recipe5 E1 6 13 0 19.5 10.1 E2 8.5 0 0 0 0 E3 0 0 14.1 19.6 0 E4 6.4 15.5 0 23.1 7.5 E5 0 14 0 20.2 8
[0079] Table 5
[0080] It should be noted that, in this embodiment, the calculation process of Tables 1 to 5 is only an iterative process, and the iteration continues until the set number of iterations is 40. This part can be implemented by software, where ω, c1, c2, r1 and r2 are all taken as 1 for the sake of simplicity in calculation.
[0081] Step S104: a global optimal particle is screened from the N populations, and all solutions corresponding to the global optimal particle are determined as parameters for maximizing production capacity.
[0082] In this embodiment, the particle swarm method can be used to find a better combination of production capacity, providing a scientific method for achieving optimal production capacity.
[0083] Figure 2 It is a structural schematic diagram of a production capacity optimization system in an embodiment of the present invention.
[0084] Correspondingly, the technical solution of the present invention also provides a production capacity optimization system, please refer to Figure 2 , including: a modeling module 100, used to establish a capacity maximization model based on the efficiency and equipment configuration of processing different products, and the capacity maximization model includes an objective function and constraints; a particle swarm algorithm module 101, used to randomly generate N initial populations according to the constraints, each of the initial populations includes M particles, and each of the particles is a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2; an iterative algorithm module 102, which performs S iterations on the N initial populations to obtain N populations, wherein S is an integer greater than or equal to 2; an output module 103, which selects the global optimal particle from the N populations, and determines all solutions corresponding to the global optimal particle as the parameters for capacity maximization.
[0085] In this embodiment, the particle swarm method can be used to find a better combination of production capacity, providing a scientific method for achieving optimal production capacity.
[0086] Please continue to refer to Figure 2 In this embodiment, the iterative algorithm module 102 includes: a simulated annealing algorithm module 1021, which is used to calculate the fitness value of the initial population using a simulated annealing algorithm; a genetic algorithm module 1022, which is used to update the speed and position of each particle in the initial population, run the genetic algorithm to randomly screen each particle, and cross and mutate the screened particles; a judgment module 1023, which is used to judge whether the current number of iterations reaches the preset number of iterations S. By combining the genetic algorithm and the simulated annealing algorithm with the particle swarm algorithm, the shortcomings of the "early optimization" and local optimization of the particle swarm itself can be effectively avoided.
[0087] It should be noted that, in this embodiment, when the judgment module 1023 determines that the current number of iterations has not reached the preset number of iterations S, it returns to the simulated annealing algorithm module 1021 to continue the loop processing; when the judgment module 1023 determines that the current number of iterations has reached the preset number of iterations S, it runs to the output module 103.
[0088] Please continue to refer to Figure 2 In this embodiment, the genetic algorithm module 1022 includes: a screening module 1022a, which is used to randomly screen each particle; a crossover processing module 1022b, which is used to randomly select some particles from the screened particles and put them into the crossover pool, and perform crossover processing on the selected particles to obtain the next generation of particles after crossover changes; a recombination module 1022c, which is used to select the next generation of particles after mutation from the next generation of particles after crossover changes, and form a new population together with the particles that have not entered the crossover pool, which can effectively improve the particles' ability to search for the global optimal state.
[0089] In this embodiment, the formula for updating the speed and position of each particle in the initial population is:
[0090] Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k));
[0091] Xi(k+1)=Xi(k)+Vi(k+1);
[0092] Among them, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i in the kth iteration can find; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
[0093] In this embodiment, the objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
[0094] In this embodiment, the constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the variation range of the equipment configuration.
[0095] Correspondingly, an embodiment of the present invention further provides a storage medium on which computer instructions are stored, characterized in that when the computer instructions are executed, the steps of the method described in any of the above embodiments are executed.
[0096] Correspondingly, an embodiment of the present invention further provides a terminal, including a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, and when the processor executes the computer instructions, the steps of the method described in any one of the above embodiments are executed.
[0097] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.
Claims
1. A method for optimizing production capacity, characterized in that: include: Establishing a capacity maximization model based on the efficiency and equipment configuration of processing different products, wherein the capacity maximization model includes an objective function and constraint conditions; Randomly generate N initial populations according to the constraint conditions, each of the initial populations includes M particles, each of the particles is a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2; Performing S iterations on the N initial populations to obtain N populations, where S is an integer greater than or equal to 2; A global optimal particle is obtained by screening from the N populations, and all solutions corresponding to the global optimal particle are determined as parameters for maximizing production capacity.
2. The method for optimizing production capacity according to claim 1, characterized in that: Each iteration process of the initial population includes: using a simulated annealing algorithm to calculate the fitness value of the initial population; updating the speed and position of each particle in the initial population, running a genetic algorithm to randomly screen each particle, and performing crossover and mutation on the screened particles.
3. The capacity optimization method according to claim 2, characterized in that: The method of running a genetic algorithm to randomly screen each particle and performing crossover and mutation on the screened particles includes: randomly selecting some particles from the screened particles and placing them in a crossover pool, performing crossover processing on the selected particles, and obtaining the next generation of particles after crossover changes; selecting the mutated next generation of particles from the next generation of particles after crossover changes, and forming a new population together with the particles that have not entered the crossover pool.
4. The method for optimizing production capacity according to claim 2, characterized in that: The formula for updating the velocity and position of each particle in the initial population is: Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k)); Xi(k+1)=Xi(k)+Vi(k+1); in, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i can find in the k-th iteration; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
5. The method for optimizing production capacity according to claim 1, characterized in that: The objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
6. The method for optimizing production capacity according to claim 1, characterized in that: The constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the range of variation of the equipment configuration.
7. A production capacity optimization system, characterized in that: include: A modeling module, for establishing a capacity maximization model based on the efficiency and equipment configuration of processing different products, wherein the capacity maximization model includes an objective function and constraints; A particle swarm algorithm module, used for randomly generating N initial populations according to the constraint conditions, each of the initial populations respectively comprising M particles, each of the particles being a solution to the capacity maximization model, wherein N and M are integers greater than or equal to 2 respectively; An iterative algorithm module, performing S iterations on the N initial populations to obtain N populations, where S is an integer greater than or equal to 2; The output module selects a global optimal particle from the N populations, and determines all solutions corresponding to the global optimal particle as parameters for maximizing production capacity.
8. The production capacity optimization system according to claim 7, characterized in that: The iterative algorithm module includes: a simulated annealing algorithm module, which is used to calculate the fitness value of the initial population using a simulated annealing algorithm; a genetic algorithm module, which is used to update the speed and position of each particle in the initial population, run the genetic algorithm to randomly screen each particle, and cross and mutate the screened particles; and a judgment module, which is used to judge whether the current number of iterations reaches the preset number of iterations S.
9. The production capacity optimization system according to claim 8, characterized in that: The genetic algorithm module includes: a screening module, which is used to randomly screen each particle; a crossover processing module, which is used to randomly select some particles from the screened particles and put them into the crossover pool, and perform crossover processing on the selected particles to obtain the next generation of particles after crossover changes; and a recombination module, which is used to select the next generation of particles after mutation from the next generation of particles after crossover changes, and form a new population together with the particles that have not entered the crossover pool.
10. The production capacity optimization system according to claim 8, characterized in that: The formula for updating the velocity and position of each particle in the initial population is: Vi(k+1)=ωVi(k)+c1r1(pbesti(k)-Vi(k))+c2r2(gbest(k)-Xi(k)); Xi(k+1)=Xi(k)+Vi(k+1); in, Xi(k+1) is the position vector of particle i in the k+1th iteration; Vi(k+1) is the velocity vector of particle i in the k+1th iteration; ω is the inertia weight; c1 and c2 are learning factors; pbesti(k) is the optimal solution that particle i can find in the k-th iteration; gbest(k) represents the optimal solution currently found by the entire population in the kth iteration.
11. The production capacity optimization system according to claim 7, characterized in that: The objective function of the capacity maximization model is determined according to the efficiency and equipment configuration of processing different products.
12. The production capacity optimization system according to claim 7, characterized in that: The constraint conditions of the capacity maximization model are determined according to the efficiency of the capacity and the range of variation of the equipment configuration.
13. A storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed, the steps of the method according to any one of claims 1 to 6 are executed.
14. A terminal comprising a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, characterized in that: When the processor runs the computer instructions, the steps of the method according to any one of claims 1 to 6 are performed.