Method and system for quickly solving set parameters in multi-process production process

Abstract

The invention provides a multi-process production process-oriented setting parameter rapid solving method and system, and relates to the technical field of data processing, and the method comprises the steps: constructing a multi-process description mathematical model, and determining the basic information of each process; constructing a multi-objective optimization model by taking maximization of the production efficiency and the production benefit as an objective; solving the multi-objective optimization model through an NSGA-II algorithm, and determining an optimal dynamic setting strategy; converting the multi-objective optimization model into a single-objective optimization model, and determining constraint conditions in the conversion process; an optimal dynamic setting strategy is adopted, optimal setting parameters under different input parameters are determined through a particle swarm optimization algorithm, and a training set is constructed; fusing constraint conditions in the conversion process into a neural network as physical information, and constructing a physical information neural network model; training through the training set; and determining set parameter values of the input parameters through the trained physical information neural network model.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for quickly solving setting parameters for a multi-process production process. Background Art

[0002] The multi-process production process is the core link in modern industrial manufacturing, covering multiple key steps from raw material processing to finished product assembly. This type of production process is widely used in many fields, including machinery manufacturing, aerospace, automotive industry and electronic equipment manufacturing. In the multi-process production process, a variety of complex process steps are involved, such as smelting, casting, rolling, heat treatment and surface treatment. The parameter setting of each process (such as temperature, time, pressure, etc.) will have an important impact on the final performance and quality of the product. For example, in the smelting process, improper temperature control may lead to deviations in alloy composition; in the casting process, slight changes in cooling rate may affect the grain size and distribution of the alloy; in the rolling and heat treatment processes, improper parameter setting will lead to uneven distribution of internal stress and hardness of the material. Therefore, the reasonable setting of process parameters has become the key to improving product quality and production efficiency.

[0003] However, the traditional process parameter setting method usually relies on experience. Operators adjust process parameters based on past experience or step-by-step tests. This method is not only less accurate, but also consumes a lot of manpower and time. In addition, due to the reliance on the experience of operators, it is difficult to adjust and optimize parameters in a timely manner in the face of complex and changing production environments, which affects the continuity of production and the consistency of products. Summary of the invention

[0004] In order to solve the problem that the traditional process parameter setting method usually relies on experience, the operator adjusts the process parameters through past experience or step-by-step experiments. This method is not only less accurate, but also consumes a lot of manpower and time costs. In addition, due to the reliance on the experience of the operator, it is difficult to adjust and optimize the parameters in a timely manner in the face of a complex and changeable production environment, which affects the technical problem of production continuity and product consistency. The present invention provides a method and system for quickly solving the setting parameters for a multi-process production process.

[0005] The technical solution provided by the embodiment of the present invention is as follows:

[0006] First aspect:

[0007] An embodiment of the present invention provides a method for quickly solving setting parameters for a multi-step production process, comprising:

[0008] S1: Construct a mathematical model to describe multiple processes and determine the basic information of each process;

[0009] S2: Construct a multi-objective optimization model with the goal of maximizing production efficiency and production benefits;

[0010] S3: solving the multi-objective optimization model through the NSGA-II algorithm to determine the optimal dynamic setting strategy;

[0011] S4: converting the multi-objective optimization model into a single-objective optimization model, and determining the constraints in the conversion process;

[0012] S5: adopting the optimal dynamic setting strategy and determining the optimal setting parameters under different input parameters through a particle swarm optimization algorithm;

[0013] S6: constructing a training data set according to each input parameter and the optimal setting parameter corresponding to the input parameter;

[0014] S7: Integrate the constraints in the transformation process into the neural network as physical information to build a physical information neural network model;

[0015] S8: Training the physical information neural network model using the training data set;

[0016] S9: Get input parameters;

[0017] S10: Determine the set parameter values ​​of the input parameters through the trained physical information neural network model.

[0018] Second aspect:

[0019] An embodiment of the present invention provides a system for quickly solving setting parameters for a multi-step production process, comprising:

[0020] processor;

[0021] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method for quickly solving setting parameters for a multi-step production process as described in the first aspect is implemented.

[0022] The third aspect:

[0023] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for quickly solving setting parameters for a multi-step production process as described in the first aspect is implemented.

[0024] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0025] (1) In the present invention, a multi-objective optimization model is constructed with the goal of maximizing production efficiency and production benefits, and the multi-objective optimization model is solved by the NSGA-II algorithm to determine the optimal dynamic setting strategy. The optimal dynamic setting strategy is adopted, and the optimal setting parameters under different input parameters are determined by the particle swarm optimization algorithm. By constructing a multi-objective optimization model based on mathematical modeling, the dependence on experience and experiments is significantly reduced, a large number of trial and error experiments are avoided, the accuracy of process parameter setting is improved, and labor costs are reduced.

[0026] (2) In the present invention, the constraints in the transformation process are integrated into the neural network as physical information to construct a physical information neural network model, which is trained by using a training data set. The set parameter values ​​of the input parameters are determined by the trained physical information neural network model. This method realizes the rapid and automatic adjustment of parameters, ensures the continuity of production and the consistency of products, effectively improves production efficiency and ensures the stability of product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 A schematic diagram of a flow chart of a method for quickly solving set parameters for a multi-step production process provided by an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of the structure of a system for quickly solving setting parameters for a multi-step production process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0031] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0032] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0033] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0034] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0035] Reference Manual Attached Figure 1 , showing a flow chart of a method for quickly solving setting parameters for a multi-step production process provided by an embodiment of the present invention.

[0036] The embodiment of the present invention provides a method for quickly solving setting parameters for a multi-process production process. The method can be implemented by a device for quickly solving setting parameters for a multi-process production process. The device for quickly solving setting parameters for a multi-process production process can be a terminal or a server. The processing flow of the method for quickly solving setting parameters for a multi-process production process can include the following steps:

[0037] S1: Construct a mathematical model to describe multiple processes and determine the basic information of each process.

[0038] In a possible implementation, the multi-process description mathematical model is specifically:

[0039]

[0040]

[0041]

[0042]

[0043]

[0044] i=1,2,,m

[0045] Among them, Q i represents the output of the ith process, C i represents the cost of the ith process, T i represents the completion time of the ith process, P irepresents the output quality of the i-th process, R i represents the resource consumption of the i-th process, Q i-1 represents the output of the i-1th process, C i-1 represents the cost of the i-1th process, T i-1 represents the completion time of the i-1th process, P i-1 represents the output quality of the i-1th process, R i-1 represents the resource consumption of the i-1th process, m represents the total number of processes, θ represents the input parameter, represents the kth setting parameter of the i-th process, k = 1, 2, ..., n i , n i Represents the total number of setting parameters for the i-th process.

[0046] In the present invention, by constructing a multi-process description mathematical model, a clear and systematic structure is provided for the multi-process production process, which helps optimize the production process, reduce costs, and improve efficiency and quality. At the same time, it provides a solid foundation for subsequent parameter optimization and intelligent control.

[0047] S2: Construct a multi-objective optimization model with the goal of maximizing production efficiency and production benefits.

[0048] In a possible implementation, the multi-objective optimization model is specifically:

[0049]

[0050]

[0051] tT i ≤T i_max ,

[0052] R i ≤R i_max ,

[0053] P i ≥P i_min ,i=1,2,…,m

[0054] Among them, max represents the maximum value, X represents the independent variable, Ω represents the range of X values, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-maxrepresents the maximum resource consumption of the ith process, P i-min Represents the minimum output quality of the i-th process.

[0055]

[0056] Where X is an r-dimensional vector, m represents the total number of processes, represents the kth setting parameter of the i-th process, k = 1, 2, ..., n i , n i represents the total number of setting parameters for the i-th process, and T represents the transpose of the matrix.

[0057] In the present invention, by considering multiple optimization goals (such as output, completion time, output quality, cost and resource consumption) at the same time, the entire production process can be comprehensively optimized, and the conflicts between different goals can be balanced to ensure that both production efficiency can be improved and production benefits can be maximized. At the same time, by setting multiple parameters for different processes, the model can flexibly adjust the setting parameters of each process according to the production situation to find the optimal solution for a specific process or a specific production task. This method can achieve personalized optimization under different production conditions and ensure that the system can quickly adapt to various complex production needs.

[0058] S3: The multi-objective optimization model is solved through the NSGA-II algorithm to determine the optimal dynamic setting strategy.

[0059] Among them, NSGA-II (Non-dominated Sorting Genetic Algorithm II) is a multi-objective evolutionary algorithm used to solve multi-objective optimization problems. Unlike traditional single-objective optimization, NSGA-II is suitable for scenarios where multiple conflicting objectives need to be optimized simultaneously. NSGA-II is an improved version of NSGA (Non-dominated Sorting Genetic Algorithm), which improves computational efficiency and convergence performance and is widely used in multi-objective optimization problems in engineering, economics, and production.

[0060] In the present invention, by applying the NSGA-II algorithm, the multi-objective optimization model can simultaneously handle multiple conflicting production objectives, quickly and efficiently find the optimal solution set, and ensure the accuracy, flexibility and global optimality of production parameter settings. This process reduces the reliance on manual experience, improves production efficiency and economic benefits, adapts to changing production environments, and realizes a more intelligent dynamic setting strategy.

[0061] S4: Convert the multi-objective optimization model into a single-objective optimization model and determine the constraints in the conversion process.

[0062] It should be noted that Input parameter for the constraint value of production benefit or production efficiency When the optimal setting parameters are different, they also change accordingly, that is, they need to be adjusted according to To obtain the optimal setting parameters, it is necessary to transform the multi-objective optimization model into a single-objective optimization model.

[0063] In a possible implementation, the single-objective optimization model is specifically:

[0064]

[0065]

[0066] T i ≤T i_max ,

[0067] R i ≤R i_max ,

[0068] P i ≥P i_min ,i=1,2,…,m

[0069] Among them, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-max represents the maximum resource consumption of the ith process, P i-min represents the minimum output quality of the ith process, ε represents the constraint value, Ω represents the value range of X, and X represents the independent variable.

[0070] In the present invention, constraints are introduced into the optimization problem to ensure that the optimization result meets the restrictions in the actual production process. The setting of such constraints can ensure the feasibility of the optimization solution and avoid generating solutions that do not meet the actual production requirements.

[0071] In a possible implementation manner, the constraint conditions in the conversion process in S4 are determined in the following manner:

[0072] Determine the first objective function in the single-objective optimization model:

[0073]

[0074] Among them, f1() represents the first objective function, m represents the total number of processes, Q i() represents the output function of the ith process, P i () represents the output quality function of the i-th process, C i () represents the cost function of the i-th process.

[0075] Convexity analysis is performed on the first objective function to determine the set of independent variables whose first-order derivative function exists and is continuous:

[0076]

[0077] in, represents the set of independent variables for which the first-order derivative of the first objective function exists and is continuous, D1 represents the domain of the first objective function, and e j represents the jth standard basis vector, represents the partial derivative, h represents the change in the independent variable, and r represents the total number of independent variables.

[0078] Based on the first-order derivative function, a second-order derivative is performed, and the set of independent variables for which the second-order derivative function of the first objective function exists and is continuous is determined as:

[0079]

[0080] in, The set of independent variables for which the second-order derivative of the first objective function exists and is continuous, x j represents the j-th independent variable of the first objective function.

[0081] It should be noted that the positive definiteness of the Hessian moment can be used to determine that if the Hessian matrix is ​​a semi-positive definite matrix, then the first objective function is a convex function.

[0082] Construct the Hessian matrix based on the second-order derivative of the first objective function:

[0083]

[0084] in, Represents the Hessian matrix constructed from the second-order derivative of the first objective function.

[0085] Based on the Hessian matrix, for all non-zero vectors All have v T H f (a) v>0, determine the set of positive Hessian matrices as constraints in the transformation process:

[0086]

[0087] in, represents the positive definite set of the Hessian matrix of the first objective function, X 0 represents a point in the domain of the first objective function, represents the domain of the first objective function, represents any vector in n-dimensional space, represents the positive definiteness condition for the Hessian matrix.

[0088] In a possible implementation, the single-objective optimization model is specifically:

[0089]

[0090]

[0091] T i ≤T i_max ,

[0092] R i ≤R i_max ,

[0093] P i ≥P i_min ,i=1,2,…,m

[0094] Among them, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-max represents the maximum resource consumption of the ith process, P i-min represents the minimum output quality of the ith process, ε represents the constraint value, Ω represents the value range of X, and X represents the independent variable.

[0095] In a possible implementation manner, the constraint conditions in the conversion process in S4 are determined in the following manner:

[0096] Determine the second objective function in the single-objective optimization model:

[0097]

[0098] Among them, f1() represents the second objective function, m represents the total number of processes, Q i () represents the output function of the ith process, T i () represents the completion time function of the i-th process.

[0099] Convexity analysis is performed on the second objective function to determine the set of independent variables whose first-order derivative function exists and is continuous:

[0100]

[0101] in, The first-order derivative of the second objective function exists and is continuous. D2 represents the domain of the second objective function. q represents the qth standard basis vector, h represents the change of the independent variable, and r represents the total number of independent variables;

[0102] Based on the first-order derivative function, a second-order derivative is obtained, and the set of independent variables for which the second-order derivative function of the second objective function exists and is continuous is determined as follows:

[0103]

[0104] in, The set of independent variables for which the second-order derivative of the second objective function exists and is continuous, represents the partial derivative, x q represents the qth independent variable of the second objective function.

[0105] Construct the Hessian matrix based on the second-order derivative function of the second objective function:

[0106]

[0107] in, Represents the Hessian matrix constructed from the second derivative of the second objective function.

[0108] Based on the Hessian matrix, for all non-zero vectors All have v T H f (a) v>0, determine the set of positive Hessian matrices as constraints in the transformation process:

[0109]

[0110] in, The Hessian matrix of the second objective function is a positive definite set, X 0 represents a point in the domain of the second objective function, represents the domain of the second objective function, represents any vector in n-dimensional space, represents the positive definiteness condition for the Hessian matrix.

[0111] In the present invention, convexity analysis enables the optimization algorithm to converge to the optimal solution stably and avoid falling into the local optimal solution, thereby improving the robustness and efficiency of the solution. At the same time, the input parameters will change with time or environmental changes. Through the optimization model, combined with the constraints, the optimal setting parameters of the production process can be dynamically adjusted according to different input parameter values, thereby improving the flexibility and adaptability of the production system.

[0112] Furthermore, after being converted into a single-objective optimization model, each key parameter in production can be systematically controlled. This systematic control method avoids the uncertainty brought by the traditional method of relying on experience to set parameters, making the production process more stable and controllable, and can greatly improve the overall production efficiency.

[0113] S5: Adopt the best dynamic setting strategy and determine the optimal setting parameters under different input parameters through the particle swarm optimization algorithm.

[0114] Among them, Particle Swarm Optimization (PSO) is an intelligent optimization algorithm that simulates group behavior. It is inspired by the group behavior of bird flocks or fish schools in nature and uses the mutual cooperation and information sharing of individuals in the group to find the optimal solution to the problem.

[0115] In the present invention, through PSO, the setting parameters can be dynamically adjusted so that the optimal solution can be flexibly found according to different input parameters. This dynamic optimization can improve the flexibility of the production system and ensure that the efficiency and effect of production under different conditions are in the optimal state. At the same time, through the automatic optimization of the particle swarm algorithm, the dependence on manual experience and intervention can be significantly reduced. PSO can automatically adjust the setting parameters through the algorithm, reduce the uncertainty caused by the artificial setting parameters, and can control the production process more accurately.

[0116] S6: Construct a training data set according to each input parameter and the optimal setting parameters corresponding to the input parameters.

[0117] S7: Integrate the constraints in the transformation process into the neural network as physical information to build a physical information neural network model.

[0118] Among them, Physics-Informed Neural Networks (PINNs) is a model that combines physical laws and neural networks to solve partial differential equations (PDEs) and related problems in physical fields. PINNs combines the powerful function approximation ability of neural networks with prior knowledge in physical systems, and can perform model training and prediction while satisfying physical constraints.

[0119] In this invention, the constraints are integrated into the neural network as physical information to construct a physical information neural network model (PINNs), which enables the model to make efficient and accurate predictions under the premise of satisfying physical constraints. This method not only improves the physical rationality and accuracy of the prediction results, but also reduces the dependence on large amounts of data and improves the generalization ability of the model.

[0120] S8: Train the physical information neural network model using the training data set.

[0121] In a possible implementation, training the physical information neural network model in S8 specifically includes:

[0122] Initialize the physical information neural network parameters, which include learning rate, number of iterations, and initial weight threshold.

[0123] Get the training sample set.

[0124] Each training sample in the training sample set is input into the input layer of the physical information neural network model, and the output value of each neuron is calculated.

[0125] Based on the output value, determine the error of each neuron.

[0126] Update the weights and thresholds of the physical information neural network based on the error.

[0127] Repeat the iterations and end the training when the constraint loss function converges.

[0128] In a possible implementation, the constraint loss function is specifically:

[0129]

[0130] Among them, L eq represents the constraint loss function, K represents the number of sample points used to evaluate the constraint, g() represents the constraint function, y() represents the neural network output function, and x k represents the kth input feature vector.

[0131] In the present invention, by initializing the parameters of the neural network and automatically adjusting the weights during the training process, the neural network can converge to the optimal solution after multiple iterations, which reduces the need to manually adjust the network parameters and improves the degree of automation of model optimization. At the same time, the physical information neural network learns the model through training data and always follows the laws of physics during the training process. In this way, even if the data set is noisy or insufficient, the model can give reasonable predictions under physical constraints to ensure the credibility of the results.

[0132] S9: Get input parameters.

[0133] S10: Determine the set parameter values ​​of the input parameters through the trained physical information neural network model.

[0134] In the present invention, the set parameter values ​​of the input parameters are determined by the trained physical information neural network model, which can greatly improve the accuracy, stability and efficiency of the set parameters, reduce the trial and error process, and improve the automation and intelligent level of production. The use of the physical information neural network model ensures that the system reasonably optimizes the set parameters under physical constraints, adapts to complex and dynamic production environments, and greatly improves the flexibility, reliability and economic benefits of production.

[0135] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0136] (1) In the present invention, a multi-objective optimization model is constructed with the goal of maximizing production efficiency and production benefits, and the multi-objective optimization model is solved by the NSGA-II algorithm to determine the optimal dynamic setting strategy. The optimal dynamic setting strategy is adopted, and the optimal setting parameters under different input parameters are determined by the particle swarm optimization algorithm. By constructing a multi-objective optimization model based on mathematical modeling, the dependence on experience and experiments is significantly reduced, a large number of trial and error experiments are avoided, the accuracy of process parameter setting is improved, and labor costs are reduced.

[0137] (2) In the present invention, the constraints in the transformation process are integrated into the neural network as physical information to construct a physical information neural network model, which is trained by using a training data set. The set parameter values ​​of the input parameters are determined by the trained physical information neural network model. This method realizes the rapid and automatic adjustment of parameters, ensures the continuity of production and the consistency of products, effectively improves production efficiency and ensures the stability of product quality.

[0138] Reference Manual Attached Figure 2 , showing a schematic structural diagram of a system for rapidly solving set parameters for a multi-step production process provided by the present invention.

[0139] The present invention further provides a system 20 for quickly solving setting parameters for a multi-step production process, which is applied to the above-mentioned method for quickly solving setting parameters for a multi-step production process, and comprises:

[0140] Processor 201;

[0141] The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201, the method for quickly solving setting parameters for a multi-step production process as described in the method embodiment is implemented.

[0142] The system 20 for quickly solving setting parameters for a multi-process production process provided by the present invention can execute the above-mentioned method for quickly solving setting parameters for a multi-process production process and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0143] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0144] (1) In the present invention, a multi-objective optimization model is constructed with the goal of maximizing production efficiency and production benefits, and the multi-objective optimization model is solved by the NSGA-II algorithm to determine the optimal dynamic setting strategy. The optimal dynamic setting strategy is adopted, and the optimal setting parameters under different input parameters are determined by the particle swarm optimization algorithm. By constructing a multi-objective optimization model based on mathematical modeling, the dependence on experience and experiments is significantly reduced, a large number of trial and error experiments are avoided, the accuracy of process parameter setting is improved, and labor costs are reduced.

[0145] (2) In the present invention, the constraints in the transformation process are integrated into the neural network as physical information to construct a physical information neural network model, which is trained by using a training data set. The set parameter values ​​of the input parameters are determined by the trained physical information neural network model. This method realizes the rapid and automatic adjustment of parameters, ensures the continuity of production and the consistency of products, effectively improves production efficiency and ensures the stability of product quality.

[0146] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0147] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0148] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0149] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0150] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0151] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0152] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0153] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0154] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0155] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0157] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0158] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for quickly solving setting parameters for a multi-step production process as described in the method embodiment is implemented.

[0159] A computer-readable storage medium provided by the present invention can implement the steps and effects of the method for quickly solving the setting parameters for a multi-step production process of the above-mentioned method embodiment. To avoid repetition, the present invention will not go into details.

[0160] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0161] (1) In the present invention, a multi-objective optimization model is constructed with the goal of maximizing production efficiency and production benefits, and the multi-objective optimization model is solved by the NSGA-II algorithm to determine the optimal dynamic setting strategy. The optimal dynamic setting strategy is adopted, and the optimal setting parameters under different input parameters are determined by the particle swarm optimization algorithm. By constructing a multi-objective optimization model based on mathematical modeling, the dependence on experience and experiments is significantly reduced, a large number of trial and error experiments are avoided, the accuracy of process parameter setting is improved, and labor costs are reduced.

[0162] (2) In the present invention, the constraints in the transformation process are integrated into the neural network as physical information to construct a physical information neural network model, which is trained by using a training data set. The set parameter values ​​of the input parameters are determined by the trained physical information neural network model. This method realizes the rapid and automatic adjustment of parameters, ensures the continuity of production and the consistency of products, effectively improves production efficiency and ensures the stability of product quality.

[0163] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0164] There are a few points to note:

[0165] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0166] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0167] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0168] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for quickly solving setting parameters for a multi-step production process, characterized in that: include: S1: Construct a mathematical model to describe multiple processes and determine the basic information of each process; S2: Construct a multi-objective optimization model with the goal of maximizing production efficiency and production benefits; S3: solving the multi-objective optimization model through the NSGA-II algorithm to determine the optimal dynamic setting strategy; S4: converting the multi-objective optimization model into a single-objective optimization model, and determining the constraints in the conversion process; S5: adopting the optimal dynamic setting strategy and determining the optimal setting parameters under different input parameters through a particle swarm optimization algorithm; S6: constructing a training data set according to each input parameter and the optimal setting parameter corresponding to the input parameter; S7: Integrate the constraints in the transformation process into the neural network as physical information to build a physical information neural network model; S8: Training the physical information neural network model using the training data set; S9: Get input parameters; S10: Determine the set parameter values ​​of the input parameters through the trained physical information neural network model.

2. The method for quickly solving setting parameters for a multi-step production process according to claim 1 is characterized in that: The multi-process description mathematical model is specifically: i=1,2,…,m Among them, Q i represents the output of the ith process, C i represents the cost of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, R i represents the resource consumption of the i-th process, Q i-1 represents the output of the i-1th process, C i-1 represents the cost of the i-1th process, T i-1 represents the completion time of the i-1th process, P i-1 represents the output quality of the i-1th process, R i-1 represents the resource consumption of the i-1th process, m represents the total number of processes, θ represents the input parameter, represents the kth setting parameter of the i-th process, k = 1, 2, ..., n i , n i Represents the total number of setting parameters for the i-th process.

3. The method for quickly solving setting parameters for a multi-step production process according to claim 1 is characterized in that: The multi-objective optimization model is specifically: s.t.T i ≤T i_max , R i ≤R i_max , P i ≥P i_min ,i=1,2,…,m Among them, max represents the maximum value, X represents the independent variable, Ω represents the range of X values, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-max represents the maximum resource consumption of the ith process, P i-min represents the minimum output quality of the i-th process; Where X is an r-dimensional vector, m represents the total number of processes, represents the kth setting parameter of the i-th process, k = 1, 2, ..., n i , n i represents the total number of setting parameters for the i-th process, and T represents the transpose of the matrix.

4. The method for quickly solving setting parameters for a multi-step production process according to claim 1 is characterized in that: The single-objective optimization model is specifically: T i ≤T i_max , R i ≤R i_max , P i ≥P i_min ,i=1,2,…,m Among them, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-max represents the maximum resource consumption of the ith process, P i-min represents the minimum output quality of the ith process, ε represents the constraint value, Ω represents the value range of X, and X represents the independent variable.

5. The method for quickly solving setting parameters for a multi-step production process according to claim 4 is characterized in that: The specific method for determining the constraint conditions in the conversion process in S4 is: Determine the first objective function in the single-objective optimization model: Among them, f1() represents the first objective function, m represents the total number of processes, Q i () represents the output function of the ith process, P i () represents the output quality function of the i-th process, C i () represents the cost function of the i-th process; Convexity analysis is performed on the first objective function to determine that the set of independent variables for which the first-order derivative function of the first objective function exists and is continuous is: in, represents the set of independent variables for which the first-order derivative of the first objective function exists and is continuous, D1 represents the domain of the first objective function, and e j represents the jth standard basis vector, represents the partial derivative, h represents the change in the independent variable, and r represents the total number of independent variables; A second-order derivative is performed on the basis of the first-order derivative function, and the set of independent variables for which the second-order derivative function of the first objective function exists and is continuous is determined as: Among them, D′ f ′1 represents the set of independent variables for which the second-order derivative of the first objective function exists and is continuous, x j represents the j-th independent variable of the first objective function; The Hessian matrix is ​​constructed according to the second-order derivative function of the first objective function: in, A Hessian matrix constructed by the second-order derivative function representing the first objective function; Based on the Hessian matrix, for all non-zero vectors All have v T H f (a) v>0, determine the set of positive definite Hessian matrices as constraints in the transformation process: Among them, Z f1 represents the positive definite set of the Hessian matrix of the first objective function, X 0 represents a point in the domain of the first objective function, D f1 represents the domain of the first objective function, represents any vector in n-dimensional space, represents the positive definiteness condition for the Hessian matrix.

6. The method for quickly solving setting parameters for a multi-step production process according to claim 1 is characterized in that: The single-objective optimization model is specifically: T i ≤T i_max , R i ≤R i_max , P i ≥P i_min ,i=1,2,…,m Among them, m represents the total number of processes, Q i represents the output of the ith process, T i represents the completion time of the ith process, P i represents the output quality of the i-th process, C i represents the cost of the ith process, R i represents the resource consumption of the ith process, T i-max represents the maximum completion time of the ith process, R i-max represents the maximum resource consumption of the ith process, P i-min represents the minimum output quality of the ith process, ε represents the constraint value, Ω represents the value range of X, and X represents the independent variable.

7. The method for quickly solving setting parameters for a multi-step production process according to claim 6 is characterized in that: The specific method for determining the constraint conditions in the conversion process in S4 is: Determine the second objective function in the single objective optimization model: Among them, f1() represents the second objective function, m represents the total number of processes, Q i () represents the output function of the i-th process, T i () represents the completion time function of the i-th process; Convexity analysis is performed on the second objective function to determine that the set of independent variables for which the first-order derivative function of the second objective function exists and is continuous is: Among them, D′ f2 The first-order derivative of the second objective function exists and is continuous. D2 represents the domain of the second objective function. q represents the qth standard basis vector, h represents the change of the independent variable, and r represents the total number of independent variables; A second-order derivative is performed on the basis of the first-order derivative function, and the set of independent variables for which the second-order derivative function of the second objective function exists and is continuous is determined as: in, The set of independent variables for which the second-order derivative of the second objective function exists and is continuous, represents the partial derivative, x q represents the qth independent variable of the second objective function; The Hessian matrix is ​​constructed according to the second-order derivative function of the second objective function: in, A Hessian matrix constructed by the second-order derivative function representing the second objective function; Based on the Hessian matrix, for all non-zero vectors All have v T H f (a) v>0, determine the set of positive definite Hessian matrices as constraints in the transformation process: in, The Hessian matrix of the second objective function is a positive definite set, X 0 represents a point in the domain of the second objective function, represents the domain of the second objective function, represents any vector in n-dimensional space, represents the positive definiteness condition for the Hessian matrix.

8. The method for quickly solving setting parameters for a multi-step production process according to claim 1 is characterized in that: The physical information neural network model is trained in S8, specifically including: Initializing physical information neural network parameters, wherein the physical information neural network parameters include a learning rate, a number of iterations, and an initial weight threshold; Obtaining the training sample set; Input each training sample in the training sample set into the input layer of the physical information neural network model, and calculate the output value of each neuron; Determining the error of each neuron based on the output value; Update the weight and threshold of the physical information neural network according to the error; Repeat the iterations and end the training when the constraint loss function converges.

9. The method for quickly solving setting parameters for a multi-step production process according to claim 8, characterized in that: The constraint loss function is specifically: Among them, L eq represents the constraint loss function, K represents the number of sample points used to evaluate the constraint, g() represents the constraint function, y() represents the neural network output function, and x k represents the kth input feature vector.

10. A system for quickly solving setting parameters for a multi-step production process, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, a method for quickly solving setting parameters for a multi-step production process as described in any one of claims 1 to 9 is implemented.

Images