A joint optimization method for multi-process mixed production
By building a full-process optimization algorithm model and a sparrow search algorithm optimization, the problem of mutual constraints between multiple processes in rubber tire manufacturing is solved, and the economic benefits and quality of the entire process are achieved, which is suitable for personalized product customization.
Patent Information
- Application Number
- CN202111565494.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-20
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-12-20
AI Technical Summary
During the rubber tire manufacturing process, the mutual constraints between multiple processes and external factors make the entire process difficult to achieve, especially when personalized product customization lacks effective quality control solutions.
Using the multi-process joint optimization method, by building a full-process optimization algorithm model, selecting key operation variables, using the sparrow search algorithm for optimization, iteratively update to the global optimal solution, and adjusting the parameters of each process to achieve the economic benefits and quality of the entire process .
It improves production efficiency, reduces production costs and energy consumption, ensures the maximization of product quality and economic benefits, and solves the quality control problems when customizing personalized products.
Smart Images

Figure CN114707686B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of rubber tire manufacturing, and in particular relates to a joint optimization method for multi-process mixed production. Background Art
[0002] The manufacturing of rubber tires adopts a typical mixed production model. Due to the high coupling of multiple processes, multiple external factors and product quality, the complexity of the manufacturing process and the difficulty of scheduling and control increase. In particular, rubber tire products in major projects in demanding fields have put forward customized functional requirements such as strong impact resistance, high and low temperature resistance, and light weight to adapt to complex and changeable application scenarios such as high-speed take-off and landing, complex road conditions, and higher requirements for important product performance such as balance, elasticity, and wear resistance. It is necessary to jointly regulate the influencing parameters of the continuous process and discrete production process to ultimately obtain product quality that meets certain target requirements.
[0003] The production process of rubber tires consists of several processes, such as mixing process, kneading process, tape process, vulcanization process, molding process, etc. Each key process has its own optimization goal, but the processes are mutually restricted and constrained, so the optimization of a single process cannot guarantee the optimization of the entire process. For example, the improvement of the control parameters of a certain process will directly affect the quality of the semi-finished product of the process, but it does not mean that the optimization goal of the entire process has been achieved. Therefore, the various processes in the complex industrial process are closely related and influence each other. Simply relying on the optimization of a single process cannot solve the problem of full process optimization. It is necessary to coordinate the various processes from the overall perspective of the production process to ultimately achieve full process optimization.
[0004] The whole process optimization method is to strive to obtain a set of optimal operating variables under the condition of satisfying production constraints, balance the various consumption and raw material costs in the whole process, and obtain the maximum economic benefits. This optimization technology can make the whole production process reach the best operating state by adjusting the specific operation amount of each process without changing the process flow and production equipment.
[0005] At present, there is no good solution for full-process quality modeling when the rubber tire industry conducts personalized product customization. The quality control parameters of the company's existing products mainly rely on previous experience and set parameters, but for the personalized customized products suddenly required by the market, the company is unable to provide a detailed solution. Summary of the invention
[0006] The present invention provides a multi-process joint optimization method, which obtains the quality standards and operating variables of new products by building an algorithm model and the quality standards and operating variables of existing products, thereby ensuring the optimal product quality and economic benefits.
[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions to achieve the above problems:
[0008] A joint optimization method for multi-process mixed production includes the following steps:
[0009] Step S101: selecting operating variables that play a key role in the optimization goal of the entire tire manufacturing process as decision variables of the entire process optimization algorithm model to obtain constraint conditions;
[0010] Step S102: constructing the full-process optimization algorithm model for the rubber tire production process;
[0011] Step S103: Optimizing the parameters of the decision variables of the full-process optimization algorithm model, and iteratively updating until converging to a global optimal solution;
[0012] Step S104: Verify the production energy consumption and economic benefits of the full-process optimization algorithm model in mixed production through existing unused data.
[0013] In one embodiment, the full-process optimization algorithm model is:
[0014]
[0015] Among them, f i is the model of the ith process, is the state vector, N i is the dimension of the state vector, u i is the operation vector of the ith process, z i is the association vector of the ith process, A ij is the correlation matrix between the i-th process and the j-th process, which is an N j ×N j phalanx.
[0016] In one embodiment,
[0017]
[0018] is the influence factor of the kth state variable of the jth process on the ith process, which is 0 or 1; 0 means no influence and 1 means influence.
[0019] In one embodiment, the objective function is:
[0020] maxJ=P(u)-C(u)
[0021] Among them, J is the comprehensive production index of the whole process; P is the profit; C is the production cost; u is the decision variable;
[0022] u=[u1 T,u2 T ,…,u n T ] T .
[0023] In one embodiment, the constraint condition is:
[0024] x i =f i (u i ,z i )
[0025] x i,min ≤x i ≤x i,max
[0026] u i,min ≤u i ≤u i,max ,i=1,2,,3…,n
[0027] z i,min ≤z i ≤z i,max ,i=1,2,,3…,n-1
[0028] Among them, the vector x i,min and x i,max is the value range of the quality index of the i-th process; vector u i,min and u i,max is the upper and lower limits of the range of values of the operating variable of the i-th process; vector z i,min and z i,max are the upper and lower limits of the associated variables of the i-th process.
[0029] In one embodiment, a sparrow search algorithm is used to optimize the decision variables of the full-process optimization algorithm model.
[0030] In one embodiment, the decision variable u on each process is used to generate the initial value of the population position matrix C through the sparrow search algorithm; the number of processes is n; the population position matrix C is an n-dimensional matrix,
[0031]
[0032] And calculate the fitness value at each position point to obtain the fitness matrix F(C),
[0033]
[0034] In one embodiment, the number of evolutions E of the sparrow search algorithm population is set, and the population is divided into three groups: joiners, discoverers, and scouts; the constraint processing strategy of the penalty function is used to impose a certain penalty on infeasible solutions so that the whole process gradually converges.
[0035] In one embodiment, the observer provides food resource information to the population and guides the population's foraging direction. The position changes during each round of population evolution are:
[0036]
[0037] t represents the current iteration number, c i,j represents the position of the i-th sparrow in the j-th dimension; R2 (R2∈[0,1]) and ST (ST∈[0.5,1]) represent the warning value and safety value, respectively, that is, the threshold range of the production parameters of each process; R2≥ST means that the position is safe and can be searched, otherwise a warning is issued and the current position is left; Emax is a constant, indicating the maximum number of iterations; Q is a random number that follows a normal distribution; L represents a 1×d matrix, in which all elements in the matrix are 1;
[0038] The joiner updates the foraging direction by listening to the location information of the discoverer to compete for high-quality resources. The position of the joiner is updated as follows:
[0039]
[0040] c p is the local optimal position of the finder, c worst represents the global worst position; A represents a 1×d matrix, in which each element is randomly assigned a value of 1 or -1, and A+=AT(AAT)-1; when When , it indicates that the i-th participant still has no food and has a low fitness, and needs to find other locations to obtain food resources;
[0041] The scouts provide the population with timely danger information, accounting for 10% to 20% of the population, and update their positions according to the following formula:
[0042]
[0043] c best is the global optimal position in this round of iteration; η is the step size control parameter, which is a normally distributed random number with a mean of 0 and a variance of 1; when f i >f b It means that the sparrow is vulnerable to predators and is on the edge of the population. i =f bWhen , it indicates that the sparrow is aware of the danger and needs to update its position to reduce the risk of being attacked; the complex industrial process composed of each of the above-mentioned processes must adjust the size of key parameters within a reasonable range to achieve the optimal economic benefits, quality production indicators and energy consumption indicators of the entire process, that is, after the sparrow search algorithm updates the population E times, the individual food resources in the population reach the optimal solution of the area;
[0044] K∈[-1,1] is a random number, indicating the direction of the sparrow's movement and also the step size control parameter; fi is the fitness value of the current sparrow individual; fb and fw are the current global best and worst fitness values respectively; ε is the smallest constant to avoid zero in the denominator.
[0045] In one embodiment, during each round of population position update, the decision variable u of each process is i To make adjustments, the decision variables must satisfy u under the premise of achieving production parameters. i,min ≤u i ≤u i,max When producing, the production cost and energy consumption can be reduced to the greatest extent, and the production efficiency can be improved;
[0046] The state vector of the whole process x=[x1 T ,x2 T ,…,x n T ] T After each process is completed, it is updated and affects the next process. The change of the state vector affects the quality result of the entire industrial production.
[0047] Compared with the prior art, the advantages and positive effects of the present invention are:
[0048] The joint optimization method for multi-process mixed production of the present invention adopts a typical mixed manufacturing production mode for the rubber tire industry. The multiple processes are closely related, and the parameters of each process in the manufacturing process are difficult to control, so that the production quality is affected by various external factors. Global joint optimization of multiple process parameters in the whole process can improve the production efficiency of each process and maximize the economic benefits to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0050] Figure 1It is a flow chart of an embodiment of a joint optimization method for multi-process mixed production proposed by the present invention;
[0051] Figure 2 It is a schematic diagram of a production process consisting of n processes. DETAILED DESCRIPTION
[0052] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.
[0053] Reference Figure 1 The joint optimization method for multi-process mixed production of the present invention comprises the following steps:
[0054] Step S101: selecting operating variables that play a key role in the optimization goal of the entire tire manufacturing process as decision variables of the entire process optimization model to obtain constraint conditions;
[0055] Step S102: constructing a full-process optimization model for the rubber tire production process;
[0056] Step S103: Optimizing the parameters of the decision variables of the full-process optimization model, and iteratively updating until converging to a global optimal solution;
[0057] Step S104: Verify the production energy consumption and economic benefits of the full-process optimization model in mixed production through existing unused data.
[0058] The combined optimization method for multi-process mixed production of the present invention adopts a typical mixed manufacturing production mode for the rubber tire industry. The multiple processes are closely related, and the parameters of each process in the manufacturing process are difficult to control, so that the production quality is affected by various external factors. Global optimization of multiple process parameters in the whole process can improve the production efficiency of each process and maximize the economic benefits to a certain extent.
[0059] The overall technical solution mainly introduces the optimization idea of a single process directly into the optimization of the whole process, that is, optimizes the whole production process as a whole. Through the analysis of the whole process mechanism model, the operating variables that play a key role in the optimization goal of the whole process of tire manufacturing are selected as the decision variables of the whole process optimization model, and the constraints that should be met by the process optimization are obtained.
[0060] The optimization goal of the entire rubber tire production process is to obtain the maximum production efficiency (or maximum economic benefit, minimum cost, minimum energy consumption, and most stable quality) while the product meets the quality indicators, and to provide the operating variables under the conditions of maximum production efficiency to guide on-site practice. The optimization goal often requires a comprehensive analysis of economic benefits, quality indicators, and energy consumption indicators. For the rubber tire production process, the optimization is to maximize the economic benefits of a single batch of raw materials. In the process of constructing the full-process model optimization, first ensure that the key parameter variables that affect the product process performance in each process are within a reasonable range, and then adjust the weight parameters to achieve the optimal optimization goal.
[0061] Reference Figure 2 For a complex industrial process consisting of n processes, each process is connected to form a complete flow.
[0062] x i ,u i , and z i , these three vectors are the quality index, operation variable and correlation variable of the ith process. For a complex industrial process consisting of n processes, when the mechanism model of each process has been established separately, since each process affects the next process through related indicators and is related to each other, through the correlation between processes.
[0063] These n processes can be connected to each other one by one, or some important processes can be connected together.
[0064] The whole process model can be obtained by combining multiple process models, as follows:
[0065]
[0066] Among them, f i is the model of the ith process, is the state vector, N i is the dimension of the state vector, u i is the operation vector of the ith process, z i is the association vector of the ith process, A ij is the correlation matrix between the i-th process and the j-th process, which is an N j ×N j phalanx.
[0067] The formula is:
[0068]
[0069] is the influence factor of the kth state variable of the jth process on the ith process, which is 0 or 1. 0 means no influence and 1 means influence.
[0070] The objective function is max J(x,u) which is expressed as follows:
[0071] st(constraints)
[0072] x i =f i (u i ,z i ),i=1,2,…,n
[0073] z i =[(A i1 x1) T ,(A i2 x2) T ,…,(A i(i-1) x (i-1) ) T ,(A i(i+1) x (i+1) ) T ,…,(A in x n ) T ] T
[0074] J is the comprehensive economic benefit, x=[x1 T ,x2 T ,…,x n T ] T is the state vector of the entire process.
[0075] maxJ=P(u)-C(u)
[0076] st
[0077] x i =f i (u i ,z i )
[0078] x i,min ≤x i ≤x i,max
[0079] u i,min ≤u i ≤u i,max ,i=1,2,,3…,n
[0080] z i,min ≤z i ≤z i,max ,i=1,2,,3…,n-1
[0081] Among them, J is the comprehensive production index of the whole process, P is the profit, and C is the production cost. The production cost is the sum of the raw material cost and the consumption cost. Usually the raw material cost is a certain value, so P and C are determined by the operating variables u of each process, expressed as P(u) and C(u) respectively; the vector x i,min and x i,max is the value range of process quality index; vector u i,min and u i,max is the upper and lower limits of the range of the operating variables of the ith process; vector z i,min and z i,max are the upper and lower limits of the variables associated with the i-th process.
[0082] The present invention optimizes the process parameters of the whole process of rubber tire production based on sparrow search, combines the constraint processing mechanism, adjusts the industrial production parameters of the whole process, and uses the constraint processing strategy of the penalty function to impose a certain penalty on the infeasible solution. After continuous iterative updates, the optimization algorithm gradually converges to a feasible global optimal solution. The technical solution is implemented as follows:
[0083] (1) The optimization of the entire process of rubber tire production is a complex constrained nonlinear optimization problem. Under the premise of ensuring production quality, the key parameters of n processes in the production process are optimized within a reasonable range. According to the input-output mapping relationship of the industrial process, the production quality and comprehensive economic benefits are optimized under the constraints of the entire process. The parameter index u on each production process is used to initialize the population position of the sparrow search algorithm to generate the initial position matrix C of the population. The n-dimensional sparrow population position matrix is established according to the parameters of the n processes, and the fitness value at each position point is calculated to obtain the fitness matrix F(C) of the n processes.
[0084]
[0085] (2) Set the number of evolutions E of the sparrow search algorithm population and divide the population into three groups: joiners, discoverers, and scouts. At the same time, use the constraint processing strategy of the penalty function to impose a certain penalty on infeasible solutions, that is, adjust the non-global optimal solutions of industrial parameters, so that the whole process gradually converges.
[0086] (3) The positions of the individuals in each round of evolution are adjusted. The positions of individuals with different identities are updated as follows. The observer provides food resource information to the population and guides the foraging direction of the population. The position changes in each round of population evolution are:
[0087]
[0088] t represents the current iteration number, c i,jrepresents the position of the i-th sparrow in the j-th dimension; R2(R2∈[0,1]) and ST(ST∈[0.5,1]) represent the warning value and safety value respectively, that is, the threshold range of the production parameters of each process; R2≥ST means that the position is safe and can be searched, otherwise a warning is issued and the current position is left; Emax is a constant representing the maximum number of iterations; Q is a random number that obeys the normal distribution; L represents a 1×d matrix, in which all elements in the matrix are 1.
[0089] The joiner updates the foraging direction by listening to the location information of the discoverer to compete for high-quality resources. The position of the joiner is updated as follows:
[0090]
[0091] c p is the local optimal position of the finder, c worst represents the global worst position. A represents a 1×d matrix, in which each element is randomly assigned a value of 1 or -1, and A+=AT(AAT)-1; when When , it indicates that the i-th participant still has no food and has a low fitness, and needs to find other locations to obtain food resources.
[0092] The scouts provide the population with timely danger information, accounting for 10% to 20% of the population, and update their positions according to the following formula:
[0093]
[0094] c best is the global optimal position in this iteration. η is the step size control parameter, which is a normal distribution random number with a mean of 0 and a variance of 1. i >f b It means that the sparrow is vulnerable to predators and is on the edge of the population. i =f b , it indicates that the sparrow is aware of the danger and needs to update its position to reduce the risk of being attacked; the complex industrial process composed of each of the above-mentioned steps must adjust the size of key parameters within a reasonable range to achieve the optimal economic benefits, quality production indicators and energy consumption indicators of the entire process, that is, after the sparrow search algorithm has updated the population E times, the individual food resources in the population reach the optimal solution for the area.
[0095] K∈[-1,1] is a random number, indicating the direction of the sparrow's movement and also the step size control parameter; fi is the fitness value of the current sparrow individual; fb and fw are the current global best and worst fitness values respectively; ε is the smallest constant to avoid zero in the denominator.
[0096] (4) In each round of population position update, the decision variable u of each process is i To make adjustments, the decision variables must satisfy u under the premise of achieving production parameters. i,min ≤u i ≤u i,max When producing, the production cost and energy consumption can be reduced to the greatest extent, and the production efficiency can be improved;
[0097] The state vector of the whole process x=[x1 T ,x2 T ,…,x n T ] T After each process is completed, it is updated and affects the next process. The change of the state vector affects the quality result of the entire industrial production.
[0098] (5) When the resources in the position of the entire population occupy the global optimum, the optimization is completed. The position matrix optimized by the sparrow search algorithm is used to set the parameter values of each process, thereby realizing the overall optimization of the entire manufacturing process. The optimized process parameters are tested to verify the impact of optimizing industrial production parameters on the production results of the entire process. By comparing the production cost, production quality, and production energy consumption, a comprehensive analysis of the optimized key parameters is conducted. According to the inherent mechanism of each process itself, the basic principles of energy balance, material balance, phase equilibrium, and reaction kinetics are used to establish the parameter optimization plan for each process. With the goal of maximizing the economic benefits of a single batch of raw materials, a normalized model of the entire process of rubber tire production is established and solved, and the key parameters β of each process are globally optimized so that the input-output mapping y=f(u,β) that characterizes the entire process is optimal under the conditions of satisfying constraints such as quality indicators and industrial parameters.
[0099] The present invention proposes a method for optimizing the process parameters of the entire process of rubber tire production based on a sparrow search algorithm. Combined with the constraints of each process, the constraint processing strategy of the penalty function is used to globally optimize the parameters of each process of mixed manufacturing production. This optimization method aims to solve the problems of mutual restriction and mutual constraint among various production processes in the production process. From the perspective of the entire process, it coordinates each production process, effectively alleviates the problem of unstable product quality caused by changes in individual process parameters, and obtains the maximum production efficiency under the condition of satisfying production constraints. After the global production parameters are optimized, the production cost and production energy consumption of each process are significantly reduced, the product quality performance is improved, the key production processes meet their respective optimization standards, and the comprehensive economic benefits are significantly improved. This optimization method can quickly and accurately optimize global parameters, and provides an effective solution for the optimization of the entire process of rubber tires.
[0100] The whole process optimization method is to strive to obtain a set of optimal operating variables u under the condition of satisfying production constraints, balance the various consumption and raw material costs in the whole process, and maximize the economic benefits. This optimization technology can make the whole production process reach the best operating state by adjusting the specific operation amount of each process without changing the process flow and production equipment.
[0101] This patent focuses on using optimization algorithms, applying the full-process mechanism model, and determining the best operating variable u based on the constraints of actual production. In full-process optimization, the operating variable should be a variable that has an important impact on the objective function. The optimization goal can be achieved by selecting a set of appropriate operating variables. Therefore, the key operating variable u is extracted for the important process, so that u meets certain constraints and J reaches the optimal goal. Specific embodiments
[0103] A joint optimization method for multi-process mixed production, the production process is a rubber tire production process, comprising the following steps:
[0104] Step 1: Production process analysis, identify the final quality indicators of the tire production process, the quality indicators of each process and the process parameters within the process; the final quality indicator is tensile strength, the process includes two processes: mixing and vulcanization, and other processes are discrete processes and are not considered here.
[0105] 1) The quality indicators of the mixing process include Mooney viscosity and hardness, and the process parameters include ejector pressure, rotor speed, mixing time, and mixing and debonding temperature.
[0106] 2) The quality indicators of the vulcanization process include tensile strength, hardness and wear resistance, and the process parameters include vulcanization time, vulcanization temperature, steam pressure inside the capsule, internal hot water pressure, and internal hot water temperature;
[0107] Step 2: Use the 8,600 quality indicator data corresponding to the process parameters and batches of each process collected by the enterprise as experimental data, randomly select 80% of the samples as the training set and 20% as the test set.
[0108] Step 3: Filter out the tire final product index tensile strength, the quality index data of each process, and the process parameter data within the process from the modeling data set to form a modeling data set of quality indicators for each process.
[0109] Step 4: Based on the above tire final product quality index data set and each process quality index data set, establish each process quality index model and tire tensile strength model respectively:
[0110] 1): First, preprocess all data sets, including the removal of missing values and outliers. For missing values, delete the record directly; for outliers, remove the data according to the method of data greater than ±3σ standard deviation. If the value is not within the range of (μ-3θ, μ+3θ), it is judged to deviate from the normal value and the record needs to be removed;
[0111] 2) The preprocessed dataset is divided into a modeling dataset and a test set in an 8:2 ratio.
[0112] First, the quality model of the mixing process and the vulcanization process is established. Based on the initial mechanism model framework of process parameters and various quality indicators, the process parameter indicators are used as input and the quality indicators are used as output. The neural network identification method is used to determine the other unknown parameters of the model, forming the identification model of the Mooney viscosity and hardness indicators of the mixing process and the identification model of the tensile strength, hardness and wear resistance indicators of the vulcanization process. The remaining 20% sample data is used to verify the accuracy of each model, so that the model has good robustness and generalizability.
[0113] Step 5: Based on the optimal combination of mixing engineering and vulcanization process, the tensile strength model of the final product quality is obtained. The solution method is:
[0114] 1): Based on the tire mass tensile strength model as the equality constraint, according to the value range in the X, U, and Z constraints, that is, the inequality constraints of the optimization problem, the optimization problem is constructed with the optimal tensile strength as the objective function;
[0115] 2): According to the optimization problem constructed in step 1), the sparrow search algorithm is used to optimize the single process and the multi-process combination and solve the multi-objective optimization, and the best solution is recommended from the many feasible solutions optimized for output.
[0116] In specific applications, the production process data and quality inspection data of different models of tires produced in 2020 are exported from the enterprise ERP, MES and quality systems, and the data are summarized one by one according to the batch number to form a data set including key process parameters of each process, Mooney viscosity and hardness of the mixing process, tensile strength, hardness and wear resistance of the vulcanization process, tensile strength of the finished product and other indicators. The data set is processed by preprocessing and other methods such as data anomaly point proposal and normalization. Based on the initial mechanism model architecture of the process parameters of each process and the corresponding quality indicators, the neural network method is used to establish the quality indicator model of each process, and the sparrow search algorithm is used for optimization. The optimal tensile strength of the tire after this optimization operation is 10.5 MPa, the Mooney viscosity of the mixing process is 175, the hardness is 65a, the tensile strength of the vulcanization process is 11 MPa, the hardness is 85a and the wear resistance is 520.
[0117] The process parameters of the optimal solution are used as the actual production indicators. By producing tires according to the indicators, better product quality performance can be obtained.
[0118] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by the terms "up", "down", "front", "back", "left", "right", etc. are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.
[0119] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0120] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0121] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.
[0122] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0123] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.
Claims
1. A joint optimization method for multi-process mixed production, characterized in that: The following steps are involved: Step S101: selecting operating variables that play a key role in the optimization goal of the entire tire manufacturing process as decision variables of the entire process optimization algorithm model to obtain constraint conditions; Step S102: constructing the full-process optimization algorithm model for the rubber tire production process; Step S103: Optimizing the parameters of the decision variables of the full-process optimization algorithm model, and iteratively updating until converging to a global optimal solution; Step S104: verifying the production energy consumption and economic benefits of the full-process optimization algorithm model in mixed production through existing unused data; The full-process optimization algorithm model is: Among them, f i is the model of the ith process, is the state vector, N i is the dimension of the state vector, u i is the operation vector of the ith process, z i is the association vector of the ith process, A ij is the correlation matrix between the i-th process and the j-th process, which is an N j ×N j phalanx. is the influence factor of the kth state variable of the jth process on the ith process, which is 0 or 1; 0 means no influence and 1 means influence.
2. The joint optimization method for multi-process mixed production according to claim 1 is characterized in that: The objective function is: maxJ=P(u)-C(u) Among them, J is the comprehensive production index of the whole process; P is the profit; C is the production cost; u is the decision variable; u=[u1T,u2 T ,…,u n T ] T 。 3. The joint optimization method for multi-process mixed production according to claim 2 is characterized in that: The constraints are: x i =f i (u i ,z i ) u i,min ≤u i ≤u i,max ,i=1,2,,3…,n z i,min ≤z i ≤z i,max ,i=1,2,,3…,n-1 Among them, the vector x i,min and x i,max is the value range of process quality index; vector u i,min and u i,max is the upper and lower limits of the range of the operating variables of the ith process; vector z i,min and z i,max are the upper and lower limits of the variables associated with the i-th process.
4. The joint optimization method for multi-process mixed production according to any one of claims 1 to 3 is characterized in that: The sparrow search algorithm is used to optimize the decision variables of the full-process optimization algorithm model.
5. The joint optimization method for multi-process mixed production according to claim 4 is characterized in that: The decision variable u on each process is used to generate the initial value of the population position matrix C through the sparrow search algorithm; the number of processes is n; the population position matrix C is an n-dimensional matrix, And calculate the fitness value at each position point to obtain the fitness matrix F(C), 6. The joint optimization method for multi-process mixed production according to claim 5 is characterized in that: The number of population evolutions of the sparrow search algorithm is set to E, and the population is divided into three groups: joiners, discoverers, and scouts; the constraint processing strategy of the penalty function is used to impose a certain penalty on infeasible solutions, so that the whole process gradually converges.
7. The joint optimization method for multi-process mixed production according to claim 6 is characterized in that: The observer provides food resource information to the population and guides the population's foraging direction. The position changes during each round of population evolution are: t represents the current iteration number, c i,j represents the position of the i-th sparrow in the j-th dimension; R2 (R2∈[0,1]) and ST (ST∈[0.5,1]) represent the warning value and safety value, respectively, that is, the threshold range of the production parameters of each process; R2≥ST means that the position is safe and can be searched, otherwise a warning is issued and the current position is left; Emax is a constant, indicating the maximum number of iterations; Q is a random number that follows a normal distribution; L represents a 1×d matrix, in which all elements in the matrix are 1; The joiner updates the foraging direction by listening to the location information of the discoverer to compete for high-quality resources. The position of the joiner is updated as follows: c p is the local optimal position of the finder, c worst represents the global worst position; A represents a 1×d matrix, in which each element is randomly assigned a value of 1 or -1, and A+=AT(AAT)-1; when When , it indicates that the i-th participant still has no food and has a low fitness, and needs to find other locations to obtain food resources; The scouts provide the population with timely danger information, accounting for 10% to 20% of the population, and update their positions according to the following formula: c best is the global optimal position in this round of iteration; η is the step size control parameter, which is a normally distributed random number with a mean of 0 and a variance of 1; when f i >f b It means that the sparrow is vulnerable to predators and is on the edge of the population. i =f b When , it indicates that the sparrow is aware of the danger and needs to update its position to reduce the risk of being attacked; the complex industrial process composed of each of the above-mentioned processes must adjust the size of key parameters within a reasonable range to achieve the optimal economic benefits, quality production indicators and energy consumption indicators of the entire process, that is, after the sparrow search algorithm updates the population E times, the individual food resources in the population reach the optimal solution of the area; K∈[-1,1] is a random number, indicating the direction of the sparrow's movement and also the step length control parameter; f i is the fitness value of the current sparrow individual; f b and f w are the current global best and worst fitness values respectively; ε is the smallest constant to avoid zero in the denominator.
8. The joint optimization method for multi-process mixed production according to claim 7 is characterized in that: In each round of population position update process, the decision variable u of each process is i To make adjustments, the decision variables must satisfy u under the premise of achieving production parameters. i,min ≤u i ≤u i,max When producing, the production cost and energy consumption can be reduced to the greatest extent, and the production efficiency can be improved; The state vector of the whole process x=[x1 T ,x2 T ,…,x n T ] T After each process is completed, it is updated and affects the next process. The change of the state vector affects the quality result of the entire industrial production.
