Intelligent control method and system for optimizing production process of high-filling modified material

Abstract

The invention discloses an intelligent control method and system for production process optimization of a high-filling modified material, and relates to the technical field of production process optimization control. Historical process data and a product quality label set of each production process in each production experiment are collected in advance, a quality parameter label of each production process in each production experiment is calculated, a quality prediction model is trained for each production process, and a quality prediction function corresponding to the quality prediction model of each production process is collected. A quality threshold value is preset for the kth production process; for the kth production process of the to-be-produced material, obtaining an optimized production parameter set of the kth production process based on the quality parameter label of the (k-1) th production process and the quality threshold and the quality prediction function of the kth production process; according to the invention, the qualified rate and the production efficiency can be greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of production process optimization control, and in particular to an intelligent control method and system for optimizing the production process of highly filled modified materials. Background Art

[0002] Highly filled modified materials are high-performance composite materials, usually made by physically mixing matrix resins (such as PP, PE and other thermoplastics), high-content inorganic fillers (such as glass fiber, carbon fiber, etc.) and a small amount of additives (such as compatibilizers, flame retardants, etc.). This type of material has the advantages of both the processability of resins and the high strength and high rigidity of fillers, and is widely used in the fields of automobiles, electronics, aviation, etc.

[0003] The production of highly filled modified materials generally requires three main processes:

[0004] Mixing process: Add the base resin, filler and additives according to the proportion, and mix them thoroughly and homogenize them under high temperature and high speed stirring.

[0005] Injection molding / extrusion molding process: The mixed compound is injected into a mold or molded by an extruder under a heated state.

[0006] Cooling process: Control the cooling rate so that the molded product can cool and solidify and obtain the desired geometric shape and performance indicators.

[0007] Traditional control of process parameters mainly relies on manual experience and trial and error. It is impossible to quantify the weights of different quality indicators. The adjustment cycle is long, the labor cost is high, and there is a lack of systematic theoretical model guidance. It is difficult to control and improve the product quality level. Human operation factors will also affect the adjustment effect.

[0008] To this end, the present invention proposes an intelligent control method and system for optimizing the production process of highly filled modified materials. Summary of the invention

[0009] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an intelligent control method and system for optimizing the production process of highly filled modified materials, which greatly improves the qualified product rate and production efficiency.

[0010] To achieve the above objectives, an intelligent control method for optimizing the production process of highly filled modified materials is proposed, comprising the following steps:

[0011] Step 1: Conduct N production experiments in advance, and collect historical process data and product quality label sets of each production process in each production experiment; N is the number of selected production experiments;

[0012] Step 2: Based on the product quality label set, calculate the quality parameter label of each production process in each production experiment;

[0013] Step 3: For the k-th production process, based on the historical process data, quality parameter labels and the product quality label set of the k-1-th production process, a quality prediction model is trained for each production process, and the quality prediction function corresponding to the quality prediction model of each production process is collected; k is the number of the production process sorted in production order;

[0014] Step 4: Preset the quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, obtain the optimized production parameter set of the k-th process.

[0015] Specifically, the method of collecting the historical process data of each production process and the product quality label set in each production experiment is:

[0016] For the kth production process of each production experiment, the raw material data input into the production process and various production parameters of the production process are collected to form the historical process data.

[0017] Specifically, for the kth production process of each production experiment, the parameter values ​​of various quality parameters of the product obtained after the production process is completed are collected to form a product quality label set.

[0018] Specifically, for the mixing process, the quality parameters include filler dispersion uniformity and actual density;

[0019] For injection molding and extrusion processes, the quality parameters include actual dimensions, measured tensile strength values, and appearance integrity;

[0020] For the cooling process, the quality parameters include dimensional change rate, volume shrinkage rate, surface quality fraction, measured impact strength, measured HDT value and surface resistivity.

[0021] Specifically, the method of calculating the quality parameter label of each production process in each production experiment based on the product quality label set is:

[0022] For the mixing process, a mixing quality function is constructed based on the filler dispersion uniformity, actual density, and pre-calculated theoretical density, and the parameter values ​​of each quality parameter of the mixing process of each production experiment are substituted into the mixing quality function. The calculated function value is the quality parameter label of the mixing process;

[0023] For the injection molding extrusion process, an injection molding extrusion quality function is constructed based on the actual size, the measured tensile strength value, the appearance integrity, the preset design size and the preset design tensile strength value, and the parameter values ​​of the various quality parameters of the injection molding extrusion process of each production experiment are substituted into the injection molding extrusion quality function, and the calculated function value is the quality parameter label of the injection molding extrusion process;

[0024] For the cooling process, a cooling quality function is constructed based on the dimensional change rate, volume shrinkage rate, surface quality fraction, measured impact strength, measured HDT value and surface resistivity, the preset allowable dimensional change rate, the preset design shrinkage rate, the preset design impact strength and the collected use temperature. The parameter values ​​of various quality parameters of the cooling process of each production experiment are substituted into the cooling quality function, and the calculated function value is the quality parameter label of the cooling process.

[0025] Specifically, the method of training the quality prediction model for each production process is:

[0026] The raw material data, production parameters of the k-th production process of each production experiment and the labels of various quality parameters of the k-1-th production process are combined to form a set of training feature vectors;

[0027] For the k-th production process, the training feature vector of each production experiment corresponding to the k-th production process is used as the input of the quality prediction model, and the quality prediction model takes the predicted value of the quality parameter of the product of this production process as the output; the quality prediction model takes the quality parameter label of the product of this production process as the prediction target, and the difference between the predicted value of the quality parameter and the quality parameter label as the prediction error, and takes minimizing the sum of squares of prediction errors of all production experiments as the training target; the quality prediction model is trained until the sum of squares of prediction errors converges and the training is stopped; the quality prediction model is a polynomial model.

[0028] Specifically, the method of collecting the quality prediction function corresponding to the quality prediction model of each production process is:

[0029] The polynomial function expression corresponding to the quality prediction model trained for each production process is read as the quality prediction function.

[0030] Specifically, for the kth production process of the material to be produced, based on the quality parameter label of the k-1th production process, the quality threshold of the kth production process and the quality prediction function, the method of obtaining the optimized production parameter set of the kth process is:

[0031] Before the k-th production process is carried out on the production material, the raw material data of the product of the k-1-th production process is collected and marked as the raw material set A, and the parameter values ​​of various quality parameters of the product of the k-1-th production process are collected and marked as the quality set B;

[0032] The function expression of the quality prediction function of the k-th production process is marked as Fk, Fk=fk(A, B, x); wherein fk is the function equation corresponding to the quality prediction model of the k-th production process, and x is the set of production parameter variables of the k-th production process;

[0033] The quality threshold of the k-th production process is marked as Yk;

[0034] Substitute the quality threshold into the quality prediction function, construct the optimization constraint condition: Yk<fk(A, B, x), use genetic algorithm or ant colony algorithm to solve the optimization constraint condition, and obtain the feasible solution set of the production parameter variable set as the optimized production parameter set.

[0035] An intelligent control system for optimizing the production process of highly filled modified materials is proposed, including a test data collection module, a quality calculation module, a model training module and a production control module; wherein each module is electrically connected;

[0036] The test data collection module conducts N production experiments in advance, collects the historical process data and product quality label set of each production process in each production experiment, and sends the historical process data and product quality label set to the model training module, and sends the product quality label set to the quality calculation module;

[0037] The quality calculation module calculates the quality parameter labels of each production process in each production experiment based on the product quality label set, and sends the quality parameter labels to the model training module;

[0038] The model training module, for the k-th production process, trains a quality prediction model for each production process based on historical process data, quality parameter labels, and a set of product quality labels for the k-1-th production process, collects the quality prediction function corresponding to the quality prediction model of each production process, and sends the quality prediction function to the production control module;

[0039] The production control module presets a quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, the optimized production parameter set of the k-th process is obtained.

[0040] An electronic device is proposed, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0041] The processor executes the above-mentioned intelligent control method for optimizing the production process of highly filled modified materials by calling the computer program stored in the memory.

[0042] A computer-readable storage medium is provided, on which a rewritable computer program is stored;

[0043] When the computer program is run on a computer device, the computer device is enabled to execute the above-mentioned intelligent control method for optimizing the production process of highly filled modified materials.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention collects historical process data and product quality label sets of each production process in each production experiment by pre-performing N production experiments; N is the number of selected production experiments, and based on the product quality label set, the quality parameter label of each production process in each production experiment is calculated; for the kth production process, based on the historical process data, the quality parameter label and the product quality label set of the k-1th production process, a quality prediction model is trained for each production process, and a quality prediction function corresponding to the quality prediction model of each production process is collected, and a quality threshold is preset for the kth production process; for the kth production process of the material to be produced, based on the quality parameter label of the k-1th production process, the quality threshold and the quality prediction function of the kth production process, an optimized production parameter set of the kth process is obtained; by constructing a reasonable quality scoring function, the weights of different quality parameters are quantified; historical process data and quality inspection data are collected, and these big data are automatically analyzed by machine learning technology to train a quality prediction model for each process; and then combined with the quality threshold, an optimization algorithm is used to automatically calculate the optimal production parameter set of each process, thereby greatly improving the qualified product rate and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the intelligent control method for optimizing the production process of highly filled modified materials in Example 1 of the present invention;

[0047] Figure 2 This is a module connection relationship diagram of an intelligent control system for optimizing the production process of highly filled modified materials in Example 2 of the present invention;

[0048] Figure 3 This is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention;

[0049] Figure 4This is a schematic diagram of the computer-readable storage medium structure in Example 4 of the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Example 1

[0052] like Figure 1 As shown, the intelligent control method for optimizing the production process of highly filled modified materials includes the following steps:

[0053] Step 1: Conduct N production experiments in advance, and collect historical process data and product quality label sets of each production process in each production experiment; N is the number of selected production experiments;

[0054] Step 2: Based on the product quality label set, calculate the quality parameter label of each production process in each production experiment;

[0055] Step 3: For the k-th production process, based on the historical process data, quality parameter labels and the product quality label set of the k-1-th production process, a quality prediction model is trained for each production process, and the quality prediction function corresponding to the quality prediction model of each production process is collected; k is the number of the production process sorted in production order;

[0056] Step 4: Preset the quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, obtain the optimized production parameter set of the k-th process.

[0057] Among them, the various production processes described in the production process of the highly filled modified material include a mixing process, an injection molding extrusion process and a cooling process.

[0058] In each of the production experiments, several raw materials for producing highly filled modified materials are selected as test raw materials, and the test personnel actively control the various production parameters in each production process to complete the production process from the test raw materials to the highly filled modified materials, thereby recording the various production parameters of each production process and the product quality of the production results of each production process during each production experiment.

[0059] The method of collecting the historical process data and product quality label set of each production process in each production experiment is as follows:

[0060] For the kth production process of each production experiment, the raw material data and various production parameters of the production process are collected to form the historical process data;

[0061] Among them, the raw material data is determined according to the physical properties of the production results of the previous production process. For example, for the mixing process, the raw material data includes but is not limited to the raw material ratio, the melt fluidity (MFR) and density of the matrix resin, the size, aspect ratio, density and chemical composition of the filler, and the polarity, molecular weight, melting point, etc. of the additive; and for the injection molding and extrusion process, the raw material data includes but is not limited to the fluidity and wettability of the mixed high-filler composite material; for the cooling process, the raw material data includes but is not limited to the crystallinity, residual stress and micropores of the high-temperature molded product after injection molding / extrusion.

[0062] For the kth production process of each production experiment, the parameter values ​​of various quality parameters of the product obtained after the production process is completed are collected to form a product quality label set;

[0063] Specifically, for the mixing process, the quality parameters include filler dispersion uniformity and actual density;

[0064] For injection molding and extrusion processes, the quality parameters include actual dimensions, measured tensile strength values, and appearance integrity;

[0065] For the cooling process, the quality parameters include dimensional change rate, volume shrinkage rate, surface quality fraction, measured impact strength, measured HDT value and surface resistivity.

[0066] It should be noted that, for each quality parameter, it is measured by corresponding measuring means after each production process is completed. The measuring method of each quality parameter is a conventional means in the art, and the present invention will not go into details here; for some subjective quality parameters, such as appearance integrity, it can be set to the number of surface cracks, the appearance integrity can be set to the difference between the product shape and the expected shape, and the surface quality score can be set to the number of cracks, flatness, etc. on the surface of the final product.

[0067] Furthermore, the method of calculating the quality parameter label of each production process in each production experiment based on the product quality label set is as follows:

[0068] For the mixing process, a mixing quality function is constructed based on the filler dispersion uniformity, actual density, and pre-calculated theoretical density, and the parameter values ​​of each quality parameter of the mixing process of each production experiment are substituted into the mixing quality function. The calculated function value is the quality parameter label of the mixing process;

[0069] For the injection molding extrusion process, an injection molding extrusion quality function is constructed based on the actual size, the measured tensile strength value, the appearance integrity, the preset design size and the preset design tensile strength value, and the parameter values ​​of the various quality parameters of the injection molding extrusion process of each production experiment are substituted into the injection molding extrusion quality function, and the calculated function value is the quality parameter label of the injection molding extrusion process;

[0070] For the cooling process, a cooling quality function is constructed based on the dimensional change rate, volume shrinkage rate, surface quality fraction, measured impact strength, measured HDT value and surface resistivity, the preset allowable dimensional change rate, the preset design shrinkage rate, the preset design impact strength and the collected use temperature. The parameter values ​​of various quality parameters of the cooling process of each production experiment are substituted into the cooling quality function, and the calculated function value is the quality parameter label of the cooling process.

[0071] In a preferred embodiment, the mixing quality function can be set to: Q_mix=30×U+20×(100-5×|Dm-Dt|); wherein: Q_mix is ​​the mixing quality function, U is the filler dispersion uniformity fraction, Dm is the actual density, and Dt is the theoretical density;

[0072] The injection molding extrusion mass fraction formula can be set as: Where: Q_mold is the injection molding extrusion quality function, L is the actual size, Lt is the design size, Ts is the measured tensile strength value, Td is the design tensile strength value, and Wg is the appearance integrity;

[0073] The cooling quality function can be set as:

[0074] Among them, Q_final is the cooling quality function, Sv is the dimensional change rate, Sd is the allowable dimensional change rate, Vr is the volume shrinkage rate, Vd is the design shrinkage rate, S is the surface mass fraction, Is is the measured impact strength, Id is the design impact strength, Ht is the measured HDT value, Hu is the operating temperature, and Rs is the surface resistivity.

[0075] It should be noted that the design principles of the parameter proportions and mathematical formulas in the mixing mass function, injection extrusion mass function and cooling mass function are:

[0076] For the mixing process, filler dispersion uniformity and density are the two most critical indicators, accounting for 60% and 40% of their importance in production respectively;

[0077] The injection molding / extrusion process mainly focuses on dimensional accuracy (35%), tensile strength (35%) and appearance (30%), and their importance in production accounts for 35%, 35% and 30% respectively;

[0078] In the cooling quality function: dimensional change rate (25%) and volume shrinkage rate (25%) are key indicators of the cooling process, accounting for half of the total score, reflecting their important influence on the quality of the final product. Surface quality (15%) is second, because for many applications, good surface quality is also a basic requirement. Impact strength (25%) and heat deformation temperature HDT (20%) are two key mechanical indicators reflecting the impact resistance and thermal properties of the final product, accounting for a high proportion. Other properties, such as surface resistivity (15%), although less important than the above items, cannot be ignored.

[0079] The rationality of mathematical expression design includes:

[0080] |Dm-Dt| represents the absolute deviation of density from the theoretical value, and 5 is the empirical penalty coefficient.

[0081] It means that if the dimensional error is within 0.05mm, full marks will be awarded, and points will be deducted if it exceeds the linearity.

[0082] It represents the ratio of tensile strength to design value. If it is greater than 1, full marks will be awarded, and if it is less than 1, points will be deducted proportionally.

[0083] In the form of 1-10×|Sv-Sd| and 1-10×|Vr-Vd|, points will be awarded if dimensional changes and shrinkage are strictly controlled within the allowable range. Exceeding the range will result in serious point deductions.

[0084] It represents the ratio of impact strength to design value, and points are also given proportionally.

[0085] It means that if the difference between HDT and operating temperature is more than 30%, full marks will be awarded, otherwise points will be deducted linearly according to the percentage of the difference.

[0086] |Log(Rs)-5.5| indicates that the resistivity is (10 5 ,10 12 ) ohm range, full marks will be awarded, and points will be deducted based on the logarithmic difference when deviating from this range.

[0087] It can be understood that the mixing quality function, injection extrusion quality function and cooling quality function are actual values ​​calculated based on the facts of established physical parameters after the product is produced, and the quality parameters of the products in each production process are jointly determined by the raw material data, production parameters and the quality of the raw materials (i.e., the quality parameters of the previous production parameters). Therefore, a model is needed to predict the quality parameters of the next production process.

[0088] Furthermore, for the k-th production process, based on the historical process data, the quality parameter labels and the product quality label set of the k-1-th production process, the method of training the quality prediction model for each production process is as follows:

[0089] The raw material data, production parameters of the k-th production process of each production experiment and the labels of various quality parameters of the k-1-th production process are combined to form a set of training feature vectors;

[0090] For the k-th production process, the training feature vector of each production experiment corresponding to the k-th production process is used as the input of the quality prediction model, and the quality prediction model takes the predicted value of the quality parameter of the product of this production process as the output; the quality prediction model takes the quality parameter label of the product of this production process as the prediction target, and the difference between the predicted value of the quality parameter and the quality parameter label as the prediction error, and takes minimizing the sum of squares of prediction errors of all production experiments as the training target; the quality prediction model is trained until the sum of squares of prediction errors converges and the training is stopped; the quality prediction model is a polynomial model.

[0091] Furthermore, the method of collecting the quality prediction function corresponding to the quality prediction model of each production process is:

[0092] Read the polynomial function expression corresponding to the quality prediction model trained for each production process as the quality prediction function;

[0093] It can be understood that the quality prediction function is a polynomial function with the quality parameter as the dependent variable and the raw material data production parameter and the quality of the raw material as the independent variables.

[0094] Furthermore, the quality threshold is the minimum threshold of each quality parameter of the products of each production process, and only products that reach the quality threshold can be regarded as qualified products.

[0095] For the kth production process of the material to be produced, based on the quality parameter label of the k-1th production process, the quality threshold of the kth production process and the quality prediction function, the method of obtaining the optimized production parameter set of the kth process is:

[0096] Before the k-th production process is carried out on the production material, the raw material data of the product of the k-1-th production process is collected and marked as the raw material set A, and the parameter values ​​of various quality parameters of the product of the k-1-th production process are collected and marked as the quality set B;

[0097] The function expression of the quality prediction function of the k-th production process is marked as Fk, Fk=fk(A, B, x); wherein fk is the function equation corresponding to the quality prediction model of the k-th production process, and x is the set of production parameter variables of the k-th production process;

[0098] The quality threshold of the k-th production process is marked as Yk;

[0099] Substitute the quality threshold into the quality prediction function, construct the optimization constraint condition: Yk<fk(A, B, x), use the genetic algorithm or ant colony algorithm to find a feasible solution to the optimization constraint condition, and obtain a feasible solution set of the production parameter variable set as the optimized production parameter set;

[0100] It can be understood that the combination of various production parameters in the feasible solution set can be used to control various production parameters in the kth production process based on raw material data and raw material quality, thereby ensuring that qualified products are obtained in this production process.

[0101] Example 2

[0102] like Figure 2 As shown, the intelligent control system for optimizing the production process of highly filled modified materials includes a test data collection module, a quality calculation module, a model training module and a production control module; wherein each module is electrically connected;

[0103] The test data collection module conducts N production experiments in advance, collects the historical process data and product quality label set of each production process in each production experiment, and sends the historical process data and product quality label set to the model training module, and sends the product quality label set to the quality calculation module;

[0104] The quality calculation module calculates the quality parameter labels of each production process in each production experiment based on the product quality label set, and sends the quality parameter labels to the model training module;

[0105] The model training module, for the k-th production process, trains a quality prediction model for each production process based on historical process data, quality parameter labels, and a set of product quality labels for the k-1-th production process, collects the quality prediction function corresponding to the quality prediction model of each production process, and sends the quality prediction function to the production control module;

[0106] The production control module presets a quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, the optimized production parameter set of the k-th process is obtained.

[0107] Example 3

[0108] Figure 3 Schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, according to another aspect of the present application, an electronic device 100 is also provided. The electronic device 100 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are run by one or more processors, the intelligent control method for optimizing the production process of highly filled modified materials as described above may be executed.

[0109] The method or device according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3 As shown, the electronic device 100 may include a bus 101, one or more CPUs 102, a ROM 103, a RAM 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as the ROM 103 or the hard disk 107, may store the intelligent control method for optimizing the production process of highly filled modified materials provided in the present application.

[0110] Furthermore, the electronic device 100 may further include a user interface 108. Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0111] Example 4

[0112] Figure 4 Schematic diagram of the computer-readable storage medium structure provided by an embodiment of the present application. Figure 4As shown, it is a computer-readable storage medium 200 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, the intelligent control method for optimizing the production process of highly filled modified materials according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0113] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0114] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0115] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0116] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0117] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0118] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent control method for optimizing the production process of highly filled modified materials, characterized in that: The following steps are involved: Step 1: Conduct N production experiments in advance, and collect historical process data and product quality label sets of each production process in each production experiment; N is the number of selected production experiments; Step 2: Based on the product quality label set, calculate the quality parameter label of each production process in each production experiment; Step 3: For the k-th production process, based on the historical process data, quality parameter labels and the product quality label set of the k-1-th production process, a quality prediction model is trained for each production process, and the quality prediction function corresponding to the quality prediction model of each production process is collected; k is the number of the production process sorted in production order; Step 4: Preset the quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, obtain the optimized production parameter set of the k-th process.

2. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 1, characterized in that: For the kth production process of each production experiment, the raw material data input into the production process and various production parameters of the production process are collected to form the historical process data.

3. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 1, characterized in that: For the kth production process of each production experiment, the parameter values ​​of various quality parameters of the product obtained after the production process is completed are collected to form a product quality label set.

4. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 3, characterized in that: The method of calculating the quality parameter label of each production process in each production experiment based on the product quality label set is as follows: For the mixing process, a mixing quality function is constructed based on the filler dispersion uniformity, actual density, and pre-calculated theoretical density, and the parameter values ​​of each quality parameter of the mixing process of each production experiment are substituted into the mixing quality function. The calculated function value is the quality parameter label of the mixing process; For the injection molding extrusion process, an injection molding extrusion quality function is constructed based on the actual size, the measured tensile strength value, the appearance integrity, the preset design size and the preset design tensile strength value, and the parameter values ​​of the various quality parameters of the injection molding extrusion process of each production experiment are substituted into the injection molding extrusion quality function, and the calculated function value is the quality parameter label of the injection molding extrusion process; For the cooling process, a cooling quality function is constructed based on the dimensional change rate, volume shrinkage rate, surface quality fraction, measured impact strength, measured HDT value and surface resistivity, the preset allowable dimensional change rate, the preset design shrinkage rate, the preset design impact strength and the collected use temperature. The parameter values ​​of various quality parameters of the cooling process of each production experiment are substituted into the cooling quality function, and the calculated function value is the quality parameter label of the cooling process.

5. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 4, characterized in that: The method of training the quality prediction model for each production process is as follows: The raw material data, production parameters of the k-th production process of each production experiment and the labels of various quality parameters of the k-1-th production process are combined to form a set of training feature vectors; For the k-th production process, the training feature vector of each production experiment corresponding to the k-th production process is used as the input of the quality prediction model, and the quality prediction model takes the predicted value of the quality parameter of the product of this production process as the output; the quality prediction model takes the quality parameter label of the product of this production process as the prediction target, and the difference between the predicted value of the quality parameter and the quality parameter label as the prediction error, and takes minimizing the sum of squares of prediction errors of all production experiments as the training target; the quality prediction model is trained until the sum of squares of prediction errors converges and the training is stopped; the quality prediction model is a polynomial model.

6. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 5, characterized in that: The method of collecting the quality prediction function corresponding to the quality prediction model of each production process is: The polynomial function expression corresponding to the quality prediction model trained for each production process is read as the quality prediction function.

7. The intelligent control method for optimizing the production process of highly filled modified materials according to claim 6, characterized in that: For the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, the method of obtaining the optimized production parameter set of the k-th process is: Before the k-th production process is carried out on the production material, the raw material data of the product of the k-1-th production process is collected and marked as the raw material set A, and the parameter values ​​of various quality parameters of the product of the k-1-th production process are collected and marked as the quality set B; The function expression of the quality prediction function of the k-th production process is marked as Fk, Fk=fk(A, B, x); wherein fk is the function equation corresponding to the quality prediction model of the k-th production process, and x is the set of production parameter variables of the k-th production process; The quality threshold of the k-th production process is marked as Yk; Substitute the quality threshold into the quality prediction function, construct the optimization constraint condition: Yk<fk(A, B, x), use genetic algorithm or ant colony algorithm to solve the optimization constraint condition, and obtain the feasible solution set of the production parameter variable set as the optimized production parameter set.

8. An intelligent control system for optimizing the production process of highly filled modified materials, which is used to implement the intelligent control method for optimizing the production process of highly filled modified materials as described in any one of claims 1 to 7, characterized in that: It includes a test data collection module, a quality calculation module, a model training module and a production control module; wherein each module is electrically connected; The test data collection module conducts N production experiments in advance, collects the historical process data and product quality label set of each production process in each production experiment, and sends the historical process data and product quality label set to the model training module, and sends the product quality label set to the quality calculation module; The quality calculation module calculates the quality parameter labels of each production process in each production experiment based on the product quality label set, and sends the quality parameter labels to the model training module; The model training module, for the k-th production process, trains a quality prediction model for each production process based on historical process data, quality parameter labels, and a set of product quality labels for the k-1-th production process, collects the quality prediction function corresponding to the quality prediction model of each production process, and sends the quality prediction function to the production control module; The production control module presets a quality threshold for the k-th production process; for the k-th production process of the material to be produced, based on the quality parameter label of the k-1-th production process, the quality threshold of the k-th production process and the quality prediction function, the optimized production parameter set of the k-th process is obtained.

9. An electronic device, characterized in that: include: processor and memory, wherein The memory stores a computer program that can be called by the processor; The processor executes the intelligent control method for optimizing the production process of highly filled modified materials as described in any one of claims 1 to 7 in the background by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: A rewritable computer program is stored thereon; When the computer program is run on a computer device, the computer device executes the intelligent control method for optimizing the production process of highly filled modified materials as described in any one of claims 1 to 7 in the background.

Images