Intelligent formula method and system for rubber extrusion

Abstract

The invention relates to a rubber extrusion formula technology, and discloses a rubber extrusion intelligent formula method and system, and the method comprises the steps: an intelligent control management end carries out the clustering of recent production data through employing a mean-shift clustering algorithm, and calculates the recent optimal formula information of each specification; the intelligent control management end uses historical production data to regress influence factors of different sizing materials and different production line states on a formula through a random forest algorithm; during production, the intelligent control management end obtains specification clustering from the data platform through PLC switching rubber specifications, obtains optimal formula information and automatically issues the optimal formula information to a machine; after production for a period of time, the intelligent terminal obtains sizing material and production line information, transmits the sizing material and production line information to the random forest model, obtains standard correction and issues the standard correction to the machine. The rubber extrusion formula disclosed by the invention is high in efficiency and good in stability and robustness.

Description

Technical Field

[0001] The invention relates to a rubber extrusion formulation technology, and in particular to a rubber extrusion intelligent formulation method and system. Background Art

[0002] Semi-finished product extrusion is one of the four major steps in tire manufacturing. It is a process that allows the rubber to be continuously made into semi-finished products of various shapes through an extruder. After the rubber is filled into the barrel from the feed port, the rotation of the screw causes the rubber to be subjected to a strong extrusion effect between the screw and the barrel wall, forcing the rubber to continuously move forward and pass through the die, thereby obtaining semi-finished products of various shapes. The extrusion process generally includes rubber heat refining, extrusion, cooling, cutting and receiving processes; except for heat refining, other processes are linked operations, and the total length of the production line is 120 meters to 200 meters. Summary of the invention

[0003] The present invention aims at the problem that the rubber extrusion process in the prior art is still mainly based on manual control or expert system. The parameters and formula parameters in the production process are all recorded in the notebook by the operator, and then adjusted according to experience during production, and the effect is good or bad, and it is closely related to the operator's experience; the automated program written by the operator's experience generally cannot adapt to all specifications, and its robustness and stability are relatively poor. The present invention provides a rubber extrusion intelligent formula method and system.

[0004] In order to solve the above technical problems, the present invention is solved by the following technical solutions:

[0005] A rubber extrusion intelligent formula method includes an intelligent control management end, an intelligent terminal and a machine; the method includes:

[0006] The intelligent control management terminal uses the mean-shift clustering algorithm to cluster recent production data and calculate the recent optimal formula information for each specification;

[0007] The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm;

[0008] During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine;

[0009] After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

[0010] Preferably, the recent production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width and rear weight.

[0011] Preferably, the historical production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width, rear weight, upper extruder pressure, middle extruder pressure, lower extruder pressure, upper extruder temperature, middle extruder temperature, lower extruder temperature, upper extruder current, middle extruder current, lower extruder current, rubber Mooney value, rubber hardness, rubber specific gravity and rubber sulfur grade.

[0012] As a preferred method, the implementation of clustering recent production data using the mean-shift clustering algorithm includes:

[0013] Acquisition of recent production data: The intelligent control management terminal collects recent production data and classifies the acquired recent production data;

[0014] Preprocessing of recent production data: Preprocessing of recent production data through time-lagged cross-correlation algorithm;

[0015] Initialization of the processed recent production data. A piece of data is randomly selected as the starting center point x for the processed recent production data. 起始 ;

[0016] The bandwidth parameter h is estimated by the estimate_bandwidth function to determine the updated center point position x 更新， x 更新 =x 起始 +m(x);

[0017] m(x) is the drift vector,

[0018]

[0019] Among them, K h (u) is a Gaussian kernel function with bandwidth h,

[0020]

[0021] The kernel function satisfies

[0022] By obtaining the drift vector, when ||m(x)||<ε or the number of iterations reaches the set maximum value T, the iteration is stopped, otherwise it returns to the above steps to continue, thereby obtaining the optimal rubber formula for recent production data.

[0023] As a preferred method, the time-lagged cross-correlation algorithm pre-processes the recent production data including:

[0024] The weight sequence before setting is X = {x1, x2, ..., xnn}

[0025] The weight sequence is Y = {y1, y2, ..., y n}, where n is the length;

[0026] Time lagged cross-correlation function R XY (τ) represents the correlation between time series Y and X at a time lag of τ,

[0027]

[0028] Where, τ is the time lag; is the mean of the time series X and Y, is the mean of the time series XY; x t The value of the time series X at time points τ and t+τ, y t+τ are the values ​​of time series X and Y at time points τ and t+τ respectively; τ max = argmax τ |R XY (τ)|

[0029] Among them, τ max The amount of weight you need to put in advance for the total width of your rear shoulders.

[0030] As a preferred method, the implementation of regressing the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm includes:

[0031] Acquisition of historical production data: The intelligent control management terminal collects historical production data;

[0032] Preprocessing of historical production data: Preprocessing of historical production data through time-lagged cross-correlation algorithm;

[0033] By building a random forest model on the preprocessed historical production data;

[0034] The parameters of the random forest model are optimized, and the random forest model is trained using the optimized parameters.

[0035] As a preferred method, the parameters of the random forest model are optimized, the random forest model is trained by the optimized parameters, and the random forest model is optimized by tree-based Parson optimization. The implementation methods include:

[0036] Select an initial hyperparameter x to calculate two probability models L(x) and G(x);

[0037] l(x) is the probability that the objective function value is lower than a certain threshold γ under a given hyperparameter x; L(x) = P(x|y<γ)

[0038] G(x) is the probability that the objective function value is higher than a certain threshold γ under a given hyperparameter x (i.e., the parameter space of "bad"): G(x) = P(x|y≥γ)

[0039] p(y|x) is the conditional probability density function of the objective function value y given the hyperparameter x:

[0040] P(y|x)=L(x)·L(y<γ)+G(x)·G(y≥γ)

[0041] Use EI(x) to select hyperparameters x for evaluation and record the objective function value. EI(x) is used to measure the expected improvement of the objective function value relative to the current optimal value γ after selecting a set of hyperparameters x.

[0042]

[0043] Among them, μ g and σ g are the mean and standard deviation of the objective function value under G(x), Φ(…) and φ(…) are the cumulative distribution function and probability density function of the standard normal distribution, respectively.

[0044] In order to solve the above technical problems, the present application also provides a rubber extrusion intelligent formula system, including an intelligent control management end, an intelligent terminal and a machine; characterized in that: it also includes,

[0045] The module for obtaining the optimal formula information: the intelligent control management terminal uses the mean-shift clustering algorithm to cluster the recent production data and calculate the recent optimal formula information for each specification;

[0046] Random forest regression module: The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm;

[0047] During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine;

[0048] After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

[0049] The present invention has significant technical effects due to the adoption of the above technical solution:

[0050] This system consists of a formula model and a dynamic control algorithm, and uses clustering and iteration to adjust parameters. Through the digital platform, 120,000 pieces of production line speed, meter weight and single weight data are captured, and the Meanshift clustering model is used to analyze and generate new production formula data including meter weight, extrusion speed, rotation speed, etc., which are sent to the on-site client for production and iteratively updated on a 7-day week basis. The (Random Forest) algorithm is used to predict the extrusion results based on information such as rubber, specifications, and environment, and the optimal production parameters are corrected and recommended in real time, and automatically sent to the machine with one click; (XGBoost, MLP) The algorithm dynamically adjusts the extrusion meter weight and production line speed in real time based on the extrusion current, pressure, temperature, meter weight, etc.; the temperature and meter weight measured at 10 meters of extrusion are used by the AI ​​model to predict the weight and width of the cooled tread after 200 meters.

[0051] The rubber extrusion formula of the invention has high efficiency, good stability and robustness, shortens the response time by 95%, increases the weight qualification rate from 90.10% to 95.90%, and increases the size qualification rate from 91.86% to 97.14%. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of the rubber extrusion formula structure of the present invention

[0053] Figure 2 It is a flow chart of the present invention.

[0054] Figure 3 It is a flow chart for obtaining the optimal formula information of the present invention.

[0055] Figure 4 It is a training flow chart of the random forest module of the present invention. DETAILED DESCRIPTION

[0056] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0057] Example 1

[0058] A rubber extrusion intelligent formula method includes an intelligent control management end, an intelligent terminal and a machine; the method includes:

[0059] The intelligent control management terminal uses the mean-shift clustering algorithm to cluster recent production data and calculate the recent optimal formula information for each specification;

[0060] The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm;

[0061] During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine;

[0062] After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

[0063] Recent production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width and rear weight.

[0064] Historical production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width, rear weight, upper extruder pressure, middle extruder pressure, lower extruder pressure, upper extruder temperature, middle extruder temperature, lower extruder temperature, upper extruder current, middle extruder current, lower extruder current, rubber Mooney value, rubber hardness, rubber specific gravity and rubber sulfur change grade.

[0065] In Table 1, there are rubber specifications AH and their corresponding recent production data.

[0066] Table 1

[0067] Production line number Specification Update time Product Type Linkage line speed Upper extruder speed Medium extruder speed Lower extruder speed Shoulder width standard Overall width standard Weight Standard 1 A 16 / 3 / 2024 02：25：06 1 10.23 17.32 15.29 8.4 213.4 253.3 6.93 1 B 3 / 12 / 2023 05：3840 1 12.41 15.36 11.4 8.55 192.4 271.03 4.23 1 C 20 / 8 / 2024 09：19：19 1 10.36 17.1 15.61 8.17 198.54 297.71 6.7 1 D 11 / 9 / 2024 08：0023 1 11.97 18.8 15.43 12.09 198.73 275.63 5.74 1 E 5 / 11 / 2024 03：1854 1 13.08 14.97 8.92 8.96 189.96 268.05 4.24 1 F 26 / 4 / 2024 08：59：04 1 12.47 16.96 11.82 9.15 213.4 253.3 4.25 1 G 22 / 9 / 2024 10：5320 1 11.99 16.93 14.8 10.59 197.09 289.51 5.57 1 H 31 / 5 / 2024 08：19：14 1 10.52 16.4 14.34 9.84 198.42 283.39 5.76

[0068] The implementation of the mean-shift clustering algorithm to cluster recent production data includes:

[0069] Acquisition of recent production data: The intelligent control management terminal collects recent production data and classifies the acquired recent production data;

[0070] Preprocessing of recent production data: Preprocessing of recent production data through time-lagged cross-correlation algorithm;

[0071] Initialization of the processed recent production data. A piece of data is randomly selected as the starting center point x for the processed recent production data. 起始 ;

[0072] The bandwidth parameter h is estimated by the estimate_bandwidth function to determine the updated center point position x 更新， x 更新 =x 起始 +m(x);

[0073] m(x) is the drift vector,

[0074]

[0075] Among them, K h (u) is a Gaussian kernel function with bandwidth h,

[0076]

[0077] The kernel function satisfies

[0078] By obtaining the drift vector, when ||m(x)||<ε or the number of iterations reaches the set maximum value T, the iteration is stopped, otherwise it returns to the above steps to continue, thereby obtaining the optimal rubber formula for recent production data.

[0079] The time-lagged cross-correlation algorithm is used to pre-process the recent production data including

[0080] The weight sequence before setting is X = {x1, x2, ..., xn n}

[0081] The weight sequence is Y = {y1, y2, ..., y n}, where n is the length;

[0082] Time lagged cross-correlation function R XY (τ) represents the correlation between time series Y and X at a time lag of τ,

[0083]

[0084] Where, τ is the time lag; is the mean of the time series X and Y, is the mean of the time series XY; x t The value of the time series X at time points τ and t+τ, y t+τ are the values ​​of time series X and Y at time points τ and t+τ respectively; τ max = argmax τ |R XY (τ)|

[0085] Among them, τ max The amount of weight you need to put in advance for the total width of your rear shoulders.

[0086] The implementation of regressing the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm includes:

[0087] Acquisition of historical production data: The intelligent control management terminal collects historical production data;

[0088] Preprocessing of historical production data: Preprocessing of historical production data through time-lagged cross-correlation algorithm;

[0089] By building a random forest model on the preprocessed historical production data;

[0090] The parameters of the random forest model are optimized, and the random forest model is trained using the optimized parameters.

[0091] Optimize the parameters of the random forest model, train the random forest model with the optimized parameters, and optimize the random forest model based on the tree-based Parson optimization. The implementation methods include:

[0092] Select an initial hyperparameter x to calculate two probability models L(x) and G(x);

[0093] l(x) is the probability that the objective function value is lower than a certain threshold γ under a given hyperparameter x; L(x) = P(x|y<γ)

[0094] G(x) is the probability that the objective function value is higher than a certain threshold γ under a given hyperparameter x (i.e., the parameter space of "bad"): G(x) = P(x|y≥γ)

[0095] p(y|x) is the conditional probability density function of the objective function value y given the hyperparameter x:

[0096] P(y|x)=L(x)·L(y<γ)+G(x)·G(y≥γ)

[0097] Use EI(x) to select hyperparameters x for evaluation and record the objective function value. EI(x) is used to measure the expected improvement of the objective function value relative to the current optimal value γ after selecting a set of hyperparameters x.

[0098]

[0099] Among them, μ g and σ g are the mean and standard deviation of the objective function value under G(x), Φ(…) and φ(…) are the cumulative distribution function and probability density function of the standard normal distribution, respectively.

[0100] Example 2

[0101] Based on Example 1, this embodiment is a rubber extrusion intelligent formula system, including an intelligent control management end, an intelligent terminal and a machine; it also includes,

[0102] The module for obtaining the optimal formula information: the intelligent control management terminal uses the mean-shift clustering algorithm to cluster the recent production data and calculate the recent optimal formula information for each specification;

[0103] Random forest regression module: The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm;

[0104] During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine;

[0105] After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

[0106] Figure 1 In the process, the data from MES, 3D camera and PLC are transmitted to the process recipe model for learning, and then transmitted to the management and control platform for storage. The management and control platform sends semi-finished product AI smart terminals, and the smart terminals send them to the production line. The AI ​​predictive control model learned from the 3D camera and PLC data is deployed on the semi-finished product AI smart terminal to control the production line and achieve the effect of automatic control.

Claims

1. A rubber extrusion intelligent formula method, including an intelligent control management end, an intelligent terminal and a machine; The methods include: The intelligent control management terminal uses the mean-shift clustering algorithm to cluster recent production data and calculate the recent optimal formula information for each specification; The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm; During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine; After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

2. The method for rubber extrusion intelligent formulation according to claim 1, characterized in that: Recent production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width and rear weight.

3. The intelligent formulation method for rubber extrusion according to claim 1, characterized in that: Historical production data include rubber specifications, linkage line speed, upper extruder speed, middle extruder speed, lower extruder speed, front shoulder width, front total width, front weight, rear shoulder width, rear total width, rear weight, upper extruder pressure, middle extruder pressure, lower extruder pressure, upper extruder temperature, middle extruder temperature, lower extruder temperature, upper extruder current, middle extruder current, lower extruder current, rubber Mooney value, rubber hardness, rubber specific gravity and rubber sulfur change grade.

4. The method for rubber extrusion intelligent formulation according to claim 1, characterized in that: The implementation of the mean-shift clustering algorithm to cluster recent production data includes: Acquisition of recent production data: The intelligent control management terminal collects recent production data and classifies the acquired recent production data; Preprocessing of recent production data: Preprocessing of recent production data through time-lagged cross-correlation algorithm; Initialization of the processed recent production data. A piece of data is randomly selected as the starting center point x for the processed recent production data. 起始 ; The bandwidth parameter h is estimated by the estimate_bandwidth function to determine the updated center point position x 更新， x 更新 =x 起始 +m(x); m(x) is the drift vector, Among them, K h (u) is a Gaussian kernel function with bandwidth h, The kernel function satisfies By obtaining the drift vector, when ||m(x)||<ε or the number of iterations reaches the set maximum value T, the iteration is stopped, otherwise it returns to the above steps to continue, thereby obtaining the optimal rubber formula for recent production data.

5. The intelligent formulation method for rubber extrusion according to claim 1, characterized in that: The time-lagged cross-correlation algorithm is used to pre-process the recent production data including The weight sequence before setting is X = {x1, x2, ..., xn n } The weight sequence is Y = {y1, y2, ..., y n }, where n is the length; Time lagged cross-correlation function R XY (τ) represents the correlation between time series Y and X at a time lag of τ, Where, τ is the time lag; is the mean of the time series X and Y, is the mean of the time series XY; x t The value of the time series X at time points τ and t+τ, y t+τ are the values ​​of time series X and Y at time points τ and t+τ respectively; τ max = argmax τ |R XY (τ)| Among them, τ max The amount of weight you need to put in advance for the total width of your rear shoulders.

6. The intelligent formulation method for rubber extrusion according to claim 1, characterized in that: The implementation of regressing the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm includes: Acquisition of historical production data: The intelligent control management terminal collects historical production data; Preprocessing of historical production data: Preprocessing of historical production data through time-lagged cross-correlation algorithm; By building a random forest model on the preprocessed historical production data; The parameters of the random forest model are optimized, and the random forest model is trained using the optimized parameters.

7. The intelligent formulation method for rubber extrusion according to claim 1, characterized in that: Optimize the parameters of the random forest model, train the random forest model with the optimized parameters, and optimize the random forest model based on the tree-based Parson optimization. The implementation methods include: Select an initial hyperparameter x to calculate two probability models L(x) and G(x); l(x) is the probability that the objective function value is lower than a certain threshold γ under a given hyperparameter x; L(x) = P(x|y<γ) G(x) is the probability that the objective function value is higher than a certain threshold γ under a given hyperparameter x (i.e., the parameter space of "bad"): G(x) = P(x|y≥γ) p(y|x) is the conditional probability density function of the objective function value y given the hyperparameter x: P(y|x)=L(x)·L(y<γ)+G(x)·G(y≥γ) Use EI(x) to select hyperparameters x for evaluation and record the objective function value. EI(x) is used to measure the expected improvement of the objective function value relative to the current optimal value γ after selecting a set of hyperparameters x. Among them, μ g and σ g are the mean and standard deviation of the objective function value under G(x), Φ(…) and φ(…) are the cumulative distribution function and probability density function of the standard normal distribution, respectively.

8. A rubber extrusion intelligent formula system, comprising an intelligent control management end, an intelligent terminal and a machine; characterized in that: Also includes, The module for obtaining the optimal formula information: the intelligent control management terminal uses the mean-shift clustering algorithm to cluster the recent production data and calculate the recent optimal formula information for each specification; Random forest regression module: The intelligent control management end uses historical production data to regress the influencing factors of different rubber materials and different production line conditions on the formula through the random forest algorithm; During production, the intelligent control management end switches the rubber material specifications through PLC, obtains the specification clustering from the data platform, obtains the optimal formula information, and automatically sends it to the machine; After a period of production, the intelligent terminal obtains the rubber material and production line information, transmits it to the random forest model, obtains standard corrections and sends it to the machine.

Images