Method for increasing sidewall turn-up height in all-steel radial tire forming process

Abstract

The invention discloses a method for increasing the sidewall turn-up height in a building process of an all-steel radial tire, which comprises the following steps of: acquiring multi-source data such as temperature, pressure and angle in a turn-up process through an infrared temperature sensor, a pressure sensor and the like; and constructing a multiple linear regression model, a machine learning model and a hybrid model fused with the physical principle, analyzing the relationship between the process parameters and the anti-package height, and performing verification and optimization. According to model output parameters, a servo motor is used for accurately controlling the downward pressing angle and pressure of a reverse wrapping pressing roller, and production is executed in cooperation with an auxiliary supporting ring and a sectional type reverse wrapping technology; meanwhile, the turn-up height is detected in real time through an intelligent monitoring system, and parameters are automatically adjusted when the turn-up height does not reach the standard. According to the method, precise regulation and control of the tire sidewall turn-up height are achieved, compared with a traditional process, the product percent of pass is increased to 95% or above from 85%, the production efficiency is improved by 20%-30%, the tire structure stability is effectively enhanced, the rolling resistance and the tire burst risk are reduced, and intelligent upgrading of tire manufacturing is promoted.

Description

Technical Field

[0001] The invention relates to the technical field of tire processing, in particular to a method for increasing the sidewall turn-up height during the molding process of an all-steel radial tire. Background Art

[0002] During the manufacturing process of all-steel radial tires, sidewall turnup height is a key indicator of tire quality and performance. Sufficient sidewall turnup height strengthens the connection between the bead and the carcass, effectively dissipating the impact forces and internal air pressure experienced by the bead during driving, thereby preventing structural issues such as cord-rubber separation and bead delamination. Furthermore, a reasonable sidewall turnup height helps maintain the overall stability of the tire's shape, reduces rolling resistance, mitigates localized wear caused by tire deformation, and improves tire durability and service life. Furthermore, sidewall turnup height directly impacts tire safety and comfort, optimizing stress distribution, reducing the risk of blowouts, and enhancing shock absorption.

[0003] However, the existing all-steel radial tire molding process has significant defects in controlling the sidewall turnup height. On the one hand, traditional processes often rely on the operator's experience, judgment and manual adjustment, and lack precise control of key process parameters (such as temperature, pressure, turnup speed, etc.). For example, during the turnup process, too low a temperature will lead to insufficient rubber viscosity, making it difficult to fit tightly with the cord layer; uneven pressure may cause inconsistent turnup height, affecting the overall performance of the tire. On the other hand, the existing technology has not yet established a scientific mathematical model to predict and optimize the sidewall turnup height, and it is impossible to achieve systematic analysis and precise control of complex process steps. During the production process, due to the inability to obtain the quantitative relationship between process parameters and turnup height in a timely manner, problems such as substandard turnup height and unstable tire quality often occur, resulting in low production efficiency and difficulty in improving product qualification rate.

[0004] Furthermore, existing data collection and processing methods are relatively backward. Traditional methods often rely on single-point measurement or intermittent testing, which cannot fully and effectively capture dynamic data such as temperature and pressure during the sidewall turnup process. Furthermore, they lack effective data processing and in-depth analysis, making it difficult to uncover the underlying process patterns and provide strong support for process optimization. As the tire manufacturing industry continues to increase its demands for product quality and production efficiency, a method that can precisely control the height of the sidewall turnup is urgently needed. By scientifically collecting and analyzing key process parameters, building a reliable mathematical model, and combining it with advanced automated control technology, refined management of the sidewall turnup process can be achieved, thereby improving the overall performance and market competitiveness of all-steel radial tires.

[0005] Based on this, a method for increasing the sidewall turn-up height during the molding process of an all-steel radial tire is now provided, which can eliminate the disadvantages of the existing device. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for increasing the sidewall turn-up height during the molding process of an all-steel radial tire, thereby solving the problem of inconvenience in use in the prior art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for increasing the sidewall turnup height during the molding process of an all-steel radial tire comprises the following steps:

[0009] Step 1: Collect data such as temperature, pressure, roll angle and speed, and auxiliary support ring telescopic displacement during the sidewall turnup process. Combined with process parameters such as tire specifications and rubber formulation, this data set is generated.

[0010] Step 2: Filter, time synchronize and normalize the collected raw data;

[0011] Step 3: Build a hybrid model that includes a multiple linear regression model, a machine learning model, and physical principles;

[0012] Step 4: Compare the anti-packet height predicted by the model with the actual measured value for verification and optimization;

[0013] Step 5: Based on the model prediction results, a high-precision turn-up roller driven by a servo motor is used to turn up the tire sidewall, and the auxiliary support ring and turn-up stage process parameters are controlled. At the same time, the intelligent monitoring system is used to adjust the turn-up process in real time.

[0014] On the basis of the above technical solutions, the present invention also provides the following optional technical solutions:

[0015] In an optional solution: in the data collection step, 3-5 measuring points are evenly arranged in the turn-up area by an infrared temperature sensor to collect temperature change data of the sidewall rubber surface in the initial, middle and final turn-up stages, and a thermocouple sensor is used to monitor the temperature of the turn-up roller; the turn-up pressure data is collected by a surface pressure sensor of the turn-up roller, and the support pressure data is collected by a micro pressure sensor at the contact point between the auxiliary support ring and the sidewall; the downward pressure angle and speed data of the turn-up roller are collected by an encoder, and the telescopic displacement of the auxiliary support ring is monitored by a displacement sensor, and the temperature collection frequency of the infrared temperature sensor is not less than 5 times per second, and the pressure collection frequency of the pressure sensor is not less than 10 times per second.

[0016] In an optional scheme: in the mathematical model construction step, the sidewall turnup height H is used as the dependent variable, and the temperature T, pressure P, turnup speed v, tensile force F, etc. are used as independent variables to construct a multivariate linear regression model: H = α0+α1T+α2P+α3v+α4F+∈, wherein α0, α1, α2, α3, α4 are regression coefficients, ∈ is the error term, and the least squares method is used to fit the historical production data to solve the regression coefficient.

[0017] In an optional scheme: in the mathematical model construction step, a convolutional neural network (CNN) or a long short-term memory network (LSTM) is used to construct a machine learning model, the preprocessed data set is divided into a training set, a validation set and a test set, the model is trained using the training set, the model parameters are adjusted by the back propagation algorithm, the mean square error (MSE) or the mean absolute error (MAE) is used as the loss function, the model parameters are adjusted using the validation set, and the model generalization ability is evaluated using the test set; when a convolutional neural network is used, the network structure includes no less than 3 convolutional layers and 2 fully connected layers; when a long short-term memory network is used, the network includes no less than 2 LSTM layers.

[0018] In an optional solution: in the mathematical model construction step, based on the viscoelastic mechanical properties of the rubber material, combined with the heat conduction equation and the stress-strain relationship, the Maxwell model or the Kelvin-Voigt model is introduced to describe the viscoelastic behavior of the rubber, and the influence of temperature on the rubber properties is analyzed by the Fourier heat conduction law. A physical model is established, and the calculation results of the physical model are combined with the machine learning model to construct a hybrid model.

[0019] In an optional solution: in the process execution and control steps, the turning-up pressure roller is controlled by a programmable logic controller to have a downward pressure accuracy of ±0.5° and a pressure accuracy of ±1N; at the initial stage of turning-up, the turning-up pressure roller pre-laminates the sidewall with a small pressure, and then gradually increases the pressure; the sidewall turning-up process is divided into an initial turning-up stage, an intermediate turning-up stage and a final stage, a lower turning-up speed and a larger tensile force are used in the initial turning-up stage, the turning-up speed is increased and the turning-up pressure is adjusted in the intermediate turning-up stage, and the speed and pressure are reduced in the final stage for fine adjustment; the temperature of the sidewall rubber is monitored in real time during the turning-up process, and the turning-up pressure is dynamically adjusted according to the mathematical model and temperature changes.

[0020] In an optional solution: in the model verification and optimization step, when the model prediction error exceeds 5% of the actual reverse packaging height, the model retraining program is triggered; the intelligent monitoring system analyzes the reverse packaging height and edge flatness through an image recognition algorithm, and when the reverse packaging height deviation exceeds ±5% of the set threshold, an alarm is issued and the process parameters of the reverse packaging mechanism are automatically adjusted.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] This technology achieves precise control of the sidewall turnup height during the molding of all-steel radial tires through real-time multi-parameter acquisition, multi-element mathematical model construction and automated intelligent control. It can increase the product qualification rate to over 95%, reduce rolling resistance by 10%-15%, and the risk of tire blowout by 30%, and reduce material waste and equipment loss. At the same time, it promotes the intelligent upgrade of tire manufacturing and significantly improves production efficiency and overall product performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A logic block diagram of the processing of the invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0025] In an embodiment of the present invention, a method for increasing the sidewall turnup height during the molding process of an all-steel radial tire is as follows:

[0026] Data collection and preprocessing implementation method:

[0027] Data Collection Equipment and Operation: Data collection equipment was deployed on the all-steel radial tire sidewall turnup molding equipment. Temperature collection: A non-contact infrared temperature sensor (Raytek Raynger 3i Plus) was used. Five measurement points were evenly spaced across the sidewall turnup area, located in the center and at the upper and lower edges. The sidewall rubber surface temperature was measured eight times per second. A K-type thermocouple (accuracy ±0.5°C) was used to monitor the surface temperature of the turnup roller, providing real-time information on roller temperature changes. Pressure collection: A high-precision pressure sensor (range 0-500N, accuracy ±0.1N) was embedded on the turnup roller surface, collecting turnup pressure data 12 times per second. A micro-thin film pressure sensor (1mm thickness, measurement range 0-100N) was installed at the contact point between the auxiliary support ring and the sidewall to monitor support pressure changes. Other data collection: An incremental encoder (resolution 1000 lines) was used to collect the turnup roller's downward pressure angle and speed data. A magnetostrictive displacement sensor (accuracy ±0.01mm) was used to monitor the expansion and contraction displacement of the auxiliary support ring. At the same time, the equipment control system retrieves process parameters such as tire specifications (such as tire section width and aspect ratio), rubber formula number, and integrates them into a complete data set.

[0028] After the data collected by the data preprocessing process is transferred to the industrial computer, the following operations are performed: Filtering: Use the Kalman filter algorithm to write a program to filter the noisy data such as temperature and pressure, remove abnormal fluctuations, and retain the true signal characteristics. Time synchronization: Use the timestamp function of the data acquisition card to align the time of different sensor data to ensure the consistency of each data in the time dimension. Normalization: Use the Min-Max normalization method to uniformly map data such as temperature (range 0-200℃) and pressure (range 0-500N) to the [0,1] interval. The formula is:

[0029] Mathematical Model Construction and Implementation Methods: Data Preparation for the Multiple Linear Regression Model: Nearly 1,000 sets of historical production data were collected, including measured values ​​of sidewall turnup height, temperature, pressure, turnup speed, tensile force, and other parameters. Model Training: Using Python's Scikit-learn library, a multiple linear regression program was written, with sidewall turnup height H as the dependent variable and temperature T, pressure P, turnup speed v, and tensile force F as independent variables. The regression coefficients were solved using the least squares method, resulting in the model formula: H = 0.2T + 0.15P - 0.05v + 0.03F + 5. (II) Machine Learning Model: Convolutional Neural Network (CNN): Network Structure: A network structure was constructed consisting of four convolutional layers (with kernel sizes of 3×3, 3×3, 5×5, and 5×5, respectively), two pooling layers (max pooling with a pooling window of 2×2), and three fully connected layers. Training parameters: Adam optimizer with a learning rate of 0.001 and mean squared error (MSE) as the loss function. 80% of the data was used as the training set, 10% as the validation set, and 10% as the test set, for 100 epochs. Long Short-Term Memory (LSTM) network: Network structure: Two LSTM layers (128 neurons each) followed by two fully connected layers. Training parameters: RMSProp optimizer with a learning rate of 0.0005. Early stopping was used during training to prevent overfitting.

[0030] The hybrid model is based on the viscoelastic theory of rubber materials. The Maxwell model and Kelvin-Voigt model are written in MATLAB to simulate the stress-strain relationship of rubber under different temperatures and pressures. The calculation results of the physical model are weightedly fused with the predicted values ​​of the machine learning model. The weights are determined through cross-validation to construct a hybrid prediction model.

[0031] Process Execution and Control Methodology: The roll-up rollers are driven by a servo motor (1.5kW, 5N·m torque) coupled with a high-precision ball screw (5mm lead, C5 accuracy) to precisely control the angle and pressure of the roll-up rollers. A programmable logic controller (PLC, Siemens S7-1500) receives model predictions and controls the roll-up angle to an accuracy of ±0.3° and the pressure to ±0.5N. The auxiliary support rings utilize a pneumatic retractable structure, driven by a cylinder (50mm diameter, 100mm stroke). The support rings are coated with polytetrafluoroethylene to reduce friction and automatically extend 0.5 seconds before roll-up and retract 1 second after roll-up is complete.

[0032] Process parameter control of the turn-up stage: Initial turn-up stage: the turn-up speed is set to 2m / min, the tensile force is 60N, and the turn-up roller pre-laminates the sidewall with a pressure of 20N, which lasts for 10 seconds. Intermediate turn-up stage: the speed is increased to 4m / min, and the pressure is dynamically adjusted according to the model prediction, ranging from 30-50N. Final stage: the speed is reduced to 1.5m / min, the pressure is maintained at 35N, and fine adjustments are made, which lasts for 8 seconds. Temperature-pressure coordinated control: When the infrared temperature sensor detects that the sidewall rubber temperature is lower than 120°C, the PLC controls the turn-up roller pressure to increase by 5N; when the temperature is higher than 150°C, the pressure is reduced by 3N.

[0033] The intelligent monitoring system uses an industrial-grade high-definition camera (1920×1080 resolution, 30fps) and a laser ranging sensor (0-500mm measurement range, ±0.1mm accuracy) to capture real-time images and distance data of the tire sidewall's folded-over area. If the system determines, through an image recognition algorithm, that the folded-over height deviation exceeds a set threshold of ±3mm, it immediately triggers an alarm and automatically adjusts the folded-over roller pressure and speed parameters based on hybrid model analysis.

[0034] The model validation and optimization implementation method selects 20 sets of production data daily as validation samples, comparing the model-predicted turn-up height with the actual value measured by a three-dimensional coordinate measuring machine (accuracy ±0.05mm). If the average prediction error exceeds 5%, the model retraining process is automatically triggered, updating the historical database and readjusting the model parameters to ensure that the model continues to adapt to changes in the production process. Through the above-mentioned specific implementation method, the technical solution of increasing the turn-up height of the sidewall of an all-steel radial tire as claimed in the claims can be effectively implemented, significantly improving tire molding quality and production efficiency.

[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for increasing the sidewall turnup height during the molding process of an all-steel radial tire, characterized by: The following steps are involved: Step 1: Collect data such as temperature, pressure, roll angle and speed, and auxiliary support ring expansion and contraction displacement during the sidewall turnup process. Combined with process parameters such as tire specifications and rubber formulation, this data set is generated. Step 2: Filter, time synchronize and normalize the collected raw data; Step 3: Build a hybrid model that includes a multiple linear regression model, a machine learning model, and physical principles; Step 4: Compare the anti-packet height predicted by the model with the actual measured value for verification and optimization; Step 5: Based on the model prediction results, a high-precision turn-up roller driven by a servo motor is used to turn up the tire sidewall, and the auxiliary support ring and turn-up stage process parameters are controlled. At the same time, the intelligent monitoring system is used to adjust the turn-up process in real time.

2. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the data collection step, 3-5 measurement points are evenly arranged in the turn-up area using an infrared temperature sensor to collect temperature change data of the sidewall rubber surface during the initial, intermediate, and final turn-up stages. A thermocouple sensor is used to monitor the temperature of the turn-up roller. A pressure sensor on the turn-up roller surface collects turn-up pressure data, and a micro pressure sensor at the contact point between the auxiliary support ring and the sidewall collects support pressure data. The encoder is used to collect the downward pressure angle and speed data of the reverse pressure roller, and the displacement sensor is used to monitor the telescopic displacement of the auxiliary support ring. The temperature collection frequency of the infrared temperature sensor is not less than 5 times per second, and the pressure collection frequency of the pressure sensor is not less than 10 times per second.

3. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the mathematical model construction step, the sidewall turnup height H is used as the dependent variable, and the temperature T, pressure P, turnup speed v, tensile force F, etc. are used as independent variables to construct a multivariate linear regression model: H = α0 + α1T + α2P + α3v + α4F + ∈, wherein α0, α1, α2, α3, α4 are regression coefficients, ∈ is the error term, and the least squares method is used to fit the historical production data to solve the regression coefficient.

4. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the mathematical model construction step, a convolutional neural network or a long short-term memory network is used to construct a machine learning model, the preprocessed data set is divided into a training set, a validation set and a test set, the model is trained using the training set, the model parameters are adjusted by a back propagation algorithm, the mean square error or the mean absolute error is used as the loss function, the model parameters are adjusted using the validation set, and the model generalization ability is evaluated using the test set; when a convolutional neural network is used, the network structure includes no less than 3 convolutional layers and 2 fully connected layers; when a long short-term memory network is used, the network includes no less than 2 LSTM layers.

5. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the mathematical model construction step, based on the viscoelastic mechanical properties of the rubber material, combined with the heat conduction equation and the stress-strain relationship, the Maxwell model or the Kelvin-Voigt model is introduced to describe the viscoelastic behavior of the rubber. The influence of temperature on the rubber performance is analyzed by the Fourier heat conduction law. A physical model is established, and the calculation results of the physical model are combined with the machine learning model to construct a hybrid model.

6. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the process execution and control steps, the programmable logic controller is used to control the downward pressure angle accuracy of the turn-up pressure roller to ±0.5° and the pressure accuracy to ±1N; at the initial stage of turn-up, the turn-up pressure roller pre-laminates the sidewall with a small pressure, and then gradually increases the pressure; the sidewall turn-up process is divided into an initial turn-up stage, an intermediate turn-up stage and a final stage, a lower turn-up speed and a larger tensile force are used in the initial turn-up stage, the turn-up speed is increased and the turn-up pressure is adjusted in the intermediate turn-up stage, and the speed and pressure are reduced in the final stage for fine adjustment; the temperature of the sidewall rubber is monitored in real time during the turn-up process, and the turn-up pressure is dynamically adjusted according to the mathematical model and temperature changes.

7. The method for increasing the sidewall turnup height during the molding process of an all-steel radial tire according to claim 1, characterized in that: In the model verification and optimization step, when the model prediction error exceeds 5% of the actual reverse wrapping height, the model retraining program is triggered; the intelligent monitoring system analyzes the reverse wrapping height and edge flatness through an image recognition algorithm, and when the reverse wrapping height deviation exceeds ±5% of the set threshold, an alarm is issued and the process parameters of the reverse wrapping mechanism are automatically adjusted.

Images