Intelligent control method for interface temperature of tire multi-composite extrusion compound

Abstract

The application discloses a kind of intelligent control methods of tire multi-composite extrusion rubber interface temperature, it is related to tire manufacturing technical field, to solve the technical problem of current experience and fixed parameter control resulting in regulation and control lag, low precision and poor adaptability, S100, constructs and optimizes extrusion dynamic condition pre-judgment model;S200, issue multi-device advanced linkage control instruction;S300, issue instruction to realize the dynamic adjustment of multiple devices cooperation;S400, continuously collect control feedback data until the rubber interface temperature reaches the preset control requirement.The application is through normalization algorithm preprocessing, extrusion dynamic condition intelligent pre-judgment, and the integrated control structure of adaptive fuzzy PID core algorithm dynamic parameter setting, combined with multi-source data acquisition, condition change identification, and controller parameter pre-setting and autonomous setting collaborative operation, so that temperature control can follow extrusion process real-time change, accurately control rubber interface temperature in process preset interval.

Description

Technical Field

[0001] This invention relates to the field of tire manufacturing technology, and more specifically, to an intelligent control method for the interface temperature of multi-composite extruded rubber compounds for tires. Background Technology

[0002] Multi-component extrusion is one of the core processes in tire manufacturing. The precision of temperature control at the rubber compound interface directly determines the adhesion performance and structural stability between rubber layers, thus affecting the overall mechanical properties and safety of the finished tire. During multi-component extrusion, the temperature at the interface between different rubber compounds is a key control parameter in the compounding process. It directly affects the melting state of the rubber compound and the diffusion and bonding effect between molecules. Only by stably controlling this temperature within a reasonable range can a strong bond between different rubber layers be achieved, fundamentally preventing quality defects such as delamination and cracking in the compound.

[0003] Currently, the industry's control of the interface temperature of multi-compound extrusion rubber compounds for tires still mainly relies on operator experience and conventional single-loop control strategies, independently adjusting individual equipment parameters such as extruder barrel temperature, screw speed, or die temperature. In actual multi-compound extrusion processes, various factors such as rubber compound conveying speed, screw speed, and ambient workshop temperature cause dynamic changes in the rubber compound interface temperature. Fixed PID proportional, integral, and derivative parameters cannot autonomously correct and dynamically adjust according to real-time temperature changes; they can only perform mechanical control according to preset fixed rules. This not only makes it difficult to accurately and stably control the rubber compound interface temperature within the reasonable range required by the process, but also easily leads to over- or under-control, resulting in continuous and irregular fluctuations in the rubber compound interface temperature. Such temperature fluctuations directly disrupt the molecular diffusion bonding effect between different rubber compound layers, causing serious quality defects such as rubber compound delamination and poor adhesion. This not only restricts the improvement of product quality in multi-compound extrusion processes for tires but also fails to adapt to diverse production needs such as tire specification switching and rubber compound batch changes, making it difficult to meet the high-precision and high-stability temperature control requirements of modern tire production. In view of this, we propose an intelligent control method for the interface temperature of multi-composite extruded rubber compounds for tires. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art, adapt to practical needs, and provide an intelligent control method for the interface temperature of multi-composite extruded rubber compounds for tires, so as to solve the technical problems of lag in regulation, low precision and poor adaptability caused by the current reliance on experience and fixed parameter control.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for intelligent control of the interface temperature of multi-composite extruded rubber compounds for tires, comprising the following steps:

[0006] S100: Collect process, equipment and rubber interface temperature field data of tire multi-composite extrusion process, and construct and optimize the extrusion dynamic working condition prediction model after preprocessing by linear normalization algorithm.

[0007] S200 identifies trigger signals for changes in working conditions through a model, predicts the evolution trend of interface temperature, pre-tunes the parameters of the adaptive fuzzy PID controller, and issues multi-equipment advanced linkage control commands to the extruder temperature control system, screw speed adjustment mechanism, and die heating device.

[0008] S300: After determining the change in operating conditions, switch to closed-loop control mode. Obtain the total temperature deviation data through temperature deviation fusion algorithm, input the controller to complete the parameter self-tuning based on the adaptive fuzzy PID core algorithm, and then issue instructions to realize the collaborative dynamic adjustment of multiple devices.

[0009] S400 continuously collects control feedback data to determine whether the control effect meets the standard. If it does not meet the standard, it iteratively optimizes the loss function and parameter update formula, and combines the small batch gradient descent method to iteratively optimize the controller parameters and multi-device linkage logic until the rubber interface temperature reaches the preset control requirements.

[0010] Preferably, step S100 includes the following sub-steps:

[0011] S101. Non-contact acquisition of rubber interface temperature field data through infrared thermal imaging equipment, and simultaneous acquisition of process operation data such as temperature and flow rate of each rubber flow, as well as equipment operation status data such as screw speed and melt pressure.

[0012] S102. Based on historical process, equipment, operating condition changes and corresponding temperature evolution data, integrate rubber compounding process and equipment operation rules to train and optimize the extrusion dynamic operating condition prediction model.

[0013] S103. The collected real-time data is subjected to noise reduction, linear normalization and completion preprocessing operations in sequence. The linear normalization algorithm is used to complete the normalization processing of the data. The preprocessed real-time data serves as effective data support for the identification and prediction of the extrusion dynamic working condition prediction model.

[0014] Preferably, the normalization algorithm is a linear normalization algorithm, and its formula is:

[0015] ;

[0016] in, For the normalized processed data, The raw data collected, The minimum value of the original data in this dimension. This represents the maximum value of the original data for this dimension.

[0017] Preferably, step S200 includes the following sub-steps:

[0018] S201. Input the pre-processed real-time data into the prediction model to identify one or more working condition change trigger signals in tire specification switching, rubber batch change, production speed adjustment, and formula change. The signals are transmitted by the production line control system in the form of data messages.

[0019] S202. The model combines the logic of rubber compound extrusion process to deduce the direction and magnitude of interface temperature change after the working conditions change, and obtain the temperature evolution trend results.

[0020] S203. Transform the temperature evolution trend into a controller input variable, and pre-tune the proportional, integral, and derivative parameters of the adaptive fuzzy PID controller in combination with the interlayer bonding requirements of the adhesive.

[0021] S204. Based on the pre-set parameters, issue step-by-step advanced linkage control commands to each device, first adjusting the die heating device, then adjusting the extruder temperature control system, and finally fine-tuning the screw speed.

[0022] Preferably, step S300 includes the following sub-steps:

[0023] S301: Real-time acquisition of equipment operating status, process operation and rubber interface temperature field data; based on the conventional steady-state threshold of each data, comprehensively determine whether the change of operating condition is completed; after the determination is completed, trigger the closed-loop control mode switching command.

[0024] S302. Collect the actual temperature of each monitoring point on the rubber interface, calculate the numerical deviation between the actual average temperature of the interface and the preset temperature, and at the same time, combine the uniformity deviation with the temperature field distribution characteristics of the interface to obtain the uniformity deviation. Use the temperature deviation fusion algorithm to weight and fuse the two types of deviations into the total temperature deviation data.

[0025] S303: Based on the adaptive fuzzy PID core algorithm, the total temperature deviation data and the deviation change rate are used as dual input variables and input to the pre-tuned adaptive fuzzy PID controller, which then completes the autonomous tuning of the proportional, integral, and derivative parameters.

[0026] S304. Based on the parameters after the controller is autonomously calibrated, a linkage control command is sent to the extruder temperature control system, screw speed adjustment mechanism, and die heating device to realize the coordinated dynamic adjustment of each device.

[0027] Preferably, the temperature deviation fusion algorithm formula is:

[0028] ;

[0029] ;

[0030] in, For the total temperature deviation, The average temperature deviation weighting coefficient is used. This is the weighting coefficient for temperature field uniformity deviation. This represents the actual average temperature of the interface. Set the interface temperature. The standard deviation of the interface temperature field. This refers to the number of monitoring points for the infrared thermal imaging equipment. For the first The actual temperature at each monitoring point.

[0031] Preferably, the adaptive fuzzy PID core algorithm formula includes the PID control output formula and the parameter update formula after fuzzy correction;

[0032] The PID control output formula is as follows:

[0033] ;

[0034] The parameter update formula after fuzz correction is as follows:

[0035] ;

[0036] ;

[0037] ;

[0038] in, For the controller in Real-time control output at all times. The scaling factor is the fuzzy correction factor. for Total temperature deviation at any given time These are the integral coefficients after fuzzy correction. For integration variables, These are the differential coefficients after fuzzy correction. The initial value of the proportional coefficient pre-tuned in S203, These are the initial values ​​of the integral coefficients pre-tuned in S203. These are the initial values ​​of the differential coefficients pre-tuned in S203. This is the correction amount for the proportional coefficient obtained from fuzzy reasoning. This is the correction amount for the integral coefficients obtained from fuzzy inference. This is the correction amount for the differential coefficients obtained from fuzzy reasoning. for The rate of change of the total temperature deviation at any given time.

[0039] Preferably, step S400 includes the following sub-steps:

[0040] S401. According to the preset acquisition frequency, synchronously acquire the operating parameters of each device after adjustment, the temperature field distribution data of the rubber interface and the overall temperature data to form control feedback data.

[0041] S402. Compare the control feedback data with the preset control requirements for the interface temperature of the rubber compound to determine whether the control effect has met expectations. The preset control requirements include that the interface temperature value is within a reasonable range, the temperature field uniformity meets the standard, and the temperature stability meets the interlayer bonding requirements of the rubber compound.

[0042] S403. If the control effect does not meet expectations, the control feedback data will be re-input into the adaptive fuzzy PID controller, the controller parameters will be autonomously tuned again, and the multi-device linkage control logic will be adjusted before the control command is reissued.

[0043] S404. Using the feedback data from each adjustment as training samples, with the optimization objectives of temperature control accuracy, temperature field uniformity, and equipment operation coordination, an iterative optimization loss function and parameter update algorithm based on the mini-batch gradient descent method is adopted to continuously optimize the controller's tuning parameters and multi-device linkage control logic until the control effect reaches the expected level.

[0044] Preferably, the iterative optimization loss function is:

[0045] ;

[0046] The parameter update algorithm is as follows:

[0047] ;

[0048] in, This is the total loss value. This is a weighting coefficient for temperature control accuracy. As a weighting coefficient for the rationality of equipment regulation, This is the weighting coefficient for equipment operation coordination. To average the output of multiple devices, This serves as a reference value for the equipment's output regulation. To control the standard deviation of output from multiple devices, For the first The parameters to be optimized for the controller's fuzzy inference rules after the next iteration. For the first The parameters to be optimized for the controller's fuzzy inference rules in the next iteration. The learning rate has a range of values. Control parameter update step size This refers to the number of training samples in a small batch. For the loss function with respect to parameters gradient, For the first Input features of each training sample, For the first The expected output of each training sample.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. This invention, through the design of an integrated control structure that combines linear normalization algorithm preprocessing, intelligent prediction of extrusion dynamic conditions, and adaptive fuzzy PID core algorithm dynamic parameter tuning, combined with multi-source data acquisition, condition change identification, and collaborative operation of controller parameter pre-tuning and autonomous tuning, achieves advanced response, dynamic adaptation, and precise control of the interface temperature of multi-composite rubber compounds in tire extrusion. This allows temperature control to closely follow the real-time changes in the extrusion process, accurately controlling the rubber compound interface temperature within the preset process range, thus solving the problems of lag, low accuracy, and poor adaptability caused by current reliance on experience and fixed parameter control.

[0051] 2. This invention also achieves full-domain uniform control of the rubber compound interface temperature and precise coordinated operation of various production equipment by designing an infrared thermal imaging full-domain temperature detection, temperature deviation fusion algorithm quantification, and adaptive fuzzy PID controller multi-device linkage control refined management structure. Relying on full-domain temperature field data acquisition, weighted fusion of average temperature deviation and uniformity deviation, and coordinated adjustment command issuance of extruder temperature control system, screw speed adjustment mechanism, and die heating device, this invention achieves full-domain uniform control of rubber compound interface temperature and precise coordinated operation of various production equipment, making the full-domain temperature distribution of rubber compound interface more uniform and the equipment operation more in line with process requirements.

[0052] 3. This invention also designs a full-process intelligent control system with closed-loop regulation feedback and iterative optimization of loss function and parameter update algorithm for self-optimization. Combined with the synchronous acquisition of multi-dimensional feedback data after regulation, the construction of a multi-objective loss function with the goals of temperature control accuracy and equipment synergy, and the continuous optimization of controller parameters and equipment linkage logic using small-batch gradient descent method, this invention achieves self-optimization, self-adaptation and long-term stability of the rubber interface temperature control strategy. This allows the temperature control system to continuously iterate and upgrade with process changes, ensuring the stability of the production process. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating the overall method of the present invention;

[0054] Figure 2 This is a flowchart illustrating the construction of the extrusion dynamic working condition prediction model and the advanced control of working conditions according to the present invention.

[0055] Figure 3 This is a flowchart of the closed-loop control and multi-device coordinated adjustment of the rubber interface temperature of the present invention;

[0056] Figure 4 This is a flowchart illustrating the iterative optimization process for controlling the interface temperature of the adhesive material according to the present invention. Detailed Implementation

[0057] Example: Figures 1 to 4 As shown, the present invention relates to an intelligent control method for the interface temperature of multi-composite extruded rubber compounds for tires, comprising the following steps:

[0058] S100: Collect process, equipment and rubber interface temperature field data during the multi-composite extrusion process of tires, and construct a dynamic extrusion condition prediction model;

[0059] S100 specifically includes the following sub-steps:

[0060] S101. Real-time acquisition of process operation data, equipment operation status data, and temperature field data of the rubber compound interface area. The temperature field data of the rubber compound interface area is acquired by non-contact monitoring using infrared thermal imaging equipment. The infrared thermal imaging equipment is deployed in the interface monitoring area of ​​the die outlet of the rubber compound composite extrusion, and the monitoring angle covers all interlayer composite interfaces of the rubber compound, realizing full-area monitoring of the interface temperature field without blind spots. The process operation data includes the real-time temperature and real-time flow rate of each rubber compound flow, and the equipment operation status data includes the screw speed of each extruder, the structural parameters of the die, the real-time operating power of each heating device, and the melt pressure data of the extruder.

[0061] S102. Based on the historical process, equipment, operating condition changes, and corresponding interface temperature evolution data of the extrusion process, and integrating the characteristics of rubber compounding process and equipment operation rules, the training and optimization of the extrusion dynamic operating condition prediction model are completed. During the model training process, the type of operating condition change is used as the input feature and the corresponding interface temperature evolution trend is used as the output label. The model is constrained by the material flow and heat conduction laws of tire multi-composite extrusion, which improves the model's ability to identify the correlation between operating condition changes and temperature trends, and enables the model to accurately identify the operating condition change trend and predict the interface temperature evolution.

[0062] S103. The collected real-time data is preprocessed to form effective data support for model recognition and prediction. The preprocessing operations include data denoising, data normalization, and data completion. The moving average method is used to remove random noise in the data, the linear normalization method is used to map data of different dimensions and scales to the same numerical range, and the neighborhood interpolation method is used to complete the missing data that appears during the collection process. Through preprocessing, invalid and abnormal data are completely removed to ensure that the data input to the model can accurately reflect the actual state of the extrusion process.

[0063] The formula for the linear normalization algorithm is as follows:

[0064] ;

[0065] in, For the normalized processed data, The raw data collected (such as temperature, rotation speed, pressure, etc.). The minimum value of the original data in this dimension. The formula represents the maximum value of the original data in this dimension. It achieves the normalization mapping of data with different dimensions (such as temperature, speed, and pressure), eliminates the impact of differences in data dimensions on the accuracy of model prediction, and improves the feature recognition capability of the working condition prediction model.

[0066] S200 identifies the trigger signal of the change in working conditions through the model, predicts the evolution trend of interface temperature, pre-tunes the parameters of the adaptive fuzzy PID controller, and issues multi-device advanced linkage control instructions.

[0067] S200 specifically includes the following sub-steps:

[0068] S201. Input the preprocessed real-time data into the extrusion dynamic working condition prediction model. The model identifies the trigger signals for changes in working conditions in real time. The trigger signals for changes in working conditions in S201 include one or more of the following: production line instruction signals for tire specification switching, material supply signals for rubber batch change, extrusion production speed adjustment instruction signals, and process instruction signals for rubber formula change. The signals are transmitted in real time to the extrusion dynamic working condition prediction model in the form of data messages by the control system of the tire multi-composite extrusion intelligent manufacturing production line. The model performs feature extraction and feature matching on the received signals to complete the accurate identification of the trigger signals for changes in working conditions and avoid ineffective control caused by signal misjudgment.

[0069] S202. Based on the identified trigger signals and combined with the compound extrusion process logic, the model predicts the evolution trend of the rubber interface temperature after the working conditions change. Based on the trained correlation between working conditions and temperature, combined with the real-time process and equipment data of the current extrusion process, the model infers the direction and magnitude of the change in the rubber interface temperature after the working conditions change, forming a visualized temperature evolution trend result, which provides a basis for subsequent parameter pre-tuning.

[0070] S203. Based on the predicted temperature evolution trend, the parameters of the adaptive fuzzy PID controller are pre-tuned to enable the controller to adapt to the temperature control requirements after the change of working conditions. Specifically, the parameter pre-tuning of the adaptive fuzzy PID controller in S203 involves converting the predicted evolution trend of the rubber interface temperature after the change of working conditions into the input variable of the controller. Combined with the core requirements of temperature control for interlayer bonding of rubber, the proportional, integral, and derivative parameters of the adaptive fuzzy PID controller are pre-tuned to ensure that the control logic of the controller matches the process requirements after the change of working conditions, so that the controller has the ability to control temperature fluctuations after the change of working conditions.

[0071] S204. Based on the pre-tuned controller parameters, issue multi-equipment advance linkage control commands to the extruder temperature control system, screw speed adjustment mechanism, and die heating device to complete the advance adjustment of the equipment. The advance control follows the material flow and heat conduction process logic of compound extrusion and adopts a step-by-step adjustment method. First, the die heating device is adjusted for preheating or temperature fine-tuning. Then, the heating / cooling power of the extruder temperature control system is adjusted. Finally, the screw speed is fine-tuned according to the temperature adaptation requirements to avoid the disorder of the rubber flow state caused by the simultaneous adjustment of the equipment. This allows the operating status of each piece of equipment to adapt to the interface temperature control requirements after the change of working conditions in advance.

[0072] S300: After determining the change in operating conditions, switch to closed-loop control mode, obtain the interface temperature deviation and input it into the controller to complete the parameter self-tuning, and issue commands to realize the collaborative dynamic adjustment of multiple devices.

[0073] S300 specifically includes the following sub-steps:

[0074] S301: Real-time acquisition of equipment operating status, process operation, and interface temperature field data; comprehensive determination of whether the change in operating conditions is complete; when the fluctuation range of each data is within the preset stable range, the change in operating conditions is determined to be complete; the specific determination logic is based on the conventional steady-state threshold of each equipment and process data; when the fluctuation range of all data within a continuous preset time does not exceed the corresponding steady-state threshold, and there is no obvious sudden change in the temperature field of the rubber compound interface, the determination signal of the completion of the change in operating conditions is triggered, ensuring that the extrusion process after the change in operating conditions enters a stable state before switching the control mode.

[0075] S302, trigger the closed-loop control mode switching command, compare the actual temperature of each monitoring point on the rubber interface collected in real time with the preset temperature, and obtain the temperature deviation data through the temperature deviation fusion algorithm;

[0076] When acquiring temperature deviation data in S302, the overall average temperature of the rubber interface is first calculated based on the temperature of all points monitored by infrared thermal imaging, and the numerical deviation between the average temperature and the preset temperature is obtained. Then, the uniformity deviation represented by the standard deviation of the interface temperature field is introduced through the temperature deviation fusion algorithm. The two types of deviations are weighted and fused into the total temperature deviation, so that the final temperature deviation data not only reflects the overall temperature deviation, but also reflects the uniformity of the temperature distribution across the entire interface, comprehensively characterizing the actual control state of the rubber interface temperature, and avoiding the problem of local temperature runaway caused by judging only the average temperature.

[0077] The formula for the temperature deviation fusion algorithm is as follows:

[0078] ;

[0079] ;

[0080] in, The total temperature deviation comprehensively characterizes the overall deviation state of the interface temperature. This is the weighting coefficient for the average temperature deviation, with a value range of (0,1). The weighting coefficient for temperature field uniformity deviation satisfies It is set according to the requirements of the interlayer bonding process of the adhesive. is the actual average temperature of the interface, and is the arithmetic mean of the temperatures at all monitoring points in the infrared thermal imaging. The preset temperature for the interface is determined based on the type of adhesive and the lamination process. The standard deviation of the interface temperature field characterizes the uniformity of the temperature field; the smaller the value, the better the uniformity. The number of monitoring points for the infrared thermal imaging equipment is a positive integer. For the first The actual temperature at each monitoring point ;

[0081] This formula weights and fuses the relative deviation of average temperature with the relative deviation of temperature field uniformity, realizing the quantitative characterization of temperature deviation. It solves the problem that a single temperature value cannot reflect the temperature state of the entire interface, and provides a precise input basis for PID parameter tuning.

[0082] S303. Input the temperature deviation data into the pre-tuned adaptive fuzzy PID controller, and the controller will autonomously tune the parameters based on the core adaptive fuzzy PID algorithm.

[0083] In the S303, the controller completes the autonomous parameter tuning. Specifically, the temperature deviation and the rate of change of the deviation are used as the dual input variables of the controller. The PID control output formula is used as the basis for the regulation output calculation, and the parameter update formula after fuzzy correction is used as the core basis for parameter tuning. Through the fuzzy inference rules built into the controller, the proportional, integral, and derivative parameters are corrected and tuned in real time. The fuzzy inference rules are set in combination with the temperature control requirements of tire rubber compound. The parameters after fuzzy correction can be directly substituted into the PID control output formula to complete the accurate calculation of the regulation output. This allows the controller's regulation parameters to dynamically adapt to the real-time change of the rubber interface temperature, improving the accuracy and timeliness of regulation.

[0084] The core algorithm formula for adaptive fuzzy PID is as follows:

[0085] PID control output formula:

[0086] ;

[0087] The parameter update formula after fuzzy correction is as follows:

[0088] ;

[0089] ;

[0090] ;

[0091] in, For the controller in The real-time control output is the core calculation basis for multi-device control commands. The scaling factor is the fuzzy correction factor. for Total temperature deviation at any given time These are the integral coefficients after fuzzy correction. For integration variables, These are the differential coefficients after fuzzy correction. The initial value of the proportional coefficient pre-tuned in S203, These are the initial values ​​of the integral coefficients pre-tuned in S203. These are the initial values ​​of the differential coefficients pre-tuned in S203. This is the correction amount for the proportional coefficient obtained from fuzzy reasoning. This is the correction amount for the integral coefficients obtained from fuzzy inference. This is the correction amount for the differential coefficients obtained from fuzzy reasoning. for The rate of change of the total temperature deviation at any given time;

[0092] By using fuzzy inference to correct the pre-tuned PID parameters in real time, the controller can dynamically adapt to the nonlinear and time-varying characteristics of the rubber interface temperature, thus solving the problems of low accuracy and poor adaptability of traditional fixed-parameter PID control.

[0093] S304. Based on the parameters after the controller is autonomously calibrated, a linkage control command is sent to each device to realize the coordinated dynamic adjustment of the extruder temperature control system, screw speed adjustment mechanism and die heating device;

[0094] The multi-device collaborative dynamic adjustment in S304 is that the adjustment range and adjustment rate of each device are uniformly allocated by the adaptive fuzzy PID controller according to the temperature deviation data. The adjustment parameters of each device are linked and matched. For example, the temperature adjustment range of the extruder temperature control system is negatively correlated with the screw speed adjustment range, avoiding secondary temperature fluctuations caused by shear heat changes in the rubber material, and realizing precise linkage control of each device.

[0095] S400 continuously collects feedback data after regulation and iteratively optimizes the controller tuning parameters and multi-device linkage regulation logic until the rubber interface temperature reaches the preset control requirements, thereby achieving intelligent and precise control of the interface temperature.

[0096] S400 specifically includes the following sub-steps:

[0097] S401. According to the preset acquisition frequency, continuously acquire the operating parameters of each device after adjustment, the temperature field distribution data of the rubber interface, and the overall temperature data to form control feedback data. The data acquisition follows the principle of synchronization to ensure that the acquisition timestamps of the operating parameters of each device and the interface temperature field data are consistent, so as to ensure the correlation and effectiveness of the feedback data and provide accurate data support for the subsequent judgment of the control effect.

[0098] S402. Compare the feedback data with the preset control requirements for the temperature of the rubber interface to determine whether the control effect has met expectations. The preset control requirements include that the temperature value of the rubber interface is within a preset reasonable range, the overall uniformity of the temperature field of the rubber interface is within a preset standard, and the temperature stability required for interlayer bonding of the rubber is within a preset range. Only when all the feedback data meet the above preset control requirements is it determined that the control effect has met expectations.

[0099] S403. If the control effect does not meet expectations, the feedback data will be input into the adaptive fuzzy PID controller to re-tune the parameters and adjust the multi-device linkage control logic, and then issue the control command again. Different parameter tuning priorities will be set for different combinations of absolute temperature deviation and uniformity deviation. The core temperature deviation that affects the interlayer adhesion of the rubber compound will be corrected first, and then the secondary deviations will be optimized step by step to improve the control efficiency.

[0100] S404. Using the feedback data from each adjustment as training samples, and with temperature control accuracy, temperature field uniformity, and equipment operation coordination as optimization objectives, a multi-objective loss function is constructed. The iterative optimization loss function and parameter update formula are used as the core quantitative calculation basis to continuously iteratively optimize the controller's tuning parameters and the multi-device linkage control logic until the feedback data shows that the control effect meets the preset control requirements. During the iterative optimization process, the deviation between the control effect and the preset target is first quantitatively characterized by the multi-objective loss function. Then, the small-batch gradient descent method is used, combined with the parameter update rules in the iterative optimization loss function and parameter update formula, to iteratively update the fuzzy inference rules of the controller. At the same time, the linkage adjustment ratio and response rate of the multi-device are precisely fine-tuned based on the calculation results of the formula, so that the controller and the multi-device linkage logic can continuously adapt to the dynamic changes in the extrusion process and achieve long-term stable and precise control of the rubber interface temperature.

[0101] The iterative optimization loss function and parameter update formula are as follows:

[0102] Multi-objective loss function:

[0103] ;

[0104] Mini-batch gradient descent parameter update:

[0105] ;

[0106] in, The total loss value represents the deviation between the control effect and the preset target; the smaller the value, the better the control effect. This is the weighting coefficient for temperature control accuracy, with a value range of (0,1). This is the weighting coefficient for the rationality of equipment regulation, with a value range of (0,1). The weighting coefficient for equipment operation coordination should satisfy... , The average controlled output of multiple devices is denoted as , and the arithmetic mean of the controlled output of each device is denoted as . This is a reference value for the equipment's output regulation, determined based on the equipment's rated operating parameters. The standard deviation of the output of multiple devices is used to characterize the coordination of device operation; the smaller the value, the better the coordination. For the first The parameters to be optimized for the controller's fuzzy inference rules after the next iteration. For the first The parameters to be optimized for the controller's fuzzy inference rules in the next iteration. The learning rate has a range of values. Control parameter update step size The number of training samples in a mini-batch is a positive integer. For the loss function with respect to parameters The gradient represents the direction of parameter update. For the first Input features of each training sample, For the first The expected output of each training sample;

[0107] This multi-objective loss function incorporates temperature control accuracy, equipment regulation rationality, and equipment operation coordination into a unified optimization objective. It uses a small-batch gradient descent method to iteratively update the controller parameters, enabling the control strategy to continuously adapt to the dynamic changes in the extrusion process and achieve long-term stable and precise control of the rubber compound interface temperature.

[0108] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A method for intelligent control of the interface temperature of multi-composite extruded rubber compounds for tires, characterized in that, Includes the following steps: S100: Collect process, equipment and rubber interface temperature field data of tire multi-composite extrusion process, and construct and optimize the extrusion dynamic working condition prediction model after preprocessing by linear normalization algorithm. S200 identifies trigger signals for changes in working conditions through a model, predicts the evolution trend of interface temperature, pre-tunes the parameters of the adaptive fuzzy PID controller, and issues multi-device advanced linkage control commands to the extruder temperature control system, screw speed adjustment mechanism, and die heating device. S300: After determining the change in operating conditions, switch to closed-loop control mode. Obtain the total temperature deviation data through temperature deviation fusion algorithm, input the controller to complete the parameter self-tuning based on the adaptive fuzzy PID core algorithm, and then issue instructions to realize the collaborative dynamic adjustment of multiple devices. S400 continuously collects control feedback data to determine whether the control effect meets the standard. If it does not meet the standard, it iteratively optimizes the loss function and parameter update formula, and combines the small batch gradient descent method to iteratively optimize the controller parameters and multi-device linkage logic until the rubber interface temperature reaches the preset control requirements.

2. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 1, characterized in that, S100 includes the following sub-steps: S101. Non-contact acquisition of rubber interface temperature field data through infrared thermal imaging equipment, and simultaneous acquisition of process operation data such as temperature and flow rate of each rubber flow, as well as equipment operation status data such as screw speed and melt pressure. S102. Based on historical process, equipment, operating condition changes and corresponding temperature evolution data, integrate rubber compounding process and equipment operation rules to train and optimize the extrusion dynamic operating condition prediction model. S103. The collected real-time data is subjected to noise reduction, linear normalization, and completion preprocessing operations in sequence. The linear normalization algorithm is used to complete the normalization processing of the data. The preprocessed real-time data serves as effective data support for the identification and prediction of the extrusion dynamic working condition prediction model.

3. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 2, characterized in that: The normalization algorithm is a linear normalization algorithm, and its formula is: ； in, For the normalized processed data, The raw data collected, The minimum value of the original data in this dimension. This represents the maximum value of the original data for this dimension.

4. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 1, characterized in that, S200 includes the following sub-steps: S201. Input the pre-processed real-time data into the prediction model to identify one or more working condition change trigger signals in tire specification switching, rubber batch change, production speed adjustment, and formula change. The signals are transmitted by the production line control system in the form of data messages. S202. The model combines the logic of rubber compound extrusion process to deduce the direction and magnitude of interface temperature change after the working conditions change, and obtain the temperature evolution trend results. S203. Transform the temperature evolution trend into a controller input variable, and pre-tune the proportional, integral, and derivative parameters of the adaptive fuzzy PID controller in combination with the interlayer bonding requirements of the adhesive. S204. Based on the pre-set parameters, issue step-by-step advanced linkage control commands to each device, first adjusting the die heating device, then adjusting the extruder temperature control system, and finally fine-tuning the screw speed.

5. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 4, characterized in that, S300 includes the following sub-steps: S301: Real-time acquisition of equipment operating status, process operation and rubber interface temperature field data; based on the conventional steady-state threshold of each data, comprehensively determine whether the change of operating condition is completed; after the determination is completed, trigger the closed-loop control mode switching command. S302. Collect the actual temperature of each monitoring point on the rubber interface, calculate the numerical deviation between the actual average temperature of the interface and the preset temperature, and combine the uniformity deviation with the temperature field distribution characteristics of the interface. Use the temperature deviation fusion algorithm to weight and fuse the two types of deviations into the total temperature deviation data. S303: Based on the adaptive fuzzy PID core algorithm, the total temperature deviation data and the deviation change rate are used as dual input variables and input to the pre-tuned adaptive fuzzy PID controller, which then completes the autonomous tuning of the proportional, integral, and derivative parameters. S304. Based on the parameters after the controller is autonomously calibrated, a linkage control command is sent to the extruder temperature control system, screw speed adjustment mechanism, and die heating device to realize the coordinated dynamic adjustment of each device.

6. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 5, characterized in that, The formula for the temperature deviation fusion algorithm is as follows: ； ； in, For the total temperature deviation, The average temperature deviation weighting coefficient is used. This is the weighting coefficient for temperature field uniformity deviation. This represents the actual average temperature of the interface. Preset the temperature for the interface. The standard deviation of the interface temperature field. This refers to the number of monitoring points for the infrared thermal imaging equipment. For the first The actual temperature at each monitoring point.

7. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 5, characterized in that, The core algorithm formula for the adaptive fuzzy PID includes the PID control output formula and the parameter update formula after fuzzy correction. The PID control output formula is as follows: ； The parameter update formula after fuzz correction is as follows: ； ； ； in, For the controller in Real-time control output at all times. The scaling factor is the fuzzy correction factor. for Total temperature deviation at any given time These are the integral coefficients after fuzzy correction. For integration variables, These are the differential coefficients after fuzzy correction. The initial value of the proportional coefficient pre-tuned in S203, These are the initial values ​​of the integral coefficients pre-tuned in S203. These are the initial values ​​of the differential coefficients pre-tuned in S203. This is the correction amount for the proportional coefficient obtained from fuzzy reasoning. This is the correction amount for the integral coefficients obtained from fuzzy inference. This is the correction amount for the differential coefficients obtained from fuzzy reasoning. for The rate of change of the total temperature deviation at any given time.

8. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 1, characterized in that, S400 includes the following sub-steps: S401. According to the preset acquisition frequency, synchronously acquire the operating parameters of each device after adjustment, the temperature field distribution data of the rubber interface and the overall temperature data to form control feedback data. S402. Compare the control feedback data with the preset control requirements for the interface temperature of the rubber compound to determine whether the control effect has met expectations. The preset control requirements include that the interface temperature value is within a reasonable range, the temperature field uniformity meets the standard, and the temperature stability meets the interlayer bonding requirements of the rubber compound. S403. If the control effect does not meet expectations, the control feedback data will be re-input into the adaptive fuzzy PID controller, the controller parameters will be autonomously tuned again, and the multi-device linkage control logic will be adjusted before the control command is reissued. S404. Using the feedback data from each adjustment as training samples, with the optimization objectives of temperature control accuracy, temperature field uniformity, and equipment operation coordination, an iterative optimization loss function and parameter update algorithm based on the mini-batch gradient descent method is adopted to continuously optimize the controller's tuning parameters and multi-device linkage control logic until the control effect reaches the expected level.

9. The intelligent control method for the interface temperature of multi-composite extruded rubber compound for tires according to claim 8, characterized in that, The iterative optimization loss function is: ； The parameter update algorithm is as follows: ； in, This is the total loss value. This is a weighting coefficient for temperature control accuracy. As a weighting coefficient for the rationality of equipment regulation, This is the weighting coefficient for equipment operation coordination. To average the output of multiple devices, This serves as a reference value for the equipment's output regulation. To control the standard deviation of output from multiple devices, For the first The parameters to be optimized for the controller's fuzzy inference rules after the next iteration. For the first The parameters to be optimized for the controller's fuzzy inference rules in the next iteration. The learning rate has a range of values. Control parameter update step size This refers to the number of training samples in a small batch. For the loss function with respect to parameters gradient, For the first Input features of each training sample, For the first The expected output of each training sample.

Images