A Cable Extrusion Temperature Control Optimization Method Based on Adaptive Control Algorithm

Abstract

This invention discloses a method for optimizing temperature control in cable extrusion based on an adaptive control algorithm, comprising the following steps: S1, collecting and preprocessing multi-source process data to generate a multivariable temperature control input dataset; S2, configuring a self-tuning PID controller and setting initial PID parameters; S3, co-evolutionarily optimizing the PID parameters of each heating zone to obtain the optimal multi-zone PID parameter combination; S4, allocating the optimal PID parameters and adaptively adjusting the parameters in conjunction with the input dataset; S5, implementing closed-loop temperature regulation, monitoring deviations, managing co-zones, and collecting temperature control, energy consumption, and process data; S6, periodically feeding back temperature control and energy consumption data, optimizing parameters and dynamically adjusting them, and resetting parameters in case of anomalies. This invention achieves intelligent adaptive optimization control of multi-zone temperature in the cable extrusion process, significantly improving temperature control accuracy, energy efficiency, and product quality stability.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, and in particular to a method for optimizing cable extrusion temperature control based on an adaptive control algorithm. Background Technology

[0002] In modern cable manufacturing, temperature control during the cable extrusion process plays a crucial role in ensuring product quality, improving production efficiency, and reducing energy consumption. With the diversification of cable product specifications and the continuous improvement of process requirements, the accuracy and response speed of temperature regulation in the extrusion process have become important indicators of a company's competitiveness. Currently, cable extrusion temperature control systems mostly rely on conventional PID controllers or fixed-parameter temperature control systems, achieving temperature regulation of each heating zone through manual experience tuning or simple closed-loop feedback adjustment. While traditional PID controllers have advantages such as simple structure and ease of implementation, their fixed control parameters make them difficult to adapt to complex operating conditions such as material performance fluctuations, changes in ambient temperature and humidity, process speed adjustments, and equipment aging during extrusion. This fixed-parameter control strategy easily leads to delayed temperature control response, increased temperature fluctuations, and system instability when facing process disturbances and multi-segment coupling effects, ultimately affecting the appearance consistency and electrical performance of cable products.

[0003] For multi-segment extrusion equipment, existing technologies mostly employ distributed independent PID control, lacking an effective collaborative optimization mechanism between segment controllers and failing to fully consider inter-segment thermal coupling and overall energy consumption optimization. Furthermore, traditional temperature control systems primarily rely on single data feedback from temperature sensors, lacking comprehensive integration of multi-variable process information such as ambient humidity, extrusion speed, and energy consumption. This results in control strategies that cannot accurately reflect the dynamic changes of complex processes. For parameter disturbances and system anomalies, existing technologies mostly rely on manual intervention or shutdown for correction, exhibiting low levels of intelligent adjustment and insufficient system adaptability and fault tolerance.

[0004] Existing cable extrusion temperature control methods have significant shortcomings in multi-segment collaborative optimization, adaptive adjustment, multi-variable process fusion, and automatic fault diagnosis, making it difficult to meet the requirements of high-end cable products for process stability, energy saving and consumption reduction, and quality consistency.

[0005] Therefore, how to provide a cable extrusion temperature control optimization method based on adaptive control algorithm is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] One objective of this invention is to propose a cable extrusion temperature control optimization method based on an adaptive control algorithm. This invention fully utilizes techniques such as self-tuning PID control, adaptive weight adjustment, co-evolutionary optimization, and fusion of prior process knowledge. It details the entire process of multi-source process data acquisition, multi-segment co-optimization, and intelligent adaptive dynamic adjustment of temperature control parameters. Through deep fusion and dynamic feedback of multi-variable temperature control input data, this invention achieves high-precision closed-loop control, automatic co-adjustment, and rapid anomaly response for each heating segment, possessing advantages such as high adjustment accuracy, fast response speed, low energy consumption, and strong product quality consistency.

[0007] According to an embodiment of the present invention, a method for optimizing cable extrusion temperature control based on an adaptive control algorithm includes the following steps:

[0008] S1. Collect multi-source process data during the cable extrusion process, preprocess the multi-source process data, and form a multivariable temperature control input dataset;

[0009] S2. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone and set the initial values ​​of the PID parameters for each self-tuning PID controller.

[0010] S3. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. The individual codes of each population are the PID control parameter combinations for the corresponding section. Co-evolutionary operations are performed on each population to obtain the optimal multi-segment PID parameter combinations.

[0011] S4. Based on the optimal multi-segment PID parameter combination, the parameters are allocated to their respective tuned PID controllers. Each tuned PID controller, in conjunction with the multivariable temperature control input dataset, performs online adaptive adjustment of the PID parameters.

[0012] S5. Using the adjusted PID parameters and the self-tuning PID controllers of each section, closed-loop temperature regulation is implemented for each heating section, the deviation between the temperature of each section and the target value is monitored in real time, and the coupling effect between sections is managed in a coordinated manner. At the same time, temperature control results, energy consumption and process indicators are collected.

[0013] S6. The collected temperature control results, energy consumption and process indicators are periodically fed back to the co-evolutionary algorithm for parameter optimization and dynamic adjustment. If any temperature control anomaly is detected in any segment, the global parameter reset mechanism is automatically triggered to reinitialize the relevant population and re-execute optimization and adaptive adjustment.

[0014] Optionally, the multi-source process data specifically includes the real-time temperature of each heating zone, extrusion speed, ambient temperature and humidity, and process parameters related to the extrusion process.

[0015] Optionally, the preprocessing of multi-source process data specifically includes denoising, normalizing, and standardizing the collected multi-source process data to generate a multivariable temperature control input dataset in a unified format.

[0016] Optionally, S2 specifically includes:

[0017] S21. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone, analyze the process status and historical operating data of each heating zone, and set the initial values ​​of the PID parameters of each tuning PID controller in combination with prior process knowledge.

[0018] S22. Combining the process prior knowledge base and historical operating data, generate recommended PID parameter ranges for each heating section and determine the proportional coefficient K. p (0), Integral coefficient K i (0) and differential coefficient K d The initial value of (0);

[0019] S23. The dynamic characteristic parameters of the heating section are identified online by using the recursive least squares method and introducing a forgetting factor.

[0020] S24. In the parameter identification process, abnormal data suppression method and adaptive gain adjustment method are adopted to improve the robustness and accuracy of parameter identification in response to abnormal disturbances and data noise.

[0021] S25. Use the identified dynamic characteristic parameters of the heating section to correct the proportional coefficient K in real time. p (t), integral coefficient K i (t), differential coefficient K d (t);

[0022] S26. Introduce dynamic weighting coefficients w for the proportional coefficient, integral coefficient, and derivative coefficient, respectively. p (t), w i (t), w d (t), the weights of each parameter are dynamically adjusted according to temperature deviation, adjustment speed and process characteristics;

[0023] S27. Calculate the dynamically weighted PID control parameters, and set the output signal of the self-tuning PID controller to u(t);

[0024] S28. Combine the process prior knowledge base to apply range constraints to the real-time updated PID parameters. When the parameters exceed the recommended range or the adjustment effect is not good, the parameters are automatically adjusted with reference to prior knowledge.

[0025] S29. During operation, continuously utilize the collected multivariable temperature control input dataset and historical operating data to dynamically improve the process prior knowledge base.

[0026] S210. Periodically evaluate the regulation performance of the self-tuning PID controller configured in each heating section, and adaptively optimize the parameter identification method, dynamic weight adjustment strategy and process prior knowledge invocation method according to the actual operating status.

[0027] Optionally, S3 specifically includes:

[0028] S31. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. Each individual in each population is encoded as a combination of PID control parameters for that section. The PID control parameter combination includes the proportional coefficient K. p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient w p w i w d ;

[0029] S32. Based on prior process knowledge and historical operating data, the cooperative population corresponding to each heating section is initialized using experience-guided methods. The PID control parameter combination of each individual population is set to be distributed within the historical optimal range, and parameter perturbation is performed in combination with process characteristics.

[0030] S33. Construct a fitness function F, which integrates the temperature control error, response time, energy consumption, and synergistic coupling index of temperature changes between heating zones in each heating zone.

[0031] S34. Establish a multi-level information exchange mechanism, divide the cooperative populations corresponding to all heating sections into local cooperative groups according to their physical location and process relevance, and exchange high-frequency parameter information within each local cooperative group.

[0032] S35. Within each local cooperative group, the corresponding cooperative population individuals generate a new generation of PID control parameter combinations through crossover and mutation operations, and are screened according to the fitness function.

[0033] S36. On a global scale, parameter information is periodically synchronized between different local collaborative groups to perform parameter migration or share the best individual.

[0034] S37. During individual variation, limit the variation range of PID control parameter combinations to within the range recommended by prior process knowledge to prevent parameter invalidity or abnormal changes.

[0035] S38. Based on the real-time adjustment results of the dynamic cooperative weight and coupling penalty coefficient in the fitness function, continuously optimize the contribution of the cooperative population corresponding to each heating section in the overall fitness, and dynamically balance the impact of different heating sections on temperature control performance.

[0036] S39. Repeat steps S34 to S38 to iterate and optimize until the evolution termination condition is met, and obtain the globally optimal multi-segment PID control parameter combination, including the proportional coefficient, integral coefficient, derivative coefficient and dynamic weight coefficient of each heating segment.

[0037] S310. During operation, the collected temperature control results, energy consumption and process indicators are periodically used to adaptively optimize the multi-level information interaction mechanism, dynamic collaborative weight, coupling penalty coefficient and experience-guided mechanism.

[0038] Optionally, S4 specifically includes:

[0039] S41. Based on the optimal multi-segment PID parameter combination, the proportional coefficient K of each heating segment is... p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient w p w i w d Distribute to their respective tuned PID controllers;

[0040] S42. Each PID controller is individually tuned and combined with a multivariable temperature control input dataset to extract the target temperature r(t), current actual temperature y(t), and ambient humidity h in real time. env (t), Extrusion speed v ex (t) and other process variables are used to calculate the comprehensive temperature error e. * (t);

[0041] S43, Each PID controller is tuned based on the assigned PID parameter combination and the overall temperature error e. * (t), output control signal u(t);

[0042] S44, Each PID controller is tuned based on the comprehensive temperature error e * (t) and process variables change, and the proportional coefficient, integral coefficient, derivative coefficient and dynamic weight coefficient are adjusted online adaptively to continuously correct the PID parameters to adapt to the actual working conditions;

[0043] S45. Each PID controller is tuned and combined with prior process knowledge to apply range constraints to the adjusted PID parameters.

[0044] S46. Periodically evaluate the regulation performance of the self-tuning PID controller, and optimize the parameter adjustment method based on temperature control accuracy, energy consumption and response characteristics.

[0045] Optionally, S5 specifically includes:

[0046] S51. Using the adjusted PID parameters and the self-tuning PID controllers of each section, the closed-loop temperature regulation of each heating section is implemented, and the heating power is adjusted in real time to make the temperature of each heating section approach the target value.

[0047] S52. Real-time acquisition of actual temperature and set target temperature for each heating zone, and monitoring of the deviation between the temperature of each heating zone and the target value;

[0048] S53. Record the temperature regulation data and operating status of each heating section, and dynamically analyze the stability and response speed of the regulation process of each heating section.

[0049] S54. To address the thermal coupling effect between heating zones, implement coordinated adjustment between zones, and suppress interference and coupling effects in the temperature control process through parameter coordination and information interaction.

[0050] S55. Continuously collect temperature control results, energy consumption data, and key process indicators for each heating zone to form data records.

[0051] Optionally, S6 specifically includes:

[0052] S61. Periodically collect temperature control results, energy consumption data and process indicators of each heating zone to form a complete feedback dataset;

[0053] S62. Feed the feedback dataset to the co-evolutionary algorithm to comprehensively evaluate the temperature control effect, energy consumption level and process indicators of the current multi-segment PID parameter combination;

[0054] S63. Based on the comprehensive evaluation results of the co-evolutionary algorithm, dynamically optimize the parameter combination of the self-tuning PID controller in each heating section and update the coefficient K. p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient w p w i w d ;

[0055] S64. Continuously monitor the temperature control operation status of each heating zone and automatically detect whether there are any abnormalities or deviations from the set standards in the temperature control results, energy consumption and process indicators.

[0056] S65. If any abnormality in temperature control, energy consumption, or process parameters is detected in any section, the corresponding cooperative population is re-initialized, and the optimization and adaptive adjustment operations of the cooperative evolution algorithm are re-executed.

[0057] The beneficial effects of this invention are:

[0058] This invention overcomes many shortcomings of existing technologies in terms of multi-segment coupling, intelligent parameter adjustment, and system adaptability by introducing adaptive control algorithms, co-evolutionary mechanisms, and dynamic fusion of multi-source process data into the cable extrusion temperature control system. By configuring self-tuning PID controllers for each heating segment and utilizing co-evolutionary algorithms to achieve global optimization and coordinated adjustment of parameters between segments, this invention not only improves the adaptability of the temperature control system to changes in raw materials, environmental disturbances, and process switching, but also dynamically optimizes the temperature control strategy for each segment, thereby reducing temperature overshoot and fluctuations and ensuring the uniformity and stability of temperature in each segment. Compared with traditional independent temperature control or fixed parameter methods, this invention provides real-time feedback of temperature control results, energy consumption, and process indicators to the algorithm module, continuously performing adaptive parameter optimization and dynamic retuning, achieving closed-loop intelligent optimization of the extrusion process. When the system detects a temperature control anomaly in any segment, it can automatically trigger a parameter reset mechanism and quickly restore to the optimal control state, greatly improving the fault tolerance of the temperature control system and the continuous operation capability of the production line. This invention effectively improves the automation, intelligence, and lean management of cable extrusion temperature control, significantly improves product quality consistency and production efficiency, and provides a solid technical guarantee for the upgrading of high-end cable manufacturing processes. Attached Figure Description

[0059] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0060] Figure 1 This is a flowchart of a cable extrusion temperature control optimization method based on an adaptive control algorithm proposed in this invention;

[0061] Figure 2 This is a structural block diagram of a multi-segment self-tuning PID control and co-evolutionary algorithm fusion method for cable extrusion temperature control optimization based on adaptive control algorithm proposed in this invention. Detailed Implementation

[0062] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0063] refer to Figure 1 and Figure 2 A method for optimizing cable extrusion temperature control based on an adaptive control algorithm includes the following steps:

[0064] S1. Collect multi-source process data during the cable extrusion process, preprocess the multi-source process data, and form a multivariable temperature control input dataset;

[0065] S2. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone and set the initial values ​​of the PID parameters for each self-tuning PID controller.

[0066] S3. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. The individual codes of each population are the PID control parameter combinations for the corresponding section. Co-evolutionary operations are performed on each population to obtain the optimal multi-segment PID parameter combinations.

[0067] S4. Based on the optimal multi-segment PID parameter combination, the parameters are allocated to their respective tuned PID controllers. Each tuned PID controller, in conjunction with the multivariable temperature control input dataset, performs online adaptive adjustment of the PID parameters.

[0068] S5. Using the adjusted PID parameters and the self-tuning PID controllers of each section, closed-loop temperature regulation is implemented for each heating section, the deviation between the temperature of each section and the target value is monitored in real time, and the coupling effect between sections is managed in a coordinated manner. At the same time, temperature control results, energy consumption and process indicators are collected.

[0069] S6. The collected temperature control results, energy consumption and process indicators are periodically fed back to the co-evolutionary algorithm for parameter optimization and dynamic adjustment. If any temperature control anomaly is detected in any segment, the global parameter reset mechanism is automatically triggered to reinitialize the relevant population and re-execute optimization and adaptive adjustment.

[0070] In this embodiment, the multi-source process data specifically includes the real-time temperature of each heating zone, extrusion speed, ambient temperature and humidity, and process parameters related to the extrusion process.

[0071] In this embodiment, the preprocessing of multi-source process data specifically includes denoising, normalizing, and standardizing the collected multi-source process data to generate a multivariable temperature control input dataset in a unified format.

[0072] In this embodiment, S2 specifically includes:

[0073] S21. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone, analyze the process status and historical operating data of each heating zone, and set the initial values ​​of the PID parameters of each tuning PID controller in combination with prior process knowledge.

[0074] S22. Combining the process prior knowledge base and historical operating data, generate recommended PID parameter ranges for each heating section and determine the proportional coefficient K. p (0), Integral coefficient K i (0) and differential coefficient K d The initial value of (0);

[0075] S23. The dynamic characteristic parameters of the heating section are identified online by using the recursive least squares method and introducing a forgetting factor.

[0076] S24. In the parameter identification process, abnormal data suppression method and adaptive gain adjustment method are adopted to improve the robustness and accuracy of parameter identification in response to abnormal disturbances and data noise.

[0077] S25. Use the identified dynamic characteristic parameters of the heating section to correct the proportional coefficient K in real time. p (t), integral coefficient K i (t), differential coefficient K d (t);

[0078] S26. Introduce dynamic weighting coefficients w for the proportional coefficient, integral coefficient, and derivative coefficient, respectively. p (t), w i (t), w d (t), the weights of each parameter are dynamically adjusted according to temperature deviation, adjustment speed and process characteristics;

[0079] S27. Calculate the dynamically weighted PID control parameters. The output signal of the self-tuning PID controller is u(t):

[0080]

[0081] Where e(t) is the error between the set temperature and the actual temperature;

[0082] S28. Combine the process prior knowledge base to apply range constraints to the real-time updated PID parameters. When the parameters exceed the recommended range or the adjustment effect is not good, the parameters are automatically adjusted with reference to prior knowledge.

[0083] S29. During operation, continuously utilize the collected multivariable temperature control input dataset and historical operating data to dynamically improve the process prior knowledge base.

[0084] S210. Periodically evaluate the regulation performance of the self-tuning PID controller configured in each heating section, and adaptively optimize the parameter identification method, dynamic weight adjustment strategy and process prior knowledge invocation method according to the actual operating status.

[0085] In this embodiment, S3 specifically includes:

[0086] S31. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. Each individual in each population is encoded as a combination of PID control parameters for that section. The PID control parameter combination includes the proportional coefficient K. p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient wp w i w d ;

[0087] S32. Based on prior process knowledge and historical operating data, the cooperative population corresponding to each heating section is initialized using experience-guided methods. The PID control parameter combination of each individual population is set to be distributed within the historical optimal range, and parameter perturbation is performed in combination with process characteristics.

[0088] S33. Construct a fitness function F, which integrates the temperature control error, response time, energy consumption, and synergistic coupling index of temperature changes between heating zones in each heating zone:

[0089]

[0090] Where N is the total number of heating sections, β i (t) represents the dynamic cooperative weight of the cooperative population corresponding to the i-th heating segment, f i Let γ(t) be the sub-fitness of the i-th heating segment, and γ(t) be the coupling penalty coefficient that adaptively adjusts with the operating state. i+1 (t)-y i (t)| represents the difference in actual temperature between adjacent heating sections;

[0091] S34. Establish a multi-level information exchange mechanism, divide the cooperative populations corresponding to all heating sections into local cooperative groups according to their physical location and process relevance, and exchange high-frequency parameter information within each local cooperative group.

[0092] S35. Within each local cooperative group, the corresponding cooperative population individuals generate a new generation of PID control parameter combinations through crossover and mutation operations, and are screened according to the fitness function.

[0093] S36. On a global scale, parameter information is periodically synchronized between different local collaborative groups to perform parameter migration or share the best individual.

[0094] S37. During individual variation, limit the variation range of PID control parameter combinations to within the range recommended by prior process knowledge to prevent parameter invalidity or abnormal changes.

[0095] S38. Based on the real-time adjustment results of the dynamic cooperative weight and coupling penalty coefficient in the fitness function, continuously optimize the contribution of the cooperative population corresponding to each heating section in the overall fitness, and dynamically balance the impact of different heating sections on temperature control performance.

[0096] S39. Repeat steps S34 to S38 to iterate and optimize until the evolution termination condition is met, and obtain the globally optimal multi-segment PID control parameter combination, including the proportional coefficient, integral coefficient, derivative coefficient and dynamic weight coefficient of each heating segment.

[0097] S310. During operation, the collected temperature control results, energy consumption and process indicators are periodically used to adaptively optimize the multi-level information interaction mechanism, dynamic collaborative weight, coupling penalty coefficient and experience-guided mechanism.

[0098] In this embodiment, S4 specifically includes:

[0099] S41. Based on the optimal multi-segment PID parameter combination, the proportional coefficient K of each heating segment is... p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient w p w i w d Distribute to their respective tuned PID controllers;

[0100] S42. Each PID controller is individually tuned and combined with a multivariable temperature control input dataset to extract the target temperature r(t), current actual temperature y(t), and ambient humidity h in real time. env (t), Extrusion speed v ex (t) and other process variables are used to calculate the comprehensive temperature error e. * (t):

[0101]

[0102] Where α1, α2, and α3 are weighting coefficients. This is a reference value for ambient humidity. Here is the reference value for extrusion speed, and t is the current time.

[0103] S43, Each PID controller is tuned based on the assigned PID parameter combination and the overall temperature error e. * (t), output control signal u(t);

[0104] S44, Each PID controller is tuned based on the comprehensive temperature error e * (t) and process variables change, and the proportional coefficient, integral coefficient, derivative coefficient and dynamic weight coefficient are adjusted online adaptively to continuously correct the PID parameters to adapt to the actual working conditions;

[0105] S45. Each PID controller is tuned and combined with prior process knowledge to apply range constraints to the adjusted PID parameters.

[0106] S46. Periodically evaluate the regulation performance of the self-tuning PID controller, and optimize the parameter adjustment method based on temperature control accuracy, energy consumption and response characteristics.

[0107] In this embodiment, S5 specifically includes:

[0108] S51. Using the adjusted PID parameters and the self-tuning PID controllers of each section, the closed-loop temperature regulation of each heating section is implemented, and the heating power is adjusted in real time to make the temperature of each heating section approach the target value.

[0109] S52. Real-time acquisition of actual temperature and set target temperature for each heating zone, and monitoring of the deviation between the temperature of each heating zone and the target value;

[0110] S53. Record the temperature regulation data and operating status of each heating section, and dynamically analyze the stability and response speed of the regulation process of each heating section.

[0111] S54. To address the thermal coupling effect between heating zones, implement coordinated adjustment between zones, and suppress interference and coupling effects in the temperature control process through parameter coordination and information interaction.

[0112] S55. Continuously collect temperature control results, energy consumption data, and key process indicators for each heating zone to form data records.

[0113] In this embodiment, S6 specifically includes:

[0114] S61. Periodically collect temperature control results, energy consumption data and process indicators of each heating zone to form a complete feedback dataset;

[0115] S62. Feed the feedback dataset to the co-evolutionary algorithm to comprehensively evaluate the temperature control effect, energy consumption level and process indicators of the current multi-segment PID parameter combination;

[0116] S63. Based on the comprehensive evaluation results of the co-evolutionary algorithm, dynamically optimize the parameter combination of the self-tuning PID controller in each heating section and update the coefficient K. p Integral coefficient K i Differential coefficient K d and dynamic weighting coefficient w p w i w d ;

[0117] S64. Continuously monitor the temperature control operation status of each heating zone and automatically detect whether there are any abnormalities or deviations from the set standards in the temperature control results, energy consumption and process indicators.

[0118] S65. If any abnormality in temperature control, energy consumption, or process parameters is detected in any section, the corresponding cooperative population is re-initialized, and the optimization and adaptive adjustment operations of the cooperative evolution algorithm are re-executed.

[0119] Example 1:

[0120] To verify the feasibility of this invention in practice, it was applied to a smart manufacturing plant for cables. For a long time, temperature control in the extruder heating section of the production line has faced problems such as slow response, fixed parameters that are difficult to adapt to complex working conditions, high energy consumption, and fluctuating product yield. The factory originally used a traditional zoned independent PID temperature control system, which could not achieve comprehensive adjustment of multiple process parameters. Whenever there was high summer temperature or low winter temperature, raw material batch changes, or extrusion speed adjustments, it often led to temperature runaway or overshoot in multiple sections. Temperature differences in some heating sections reached as high as ±4.2℃, resulting in problems such as bubbles and rough surfaces on the product outer sheath. In more severe cases, the production line required frequent shutdowns for manual parameter adjustments, resulting in monthly energy consumption 7.5% higher than similar factories in the industry, and a scrap rate as high as 2.4%.

[0121] In March 2025, the factory upgraded one of its 110kV high-voltage cable extrusion production lines using a cable extrusion temperature control optimization method based on an adaptive control algorithm proposed in this invention. During the upgrade, each heating section was equipped with a PID controller with self-tuning capabilities. All controllers were networked with multi-source sensors for temperature, humidity, speed, and other parameters, as well as process data acquisition ports. Data from the entire line, after preprocessing, was fed into a co-evolutionary algorithm module, which automatically optimized the PID parameters based on the actual operating conditions and historical experience of each section. The co-evolutionary algorithm not only considers the temperature control of each section but also fully incorporates the effects of energy consumption, process coupling, and historical fluctuations, dynamically adjusting the PID parameters and weights across multiple sections. The system can detect temperature control anomalies in real time and autonomously trigger parameter resets, ensuring stable production operation at all times.

[0122] Between March and May 2025, the upgraded production line successfully trial-produced 20 batches of high-voltage cross-linked polyethylene insulated cables, covering various scenarios involving changes in raw material batches, ambient temperature and humidity, and extrusion speed. Data monitoring showed that the average temperature overshoot in each heating zone was controlled within ±1.4℃, the temperature control recovery time was significantly shortened, and the product qualification rate was significantly improved. Energy consumption was reduced by 8.2% compared to the same period before the upgrade, saving an average of 5.8 kWh per kilometer. Even during the peak production load in early May, with an ambient temperature of 35℃, the temperature control system could automatically adapt to various disturbances without manual intervention, completing adaptive parameter reset and returning to normal within 90 seconds.

[0123] Table 1 Comparison of data before and after the application of cable extrusion temperature control optimization based on adaptive control algorithm.

[0124]

[0125] As shown in Table 1, the temperature control optimization method for cable extrusion based on an adaptive control algorithm proposed in this invention significantly improves the temperature control performance and overall process performance of the cable production line. The temperature overshoot range decreased dramatically from ±4.2℃ to ±1.4℃, a reduction of 66.7%, indicating more precise temperature control and effectively preventing quality defects caused by excessive temperature fluctuations during extrusion. The temperature stabilization time was shortened from 17 minutes to 6.8 minutes, improving production response efficiency and enabling faster adaptation to changing operating conditions and products, greatly enhancing production cycle time and flexibility.

[0126] In terms of energy consumption, the energy consumption per kilometer decreased from 73.5 kWh / km to 67.7 kWh / km, representing an energy saving of 7.9%. This not only helped enterprises reduce production costs but also contributed to green manufacturing and energy efficiency improvement goals. The first-pass yield of cable products increased from 96.7% to 99.2%, a 2.5% improvement. This corresponds to the decrease in the scrap rate from 2.4% to 0.6%, indicating that optimized temperature control enhanced production stability, significantly reduced defective products, and further improved product consistency and market competitiveness.

[0127] In terms of system intelligence and anomaly response, the average automatic recovery time for temperature control anomalies has been significantly reduced from 310 seconds to 90 seconds. This indicates that the system can quickly and adaptively recover from process fluctuations or emergencies without manual intervention, greatly reducing downtime risks and workload. The number of production line shutdowns caused by changes in environmental temperature and humidity has decreased from 3.2 times per month to 0, further demonstrating the invention's adaptability and robustness under complex disturbances. Overall, these data comprehensively reflect the multi-dimensional improvements in production efficiency, energy consumption, product quality, and intelligent operation and maintenance levels brought about by this invention, strongly supporting its widespread application value in the modern cable manufacturing field.

[0128] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for optimizing cable extrusion temperature control based on an adaptive control algorithm, characterized in that, Includes the following steps: S1. Collect multi-source process data during the cable extrusion process, preprocess the multi-source process data, and form a multivariable temperature control input dataset; S2. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone and set the initial values ​​of the PID parameters for each self-tuning PID controller. S3. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. The individual codes of each population are the PID control parameter combinations for the corresponding section. Co-evolutionary operations are performed on each population to obtain the optimal multi-segment PID parameter combinations. S4. Based on the optimal multi-segment PID parameter combination, the parameters are allocated to their respective tuned PID controllers. Each tuned PID controller, in conjunction with the multivariable temperature control input dataset, performs online adaptive adjustment of the PID parameters. S5. Using the adjusted PID parameters and the self-tuning PID controllers of each section, closed-loop temperature regulation is implemented for each heating section, the deviation between the temperature of each section and the target value is monitored in real time, and the coupling effect between sections is managed in a coordinated manner. At the same time, temperature control results, energy consumption and process indicators are collected. S6. The collected temperature control results, energy consumption and process indicators are periodically fed back to the co-evolutionary algorithm for parameter optimization and dynamic adjustment. If any temperature control abnormality is detected in any segment, the global parameter reset mechanism is automatically triggered to reinitialize the relevant population and re-execute optimization and adaptive adjustment. S3 specifically includes: S31. Using a co-evolutionary algorithm, each heating section is assigned to multiple co-evolutionary populations. Each individual in each population is encoded as a combination of PID control parameters for that section. The PID control parameter combination includes a proportional gain. Integral coefficient Differential coefficients and dynamic weighting coefficients , , ; S32. Based on prior process knowledge and historical operating data, the cooperative population corresponding to each heating section is initialized using experience-guided methods. The PID control parameter combination of each individual population is set to be distributed within the historical optimal range, and parameter perturbation is performed in combination with process characteristics. S33. Constructing the fitness function The fitness function integrates the temperature control error, response time, energy consumption, and synergistic coupling index of temperature changes between heating zones in each heating zone. S34. Establish a multi-level information exchange mechanism, divide the cooperative populations corresponding to all heating sections into local cooperative groups according to their physical location and process relevance, and exchange high-frequency parameter information within each local cooperative group. S35. Within each local cooperative group, the corresponding cooperative population individuals generate a new generation of PID control parameter combinations through crossover and mutation operations, and are screened according to the fitness function. S36. On a global scale, parameter information is periodically synchronized between different local collaborative groups to perform parameter migration or share the best individual. S37. During individual variation, limit the variation range of PID control parameter combinations to within the range recommended by prior process knowledge to prevent parameter invalidity or abnormal changes. S38. Based on the real-time adjustment results of the dynamic cooperative weight and coupling penalty coefficient in the fitness function, continuously optimize the contribution of the cooperative population corresponding to each heating section in the overall fitness, and dynamically balance the impact of different heating sections on temperature control performance. S39. Repeat steps S34 to S38 to iterate and optimize until the evolution termination condition is met, and obtain the globally optimal multi-segment PID control parameter combination, including the proportional coefficient, integral coefficient, derivative coefficient and dynamic weight coefficient of each heating segment. S310. During operation, the collected temperature control results, energy consumption and process indicators are periodically used to adaptively optimize the multi-level information interaction mechanism, dynamic collaborative weight, coupling penalty coefficient and experience-guided mechanism. S4 specifically includes: S41. Based on the optimal multi-segment PID parameter combination, adjust the proportional coefficient of each heating segment. Integral coefficient Differential coefficients and dynamic weighting coefficients , , Distribute to their respective tuned PID controllers; S42. Each PID controller is individually tuned and combined with a multivariable temperature control input dataset to extract the target temperature in real time. Current actual temperature Ambient humidity Extrusion speed Process variables, calculate the overall temperature error ; S43. Each PID controller is tuned based on the assigned PID parameter combination and the overall temperature error. Output control signal ; S44, Each PID controller is tuned based on the overall temperature error. In response to changes in process variables, the proportional coefficient, integral coefficient, derivative coefficient, and dynamic weighting coefficient are adaptively adjusted online to continuously correct PID parameters to adapt to actual working conditions. S45. Each PID controller is tuned and combined with prior process knowledge to apply range constraints to the adjusted PID parameters. S46. Periodically evaluate the regulation performance of the self-tuning PID controller, and optimize the parameter adjustment method based on temperature control accuracy, energy consumption and response characteristics.

2. The cable extrusion temperature control optimization method based on adaptive control algorithm according to claim 1, characterized in that, The multi-source process data specifically includes the real-time temperature of each heating zone, extrusion speed, ambient temperature and humidity, and process parameters related to the extrusion process.

3. The cable extrusion temperature control optimization method based on adaptive control algorithm according to claim 1, characterized in that, The preprocessing of multi-source process data specifically includes denoising, normalizing, and standardizing the collected multi-source process data to generate a unified format multivariate temperature control input dataset.

4. The cable extrusion temperature control optimization method based on adaptive control algorithm according to claim 1, characterized in that, S2 specifically includes: S21. Based on the multivariate temperature control input dataset, configure a self-tuning PID controller for each heating zone, analyze the process status and historical operating data of each heating zone, and set the initial values ​​of the PID parameters of each tuning PID controller in combination with prior process knowledge. S22. Combining the process prior knowledge base and historical operating data, generate recommended PID parameter ranges for each heating section and determine the proportional coefficient. Integral coefficient and differential coefficients The initial value; S23. The dynamic characteristic parameters of the heating section are identified online by using the recursive least squares method and introducing a forgetting factor. S24. In the parameter identification process, abnormal data suppression method and adaptive gain adjustment method are adopted to improve the robustness and accuracy of parameter identification in response to abnormal disturbances and data noise. S25. Use the identified dynamic characteristic parameters of the heating section to correct the proportional coefficient in real time. Integral coefficient Differential coefficients ; S26. Introduce dynamic weighting coefficients for the proportional coefficient, integral coefficient, and derivative coefficient, respectively. , , The weights of each parameter are dynamically adjusted based on temperature deviation, adjustment speed, and process characteristics. S27. Calculate the dynamically weighted PID control parameters. The output signal of the self-tuning PID controller is: ; S28. Combine the process prior knowledge base to apply range constraints to the real-time updated PID parameters. When the parameters exceed the recommended range or the adjustment effect is not good, the parameters are automatically adjusted with reference to prior knowledge. S29. During operation, continuously utilize the collected multivariable temperature control input dataset and historical operating data to dynamically improve the process prior knowledge base. S210. Periodically evaluate the regulation performance of the self-tuning PID controller configured in each heating section, and adaptively optimize the parameter identification method, dynamic weight adjustment strategy and process prior knowledge invocation method according to the actual operating status.

5. The cable extrusion temperature control optimization method based on adaptive control algorithm according to claim 1, characterized in that, S5 specifically includes: S51. Using the adjusted PID parameters and the self-tuning PID controllers of each section, the closed-loop temperature regulation of each heating section is implemented, and the heating power is adjusted in real time to make the temperature of each heating section approach the target value. S52. Real-time acquisition of actual temperature and set target temperature for each heating zone, and monitoring of the deviation between the temperature of each heating zone and the target value; S53. Record the temperature regulation data and operating status of each heating section, and dynamically analyze the stability and response speed of the regulation process of each heating section. S54. To address the thermal coupling effect between heating zones, implement coordinated adjustment between zones, and suppress interference and coupling effects in the temperature control process through parameter coordination and information interaction. S55. Continuously collect temperature control results, energy consumption data, and key process indicators for each heating zone to form data records.

6. The cable extrusion temperature control optimization method based on adaptive control algorithm according to claim 1, characterized in that, S6 specifically includes: S61. Periodically collect temperature control results, energy consumption data and process indicators of each heating zone to form a complete feedback dataset; S62. Feed the feedback dataset to the co-evolutionary algorithm to comprehensively evaluate the temperature control effect, energy consumption level and process indicators of the current multi-segment PID parameter combination; S63. Based on the comprehensive evaluation results of the co-evolutionary algorithm, dynamically optimize the parameter combination of the self-tuning PID controller in each heating section and update the coefficients. Integral coefficient Differential coefficients and dynamic weighting coefficients , , ; S64. Continuously monitor the temperature control operation status of each heating zone and automatically detect whether there are any abnormalities or deviations from the set standards in the temperature control results, energy consumption and process indicators. S65. If any abnormality in temperature control, energy consumption, or process parameters is detected in any section, the corresponding cooperative population is re-initialized, and the optimization and adaptive adjustment operations of the cooperative evolution algorithm are re-executed.

Images