Method and system for integrated control of high pressure glass fiber pipeline pipe production

Abstract

The application relates to the technical field of composite material manufacturing, and particularly discloses a high-pressure glass fiber pipeline pipe production integrated control method and system. The method comprises the following steps: collecting parameters in real time, removing abnormal values and performing sliding average filtering; calculating and outputting a fiber-resin system health index and a winding process dynamic performance index according to the processed data; and executing a hierarchical adaptive control process according to the fiber-resin system health index and the winding process dynamic performance index. The application synchronously collects multiple variables and performs filtering processing, thereby obtaining key parameters in a production process. The fiber-resin system health index and the winding process dynamic performance index are calculated, so that complex production states are quantified into intuitive and determinable indexes. In combination with a hierarchical control execution process, the problems of single control response and lack of hierarchy and synergy in decision-making in traditional methods are effectively solved.

Description

Technical Field

[0001] This invention relates to the field of composite material manufacturing technology, specifically to an integrated control method and system for the production of high-pressure glass fiber pipelines. Background Technology

[0002] High-pressure fiberglass conduits are increasingly used in critical fields such as petrochemicals, marine engineering, and high-pressure fluid transportation due to their excellent corrosion resistance, high strength-to-weight ratio, and long service life. The core of manufacturing these pipes lies in the fiber winding process, which involves winding continuous resin-impregnated fibers at a specific angle onto the surface of a rotating mandrel, forming a multi-layered composite structure after curing. However, current production control systems face multiple technical bottlenecks.

[0003] First, there is a deep coupling relationship between mechanical and chemical system parameters. For example, fiber tension fluctuations can disrupt resin impregnation uniformity, while changes in resin viscosity can conversely affect tension transmission efficiency. Existing control systems generally employ a distributed, independent loop design, implementing individual setpoint tracking control for each parameter, such as traditional PID control strategies. This fragmented control mode cannot capture the dynamic interactions between parameters, resulting in a severe deficiency in the overall system state perception capability. Operators can only rely on lagging indicators such as yarn breakage rate statistics or offline visual inspections for experience-based judgment, making it difficult to achieve early warning and systematic intervention. This makes it difficult to guarantee product consistency and severely restricts the intelligent upgrading of high-pressure glass fiber pipeline manufacturing processes. Summary of the Invention

[0004] The purpose of this invention is to provide an integrated control method and system for the production of high-pressure glass fiber pipelines, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: An integrated control method for the production of high-pressure glass fiber pipelines, the method comprising: S1: Real-time acquisition of fiber tension, resin viscosity, resin temperature, mandrel rotation speed, actual and set values ​​of winding angle, and overall tension feedback and set values; and removal of outliers and performance of moving average filtering on the time-series data of each acquired parameter. S2: Based on the preprocessed data in S1, calculate and output the fiber-resin system health index and the dynamic performance index of the winding process; S3: Based on the fiber-resin system health index and winding process dynamic performance index output by S2, execute the hierarchical adaptive control process, including real-time stability fine-tuning, strategic parameter reset and collaborative judgment.

[0006] As a further aspect of the present invention: the calculation of the health index of the fiber-resin system in step S2 specifically includes: S21: For the preprocessed data in S1, calculate the coefficient of variation of fiber tension, the percentage deviation of resin viscosity relative to the set value, the percentage deviation of resin temperature relative to the set value, and the yarn breakage rate per unit time; wherein, the yarn breakage rate per unit time is calculated by dividing the number of yarn breaks detected in a specific time period by the theoretical number of fiber strands or lengths that should be wound in that time period; yarn breakage detection is achieved through a preset sensor. S22: Map the parameters calculated in S21 to a unified degradation range of zero to one: The degree of deterioration of the fiber tension variation coefficient is the smaller value between its division by the preset upper limit threshold and 1; The degree of deterioration of the percentage deviation of resin viscosity is the smaller value between its absolute value divided by the preset upper limit threshold and 1; The degree of degradation of the resin temperature percentage deviation is the smaller value between its absolute value divided by the preset upper limit threshold and 1. The degree of deterioration of the yarn breakage rate per unit time is the smaller value between its division by the preset upper limit threshold and 1; S23: Calculate the fiber-resin system health index using the weighted arithmetic mean method: Multiply the four deterioration degrees obtained in S22 by their respective preset weight coefficients and sum them up. The sum of the four weight coefficients is one. Subtract the sum from one to obtain the fiber-resin system health index. The closer the value is to one, the healthier the system is.

[0007] As a further aspect of the present invention: the weighting coefficients in S23 satisfy the following: the weight corresponding to the resin viscosity deviation is greater than the weight corresponding to the fiber tension variation coefficient, the weight corresponding to the fiber tension variation coefficient is greater than the weight corresponding to the yarn breakage rate, and the weight corresponding to the yarn breakage rate is greater than the weight corresponding to the resin temperature deviation.

[0008] As a further aspect of the present invention: the calculation of the dynamic performance index of the winding process in S2 specifically includes: S24: Based on the preprocessed data in S1, calculate the absolute deviation between the winding angle set value and the actual value, and the absolute deviation between the overall tension set value and the feedback value; at the same time, calculate the standard deviation of the fiber tension within the set sliding time window, and the standard deviation of the mandrel rotation speed within the same window. S25: Map the parameters calculated in S24 and the fiber-resin system health index output in S2 to a unified performance degradation range of zero to one: The performance degradation of the absolute deviation of the winding angle is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation degree of the overall tension absolute deviation is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation of the fiber tension standard deviation is the smaller value between its division by the preset upper threshold and 1; The performance degradation of the standard deviation of the core mold rotation speed is the smaller value between its division by the preset upper limit threshold and 1; The degree of influence of material health status on performance degradation is 1 minus the value of the fiber-resin system health index; S26: The dynamic performance index of the winding process is calculated using a weighted synthesis method based on exponential decay: the five degradation degrees obtained in S25 are multiplied by preset influence weights and then summed to obtain the comprehensive degradation sum; the dynamic performance index of the winding process is calculated; the closer the dynamic performance index of the winding process is to a certain value, the better the dynamic performance of the process; the specific expression of the dynamic performance index of the winding process is as follows: , This represents the dynamic performance index of the winding process, where k is a preset sensitivity coefficient. This indicates overall degradation.

[0009] As a further aspect of the present invention: the influence weights in S26 satisfy the following: the sum of the weights corresponding to the winding angle deviation and the overall tension deviation is greater than the sum of the weights corresponding to the fiber tension standard deviation and the mandrel rotation speed standard deviation, and the sum of the latter two weights is greater than the weight corresponding to the influence degree of material health status.

[0010] As a further aspect of the present invention, the sensitivity coefficient k in S26 has a value range of [2, 5].

[0011] As a further aspect of the present invention: S3 specifically includes: S31: When the standard deviation of fiber tension or the standard deviation of mandrel rotation speed exceeds their respective first threshold, the judgment process experiences short-term fluctuations. Based on the main frequency characteristics of the fluctuations, the proportional-integral-derivative parameters of the tension controller are adaptively adjusted within a preset range. S32: When the fiber-resin system health index is continuously lower than the second threshold for the first time period, or the winding process dynamic performance index is continuously lower than the third threshold for the second time period, according to the dominant factors of the deterioration of the fiber-resin system health index and the winding process dynamic performance index, the pre-stored strategy table is queried, and the basic tension setting value or resin temperature setting value is adjusted in a stepwise manner. S33: If the fine-tuning instruction of S31 and the reset instruction of S32 act on the same actuator and conflict in direction, the instruction of S32 shall be executed first, and the fine-tuning of S31 shall be paused for one adjustment cycle. The pre-stored strategy table is a two-dimensional lookup table indexed by the fiber-resin system health index and the winding process dynamic performance index. Its contents are optimized through correlation analysis between historical production data and final product quality.

[0012] The present invention also provides an integrated control system for the production of high-pressure glass fiber pipelines. This system is used to implement the aforementioned integrated control method for the production of high-pressure glass fiber pipelines. The system includes: The data preprocessing module is used to collect fiber tension, resin viscosity, resin temperature, mandrel speed, actual and set values ​​of winding angle, and overall tension feedback and set values ​​in real time. It also removes outliers and performs moving average filtering on the collected time-series data of each parameter. The condition assessment module is used to calculate and output the health index of the fiber-resin system and the dynamic performance index of the winding process based on the data preprocessed by the data processing module. The hierarchical control execution module is used to execute a hierarchical adaptive control process based on the fiber-resin system health index and winding process dynamic performance index output by the state assessment module, including real-time stability fine-tuning, strategic parameter reset, and collaborative judgment.

[0013] Compared with existing technologies, the beneficial effects of this invention are as follows: By synchronously collecting and filtering multiple variables such as fiber tension, resin viscosity, resin temperature, mandrel rotation speed, winding angle, and overall tension, this invention achieves comprehensive, real-time monitoring and high-quality data acquisition of key parameters in the production process; by calculating the fiber-resin system health index and the winding process dynamic performance index, complex production states are quantified into intuitive and verifiable indicators; combined with the hierarchical control execution process, it effectively solves the problems of single control response and lack of hierarchy and coordination in decision-making in traditional methods; for example, for short-term, local fluctuations, the system can quickly and accurately intervene through real-time stability fine-tuning to prevent the problem from escalating; while for long-term, trend-based deterioration, it can make fundamental adjustments through strategic parameter reset; when two control commands may conflict, the introduced collaborative decision-making architecture can ensure the coordination and consistency of control actions, avoiding mutual interference and instability that may occur in traditional multi-loop control. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0015] Figure 1 A flowchart of an integrated control method for the production of high-pressure glass fiber pipelines.

[0016] Figure 2 This is a structural diagram of an integrated control system for the production of high-pressure glass fiber pipelines. Detailed Implementation

[0017] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0018] Figure 1 This is a flowchart illustrating an integrated control method for the production of high-pressure fiberglass pipelines. In this embodiment of the invention, a method for integrated control of high-pressure fiberglass pipeline production includes: S1: Real-time acquisition of fiber tension, resin viscosity, resin temperature, mandrel rotation speed, actual and set values ​​of winding angle, and overall tension feedback and set values; and removal of outliers and performance of moving average filtering on the time-series data of each acquired parameter. S2: Based on the preprocessed data in S1, calculate and output the fiber-resin system health index and the dynamic performance index of the winding process; S3: Based on the fiber-resin system health index and winding process dynamic performance index output by S2, execute the hierarchical adaptive control process, including real-time stability fine-tuning, strategic parameter reset and collaborative judgment.

[0019] The above process involves three steps. Step S1 is the data acquisition and preprocessing process, which begins with real-time data acquisition. Specifically, various sensors can be used to continuously monitor key parameters on the production line. For example, tension sensors installed on the fiber conveying path acquire fiber tension values; viscometers and temperature sensors installed in the resin tank acquire resin viscosity and temperature values; encoders or speed sensors acquire mandrel speed; vision systems or angle sensors acquire the actual winding angle value and compare it with the preset winding angle setting; and sensors installed on the overall tension control unit acquire the overall tension feedback value and compare it with the overall tension setting. This real-time acquisition of parameters provides comprehensive raw data for subsequent condition assessment. Alternatively, data acquisition can be performed by an independent sensor network, with the data transmitted to the central processing unit.

[0020] Furthermore, the collected time-series data for each parameter undergoes preprocessing. Since sensors may be affected by environmental noise, transient interference, or their own malfunctions, the raw data may contain outliers or high-frequency noise. To improve the reliability and accuracy of the data, it is necessary to process it. For example, statistical methods can be used to remove outliers, such as setting a fixed threshold; when a data point exceeds this threshold, it is marked as an anomaly and replaced or deleted. Alternatively, an anomaly detection algorithm based on historical data distribution can be used. After removing outliers, the data can be processed using a moving average filter to smooth the data curve, eliminate high-frequency noise, and make the data trend clearer. For example, a fixed-length sliding window can be set, and the average value of the data within the window can be calculated as the smoothed value for the current point. These preprocessing measures ensure that the data input to the subsequent evaluation module has high quality and stability.

[0021] Step S2 is the process of calculating the state assessment index. Based on the preprocessed data, the health index of the fiber-resin system and the dynamic performance index of the winding process are calculated and output. The fiber-resin system health index aims to quantify the health status of the material system during the production process. For example, factors such as the fluctuation of fiber tension, the deviation of resin viscosity and temperature, and the frequency of yarn breakage during production can be comprehensively considered. These parameters can be transformed into a unified health index through a mathematical model. The value of this index can intuitively reflect the stability and potential risks of the material system.

[0022] Simultaneously, a dynamic performance index for the winding process is calculated and output. This index aims to quantify the dynamic stability and control effectiveness of the winding process. For example, it can comprehensively consider factors such as the deviation between the actual and set values ​​of the winding angle, the deviation between the overall tension feedback value and the set value, and the fluctuations in fiber tension and mandrel rotation speed within a certain time window. These parameters are then transformed into a unified performance index through a mathematical model. The value of this index can intuitively reflect the stability of the winding process and the responsiveness of the control system. Through the calculation of these two indices, a comprehensive and quantitative perception of the production process status is achieved, overcoming the fragmented and lagging problems of status assessment in traditional methods.

[0023] Step S3 is the hierarchical control execution process. Based on the fiber-resin system health index and the dynamic performance index of the winding process calculated above, hierarchical adaptive control is executed. This control strategy includes real-time stability fine-tuning, strategic parameter resetting, and collaborative determination.

[0024] Specifically, real-time stability fine-tuning is designed to address short-term, localized fluctuations that occur during production. For example, when a slight but persistent fluctuation in fiber tension or mandrel speed is detected, the control system can make small, high-frequency adjustments to the corresponding controller parameters to quickly suppress the fluctuation and restore system stability. This fine-tuning typically operates on the underlying controller, such as adjusting the proportional, integral, or derivative parameters of the PID controller.

[0025] Strategic parameter resets are designed to address long-term trends of deterioration or persistent anomalies in the production process. For example, when the health index of the fiber-resin system or the dynamic performance index of the winding process remains below a certain threshold for an extended period, it indicates a potentially deeper problem. In this case, the control system can adjust the setpoints of key process parameters, such as the base tension setpoint or resin temperature setpoint, based on the dominant factors causing the index deterioration, to fundamentally correct the problem. This type of reset typically operates at the mid-to-high-level control level, making step-wise adjustments to the production status.

[0026] Furthermore, to ensure the consistency of control commands, a collaborative decision-making architecture is introduced. For example, when real-time stability fine-tuning commands and strategic parameter reset commands are issued simultaneously, act on the same actuator (such as a tension controller), and conflict in direction, the arbitration architecture will make a judgment based on preset priority rules. For instance, the strategic parameter reset command can be given higher priority to ensure that long-term optimization goals are achieved, and fine-tuning commands can be temporarily suspended to avoid oscillations or instability in the control system. Through this hierarchical and collaborative architecture, accurate response and effective management of different types of disturbances are achieved.

[0027] In a preferred embodiment of the present invention, the calculation of the health index of the fiber-resin system in step S2 specifically includes: S21. For the pre-processed data in S1, calculate the coefficient of variation of fiber tension, the percentage deviation of resin viscosity relative to the set value, the percentage deviation of resin temperature relative to the set value, and the yarn breakage rate per unit time. S22. Map the parameters calculated in S21 to a unified degradation range of zero to one: The degree of deterioration of the fiber tension variation coefficient is the smaller value between its division by the preset upper limit threshold and 1; The degree of deterioration of the percentage deviation of resin viscosity is the smaller value between its absolute value divided by the preset upper limit threshold and 1; The degree of degradation of the resin temperature percentage deviation is the smaller value between its absolute value divided by the preset upper limit threshold and 1. The degree of deterioration of the yarn breakage rate per unit time is the smaller value between its division by the preset upper limit threshold and 1; S23. Calculate the fiber-resin system health index using the weighted arithmetic mean method: Multiply the four deterioration degrees obtained in S22 by their respective preset weight coefficients and sum them up. The sum of the four weight coefficients is one. Subtract the sum from one to obtain the fiber-resin system health index. The closer the value is to one, the healthier the system is.

[0028] In this embodiment, in step S21, the coefficient of variation of fiber tension, the percentage deviation of resin viscosity relative to the set value, the percentage deviation of resin temperature relative to the set value, and the yarn breakage rate per unit time are calculated for the preprocessed data in S1. These parameters are key indicators for evaluating the health status of the fiber-resin system. The coefficient of variation of fiber tension can be obtained by calculating the ratio of the standard deviation to the mean of fiber tension data over a period of time. For example, within a sliding time window, N fiber tension sampling values ​​are collected, their mean and standard deviation are calculated, and then the ratio is obtained. The percentage deviation of resin viscosity relative to the set value can be calculated by comparing the real-time measured resin viscosity value with the preset ideal viscosity set value, and calculating (actual value - set value) / set value * 100%. The percentage deviation of resin temperature relative to the set value is similar to the percentage deviation of resin viscosity, calculated as (actual value - set value) / set value * 100%. The yarn breakage rate per unit time can be calculated by dividing the number of yarn breaks detected in a specific time period (e.g., one minute or one hour) by the theoretical number of fiber strands or lengths that should be wound in that time period. Yarn breakage detection can be achieved through visual sensors, abnormal drop of tension sensors, or photoelectric sensors.

[0029] In step S22, the parameters calculated in S21 are mapped to a unified deterioration range of zero to one. This step aims to standardize parameters of different dimensions, making them comparable and facilitating subsequent comprehensive evaluation. Specifically, the deterioration degree of the fiber tension variation coefficient can be the smaller value between its division by a preset upper threshold and 1, where the preset upper threshold can be set based on historical data analysis or expert experience. The deterioration degree of the resin viscosity percentage deviation can be the smaller value between its absolute value and a preset upper threshold, where the preset upper threshold defines the maximum acceptable range for viscosity deviation. The deterioration degree of the resin temperature percentage deviation can be the smaller value between its absolute value and a preset upper threshold, where the preset upper threshold defines the maximum acceptable range for temperature deviation. The deterioration degree of the yarn breakage rate per unit time can be the smaller value between its division by a preset upper threshold and 1, where the preset upper threshold defines the maximum acceptable range for yarn breakage rate. These mapping methods ensure that the deterioration degree is between 0 and 1, where 0 represents no deterioration and 1 represents maximum deterioration.

[0030] In step S23, the weighted arithmetic mean method is used to calculate the health index of the fiber-resin system. This step aims to comprehensively consider various deterioration indicators and assign different weights according to their importance to obtain a single indicator that can comprehensively reflect the health status of the system. Specifically, the health index can be obtained by subtracting the sum of the products of each deterioration degree and its respective preset weight coefficient from one, that is: Health Index = 1 - (Deterioration Degree 1 * Weight 1 + Deterioration Degree 2 * Weight 2 + Deterioration Degree 3 * Weight 3 + Deterioration Degree 4 * Weight 4), where the sum of all weight coefficients is one. The weight coefficients can be determined through expert experience, historical data analysis (such as principal component analysis, regression analysis), or machine learning methods (such as neural network training). For example, weights can be allocated according to the sensitivity of different parameters to the final product quality. This method ensures that the closer the health index is to one, the healthier the system; the closer it is to zero, the more severe the system deterioration.

[0031] This invention achieves a comprehensive and accurate quantitative assessment of the health status of a fiber-resin system through the coordinated operation of the aforementioned steps. First, step S21 extracts key characteristic parameters directly related to the health status of the fiber-resin system from the preprocessed raw data. These parameters cover material properties (resin viscosity, temperature) and process stability (fiber tension variation coefficient, yarn breakage rate). Next, step S22 maps these characteristic parameters with different dimensions and physical meanings to a deterioration degree range of zero to one, achieving parameter standardization and normalization, eliminating the influence of dimensional differences on the comprehensive assessment, and allowing direct comparison and weighting of deterioration degrees in different dimensions. Finally, step S23 synthesizes these standardized deterioration degrees using a weighted synthesis method, and based on preset weight coefficients reflecting the importance of each parameter to the system's health status, ultimately calculates the fiber-resin system health index. This index is a single, intuitive numerical value that can comprehensively and accurately quantify the overall health status of the current fiber-resin system. Through this progressive calculation method, this scheme transforms multi-source, heterogeneous process data into a comprehensive index with clear physical meaning and evaluative value, providing a solid data foundation for subsequent intelligent control decisions and solving the problems of fragmented and inaccurate state assessment in traditional methods. The output of this health index provides crucial input for the hierarchical control execution in step S3 of the aforementioned method, enabling the control system to make decisions based on quantified health status, thereby achieving more precise and intelligent control.

[0032] In a preferred embodiment of the present invention, the weighting coefficients in S23 satisfy the following: the weight corresponding to the resin viscosity deviation is greater than the weight corresponding to the fiber tension variation coefficient, the weight corresponding to the fiber tension variation coefficient is greater than the weight corresponding to the yarn breakage rate, and the weight corresponding to the yarn breakage rate is greater than the weight corresponding to the resin temperature deviation.

[0033] In this embodiment, weighting coefficients are numerical factors assigned to different parameters to quantify their relative importance or influence in the comprehensive evaluation index. In the calculation of the fiber-resin system health index, these coefficients determine the contribution of each deterioration factor (such as fiber tension variation, resin viscosity deviation, etc.) to the overall health status assessment. Reasonable setting of weighting coefficients is crucial to ensuring that the calculated index accurately reflects the true state of the system. Methods for determining these coefficients may include, but are not limited to: assignment based on domain expert experience, statistical analysis using historical production data (e.g., principal component analysis, regression analysis of parameters with final product quality), or iterative adjustment using optimization algorithms. The weight corresponding to resin viscosity deviation is important because resin viscosity directly affects the fiber wetting quality and the uniform distribution of resin in the fiber bundle, which plays a decisive role in the mechanical properties and durability of the final pipeline. Assigning a high weight to this parameter aims to highlight its core influence on the system's health status. The weight corresponding to the fiber tension variation coefficient reflects the importance of fiber tension stability during winding. Continuous and uniform fiber tension is key to ensuring fiber laying accuracy, avoiding local stress concentration, and guaranteeing the structural integrity of the composite material. The weight corresponding to the yarn breakage rate directly quantifies the frequency of fiber breakage events during the production process. Yarn breakage not only means material loss but can also create defects within the product, severely affecting its strength and reliability. The weight corresponding to the resin temperature deviation considers the impact of resin temperature deviation from the set value on system health. Resin temperature indirectly affects the resin's viscosity, curing reaction rate, and interfacial bonding performance with fibers. Although its impact may not be as immediate or direct as viscosity or tension fluctuations, long-term or significant temperature deviations can still have a cumulative effect on product quality.

[0034] In a preferred embodiment of the present invention, the calculation of the dynamic performance index of the winding process in step S2 specifically includes: S24. Based on the preprocessed data in S1, calculate the absolute deviation between the winding angle setting value and the actual value, and the absolute deviation between the overall tension setting value and the feedback value; at the same time, calculate the standard deviation of the fiber tension within the set sliding time window, and the standard deviation of the mandrel rotation speed within the same window. S25. Map the parameters calculated in S24 and the fiber-resin system health index output in S2 to a unified performance degradation range of zero to one: The performance degradation of the absolute deviation of the winding angle is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation degree of the overall tension absolute deviation is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation of the fiber tension standard deviation is the smaller value between its division by the preset upper threshold and 1; The performance degradation of the standard deviation of the core mold rotation speed is the smaller value between its division by the preset upper limit threshold and 1; The degree of influence of material health status on performance degradation is 1 minus the value of the fiber-resin system health index; S26. The dynamic performance index of the winding process is calculated using a weighted synthesis method based on exponential decay: the five deterioration degrees obtained in S25 are multiplied by preset influence weights and then summed to obtain the comprehensive deterioration sum; the value of the negative k of the natural constant e multiplied by the comprehensive deterioration sum is used as the dynamic performance index of the winding process, where k is a preset sensitivity coefficient; the closer the dynamic performance index of the winding process is to a certain value, the better the dynamic performance of the process.

[0035] In this embodiment, the absolute deviation between the set and actual winding angle values ​​is calculated to quantify the accuracy of the fiber laying angle during winding. This deviation can be obtained by measuring the actual winding angle value in real time and comparing it with the set winding angle value set by the control system, taking the absolute value of the difference. For example, the actual winding angle value can be acquired in real time using devices such as vision sensors or angle encoders. The absolute deviation between the overall tension set value and the feedback value is calculated to evaluate the accuracy of fiber winding tension control. This deviation can be obtained by comparing the overall tension value fed back in real time by a tension sensor with the overall tension set value set by the control system, taking the absolute value of the difference. For example, the tension sensor can be installed on the fiber guide roller or the take-up / unwinding mechanism. The standard deviation of fiber tension within a set sliding time window is calculated to reflect the degree of fluctuation of fiber tension in a short period of time, which is an important indicator for measuring process stability. This standard deviation can be calculated by continuously collecting fiber tension data within a preset time window (e.g., 5 seconds, 10 seconds, or longer) and using statistical methods to calculate its dispersion. The standard deviation of the mandrel rotation speed within the same window is calculated to reflect the stability of the mandrel rotation speed, which has a direct impact on winding quality. The standard deviation can also be calculated by continuously collecting mandrel rotation speed data within the same sliding time window as the fiber tension and using statistical methods to calculate its dispersion.

[0036] Mapping the calculated parameters and the fiber-resin system health index to a unified performance degradation range of zero to one eliminates the influence of different physical dimensions and numerical ranges on the comprehensive evaluation, allowing all indicators to be compared and weighted on the same scale. This mapping is achieved by dividing the absolute value or standard deviation of each parameter by its respective preset upper limit threshold, taking the smaller value between it and 1, ensuring that the degradation degree is within the range of zero to one. For example, when the absolute deviation of the winding angle is 0.2 degrees and its preset upper limit threshold is 0.5 degrees, the degradation degree is 0.2 / 0.5 = 0.4. The impact of material health status on performance degradation is obtained by subtracting the value of the fiber-resin system health index from one, thereby incorporating the material's own health status into the dynamic performance evaluation, making the evaluation more comprehensive.

[0037] A weighted synthesis method based on exponential decay is employed to calculate the dynamic performance index of the winding process, aiming to more sensitively reflect subtle changes in process dynamic performance through nonlinear weighting. This method first multiplies the five degradation degrees obtained in S25 by preset influence weights and then sums them to obtain a comprehensive degradation sum. The weight coefficients can be set according to the importance of each factor's impact on dynamic performance. Subsequently, the final dynamic performance index of the winding process is obtained by multiplying the negative k of the natural constant e by the power of the comprehensive degradation sum. Here, k is a preset sensitivity coefficient used to adjust the index's sensitivity to degradation levels; a larger value indicates a more sensitive index. This exponential decay characteristic means that when the comprehensive degradation sum is small (i.e., when process performance is good), the index value is close to one; while when the comprehensive degradation sum increases (i.e., when process performance deteriorates), the index value drops rapidly, thus enabling timely detection of declining process performance trends.

[0038] This invention's technical solution involves real-time acquisition and preprocessing of multi-source data to calculate the absolute deviation between the set and actual winding angle values, the absolute deviation between the set and feedback values ​​of overall tension, the standard deviation of fiber tension within a set sliding time window, and the standard deviation of mandrel rotation speed within the same window. These indicators comprehensively capture the dynamic fluctuations of the mechanical system and process parameters during the production of high-pressure glass fiber pipelines. Subsequently, these dynamic indicators, along with the fiber-resin system health index, are mapped to a unified performance degradation range, achieving standardized processing of parameters of different types and dimensions, laying the foundation for subsequent comprehensive evaluation. Based on this, a weighted synthesis method based on exponential decay is used to sum the standardized degradation degrees using weighted summations and perform nonlinear transformation through an exponential function, ultimately obtaining the dynamic performance index of the winding process. This method not only considers the independent influence of various dynamic parameters but also incorporates the potential impact of material state on dynamic performance by introducing the fiber-resin system health index, making the evaluation results more comprehensive and accurate. The exponential decay characteristic makes this index more sensitive to minor degradations in the process, enabling it to promptly reflect the quality of process dynamic performance and providing accurate and real-time decision-making basis for subsequent hierarchical control execution. In this way, the present invention can effectively solve the problems of fragmented dynamic performance evaluation and delayed response in traditional methods, and provides key support for achieving refined control of the high-pressure glass fiber pipeline production process.

[0039] In a preferred embodiment of the present invention, the influence weights in S26 satisfy the following: the sum of the weights corresponding to the winding angle deviation and the overall tension deviation is greater than the sum of the weights corresponding to the fiber tension standard deviation and the mandrel rotation speed standard deviation, and the sum of the latter two weights is greater than the weight corresponding to the influence degree of material health status.

[0040] In this embodiment, when calculating the dynamic performance index of the winding process, the influence weights are used to quantify the contribution of each performance degradation degree to the final index. These weights can be preset constant values ​​or parameters dynamically adjusted based on historical data or expert experience. Their function is to distinguish the importance of different parameters to the dynamic performance of the process, ensuring that key factors play a dominant role in the evaluation. For example, the influence weights can be set based on the experience of process engineers, or determined through statistical methods such as regression analysis and principal component analysis of historical production data. The sum of the weights corresponding to the winding angle deviation and the overall tension deviation refers to the sum of the weights multiplied by the performance degradation degree of the absolute deviation of the winding angle and the performance degradation degree of the absolute deviation of the overall tension when calculating the dynamic performance index of the winding process. The winding angle deviation directly affects the fiber laying accuracy and the structural integrity of the pipe, while the overall tension deviation is directly related to the prestress state of the fiber and the mechanical properties of the pipe. Therefore, assigning high weights to these two parameters aims to highlight their core position in the dynamic performance evaluation of the process, ensuring that these factors that directly affect product quality are given priority. The sum of the weights corresponding to the standard deviation of fiber tension and the standard deviation of mandrel speed refers to the sum of the weights of the performance degradation degree of the standard deviation of fiber tension and the performance degradation degree of the standard deviation of mandrel speed, respectively. The standard deviation of fiber tension reflects the stability of the fiber winding process; excessive fluctuations may lead to uneven fiber arrangement or breakage. The standard deviation of mandrel speed reflects the stability of the winding speed; its fluctuations will affect the uniformity of the winding angle. These two parameters mainly reflect the stability of the process, and their sum of weights is set to be second only to the winding angle deviation and the overall tension deviation, but higher than the influence of material health status, to ensure that the stability of the process is fully considered. The weight corresponding to the influence of material health status refers to the weight of the influence of material health status on the performance degradation. The influence of material health status is calculated based on the fiber-resin system health index and reflects the indirect influence of the comprehensive state of raw materials (fiber and resin) on process performance. Because its influence is relatively indirect and is usually indirectly reflected through other parameters (such as resin viscosity and fiber tension), its weight is set to the lowest to avoid secondary factors excessively interfering with the direct assessment of process dynamic performance.

[0041] This invention ensures that the calculation of the dynamic performance index of the winding process accurately and effectively reflects key issues in the production process by clearly defining the weighting rules. Specifically, winding angle deviation and overall tension deviation directly reflect key indicators of product geometric accuracy and mechanical properties, and their fluctuations have a decisive impact on the final product quality. Therefore, the sum of their corresponding weights is set to the maximum to prioritize the capture and response to these core deviations. Secondly, fiber tension standard deviation and mandrel speed standard deviation characterize the stability of the winding process, and their stability is an important prerequisite for ensuring product consistency. Therefore, the sum of their corresponding weights is set to the next highest to ensure that the dynamic stability of the process is fully considered. Finally, although the influence of material health status has an indirect impact on process performance, its effect is relatively indirect, and some of its influence is already reflected through other parameters (such as fiber tension and resin viscosity). Therefore, its weight is set to the minimum. This hierarchical weighting structure allows the calculation of the dynamic performance index of the winding process to highlight the main contradictions and accurately identify the dominant factors leading to performance degradation. This provides a more accurate and reliable basis for subsequent hierarchical control execution, avoids evaluation distortion caused by improper weighting, and ultimately improves the decision-making efficiency and accuracy of the entire control system.

[0042] In a preferred embodiment of the present invention, the sensitivity coefficient k in S26 takes a value in the range of [2, 5].

[0043] In this embodiment, the sensitivity coefficient k is a key parameter in the weighted synthesis method based on exponential decay used to calculate the dynamic performance index of the winding process. It determines the response speed and sensitivity of the index to changes in the overall degradation. Specifically, the larger the k value, the faster the exponential decay, meaning the more sensitive the dynamic performance index is to changes in the overall degradation; conversely, the smaller the k value, the slower the exponential decay, and the less sensitive the dynamic performance index is to changes in the overall degradation. This invention limits the range of the sensitivity coefficient k to [2, 5] to ensure that the dynamic performance index achieves a moderate response characteristic when reflecting the dynamic performance of the process. For example, setting this range can prevent the index from overreacting to slight, instantaneous fluctuations in the production process due to an excessively small k value, thereby reducing unnecessary control interventions. At the same time, it also ensures that the index will not be sluggish in responding to actual, continuous performance degradation due to an excessively large k value, thereby delaying necessary corrective measures.

[0044] This method optimizes the calculation process of the dynamic performance index of the winding process by limiting the sensitivity coefficient k to the range of [2, 5]. In step S26, the calculation of the dynamic performance index of the winding process depends on the negative k of the natural constant e multiplied by the sum of the overall degradation. This sensitivity coefficient k directly affects the rate of exponential decay. When the value of k is too small, the exponential decay curve is flat, and even if the overall degradation and sum change significantly, the change in the dynamic performance index may not be obvious, resulting in a sluggish response of the system to performance degradation. Conversely, when the value of k is too large, the exponential decay curve is steep, and even small changes in the overall degradation and sum may cause drastic fluctuations in the dynamic performance index, making the system overly sensitive to normal process noise and frequently triggering control actions. Therefore, setting the value of k to the range of [2, 5] allows the dynamic performance index to operate within a reasonable sensitivity range, which can effectively filter out random noise and short-term fluctuations in the production process, avoid over-control, and promptly capture substantial performance degradation caused by material or equipment problems, ensuring that the control system can intervene in a timely manner. This optimized dynamic performance index, as an important basis for hierarchical control execution (S3), can more accurately guide real-time stability fine-tuning (S31) and strategic parameter reset (S32), thereby improving the robustness and effectiveness of the entire high-pressure glass fiber pipeline production integrated control method.

[0045] In a preferred embodiment of the present invention, step S3 specifically includes: S31. Real-time stability fine-tuning: When the standard deviation of fiber tension or the standard deviation of mandrel rotation speed exceeds their respective first threshold, the judgment process will experience short-term fluctuations. Based on the main frequency characteristics of the fluctuations, the proportional-integral-derivative parameters of the tension controller will be adaptively adjusted within a preset range. S32. Strategic parameter reset: When the fiber-resin system health index continues to be lower than the second threshold for the first time period, or the winding process dynamic performance index continues to be lower than the third threshold for the second time period, according to the dominant factors of the deterioration of the fiber-resin system health index and the winding process dynamic performance index, the pre-stored strategy table is queried, and the basic tension setting value or resin temperature setting value is adjusted in a stepwise manner. S33. Coordination Decision: If the fine-tuning instruction of S31 and the reset instruction of S32 act on the same actuator and conflict in direction, the instruction of S32 shall be executed first, and the fine-tuning of S31 shall be suspended for one adjustment cycle.

[0046] The pre-stored strategy table is a two-dimensional lookup table indexed by the fiber-resin system health index and the dynamic performance index of the winding process. Its contents are optimized through correlation analysis of historical production data and final product quality. For example, "multivariate regression analysis is used to establish the mapping relationship between health / performance index and quality indicators, and the optimal parameter adjustment table is derived by trial and error or optimization algorithm."

[0047] In this embodiment, the real-time stability fine-tuning S31 aims to address short-term, instantaneous fluctuations during production. It adaptively adjusts the proportional-integral-derivative (PID) parameters of the tension controller to quickly suppress fluctuations and restore process stability. Its function is to provide a fast-response architecture, preventing short-term disturbances from having a cumulative impact on product quality. As a possible implementation, an adaptive fuzzy PID control strategy can be employed. Through a fuzzy logic inference system, the fuzzy rules or membership functions of the PID parameters are adjusted online based on the real-time monitored fluctuation amplitude and frequency, thereby achieving flexible adaptive control for different types of short-term fluctuations.

[0048] The strategic parameter reset S32 is used to address performance degradation trends that occur during long-term system operation. By querying a pre-stored strategy table, it makes step adjustments to the base tension setpoint or resin temperature setpoint to fundamentally correct the dominant factors causing the degradation trend. Its role is to provide a macro-level, strategic adjustment framework to prevent continuous system performance decline. One possible implementation is that the pre-stored strategy table can be a rule base built based on expert experience and historical production data, containing a series of condition-action rules such as "if the fiber-resin system health index is below X and the dominant degradation factor is resin viscosity, then increase the resin temperature setpoint by Y degrees." Another possible implementation is that the strategy table can be trained and optimized using offline machine learning algorithms (e.g., decision trees, support vector machines, or neural networks). By analyzing a large amount of historical production data, including the fiber-resin system health index, winding process dynamic performance index, dominant degradation factors, and final product quality data, the optimal parameter adjustment strategy is learned and solidified into a lookup table or rule set.

[0049] The collaborative decision S33 aims to address the problem of conflicting directions when real-time stability fine-tuning commands and strategic parameter reset commands act simultaneously on the same actuator. By setting priorities, it ensures the coordination and consistency of control commands and avoids mutual interference. Its role is to guarantee the overall stability and response efficiency of the control system. As a possible implementation, a priority judgment module can be set up in the control system. When a conflict is detected between the fine-tuning command of S31 and the reset command of S32, this module will select to execute the higher-priority S32 command according to preset priority rules (e.g., the command priority of S32 is higher than that of S31), and temporarily suppress or suspend the fine-tuning command of S31 for one adjustment cycle.

[0050] This pre-stored strategy table is a key component for implementing the strategic parameter reset S32. Indexed by the fiber-resin system health index and the winding process dynamic performance index, it stores parameter adjustment suggestions for different deterioration conditions. Its content is optimized through correlation analysis of historical production data and final product quality, ensuring the scientific validity and effectiveness of the adjustment strategy. As one possible implementation, this two-dimensional lookup table can be stored in the controller's non-volatile memory in matrix form, with its rows and columns corresponding to different ranges of the fiber-resin system health index and the winding process dynamic performance index, respectively. Each cell stores one or a set of suggested parameter adjustment values ​​(e.g., the adjustment amount of the base tension setpoint or the adjustment amount of the resin temperature setpoint).

[0051] The hierarchical control execution method proposed in this invention constructs an intelligent control system that balances short-term fluctuation suppression and long-term trend correction through the organic combination of three steps: real-time stability fine-tuning (S31), strategic parameter reset (S32), and collaborative judgment (S33). This is further enhanced by a collaborative judgment architecture to ensure the coordination and consistency of control commands. Based on data acquisition and preprocessing (S1) and state evaluation index calculation (S2), the system can acquire in real-time key indicators such as the fiber-resin system health index, the winding process dynamic performance index, and the standard deviation of fiber tension and mandrel speed. These indicators provide comprehensive state awareness for subsequent hierarchical control. When the system detects that the standard deviation of fiber tension or mandrel speed exceeds its corresponding first threshold, it indicates that short-term fluctuations may have occurred in the production process. At this time, real-time stability fine-tuning (S31) is triggered. This fine-tuning architecture can adaptively adjust the proportional-integral-derivative parameters of the tension controller within a preset amplitude range according to the specific dominant frequency characteristics of the fluctuation. This adjustment is rapid and localized, aiming to quickly suppress transient disturbances and prevent them from evolving into more serious problems. Simultaneously, the system continuously monitors the fiber-resin system health index and the winding process dynamic performance index. When the fiber-resin system health index remains below the second threshold for a first time period, or the winding process dynamic performance index remains below the third threshold for a second time period, this indicates a potential deeper and more persistent performance degradation in the system. At this point, the strategic parameter reset S32 is activated. This step queries a pre-stored strategy table based on the dominant factors contributing to the degradation of the fiber-resin system health index and the winding process dynamic performance index, and makes a step-wise adjustment to the base tension setpoint or resin temperature setpoint. This adjustment is global and strategic, aiming to address the root causes of long-term degradation, such as changes in material state or drift in process parameters. The pre-stored strategy table is optimized through correlation analysis of historical production data and final product quality, ensuring the scientific basis and reliability of the adjustment decisions. When the two control architectures operate in parallel, situations may arise where the fine-tuning command of S31 and the reset command of S32 act on the same actuator and conflict in direction. For example, S31 may need to reduce tension due to short-term fluctuations, while S32 may need to increase the base tension due to a long-term degradation trend. To avoid instability or malfunction of the control system due to such conflicts, the collaborative decision-making S33 architecture comes into play. S33 stipulates that if such a conflict occurs, the instruction of S32 is executed first, and the fine-tuning of S31 is paused for one adjustment cycle. This architecture ensures that strategic, global adjustments take precedence over local, instantaneous fine-tuning, thereby avoiding mutual cancellation or oscillation of control instructions and guaranteeing the coordinated operation and overall stability of the system under complex operating conditions. Through this hierarchical, collaborative control strategy, this invention can effectively address short-term fluctuations and long-term degradation trends in the production process of high-pressure glass fiber pipelines, achieving refined management and optimized control of the production process.

[0052] As a preferred embodiment of the technical solution of the present invention, the technical solution of the present invention also provides an integrated control system for the production of high-pressure glass fiber pipelines, the system 10 comprising: The data preprocessing module 11 is used to collect fiber tension value, resin viscosity value, resin temperature value, mandrel speed, actual and set values ​​of winding angle, and overall tension feedback value and set values ​​in real time, and to remove outliers and perform moving average filtering on the collected time-series data of each parameter; it is used to execute step S1 in the above content. The condition assessment module 12 is used to calculate and output the health index of the fiber-resin system and the dynamic performance index of the winding process based on the data preprocessed by the data processing module; it is used to execute step S2 in the above content. The hierarchical control execution module 13 is used to execute the hierarchical adaptive control process based on the fiber-resin system health index and winding process dynamic performance index output by the state assessment module, including real-time stability fine-tuning, strategic parameter reset and collaborative judgment; and to execute step S3 in the above content.

[0053] In the above description, the data acquisition and preprocessing module is a component of the system that acquires and initially processes raw production data. This module aims to ensure the accuracy, reliability, and consistency of the input data, providing a high-quality data foundation for subsequent status assessments and avoiding assessment biases caused by anomalies or noise in the raw data. This module can be a hardware system consisting of sensors, a data acquisition card, an industrial Ethernet interface, and a data processing unit, programmed to perform functions such as data filtering and outlier removal. Alternatively, this module can be a software module running on an industrial control computer, acquiring data from field devices via protocols such as OPC UA and Modbus TCP / IP, and preprocessing it using digital signal processing algorithms.

[0054] The condition assessment index calculation module is responsible for quantitatively assessing the health status and dynamic performance of the production process based on preprocessed data. This module provides comprehensive quantitative assessment indicators for system condition and process dynamics, namely the fiber-resin system health index and the winding process dynamic performance index, thereby overcoming the limitations of single-parameter assessments and making control decisions more holistic and scientific. This module can be a software program running on an embedded system or industrial PC. It receives preprocessed data and calculates the corresponding fiber-resin system health index and winding process dynamic performance index based on preset mathematical models and algorithms. Alternatively, this module can be an analysis service based on cloud computing or edge computing, receiving data through a data interface, using machine learning models or expert systems for condition assessment and index calculation, and returning the results to the control system.

[0055] The hierarchical control execution module is a component of the system that adaptively adjusts production parameters hierarchically based on evaluation indices to maintain or optimize the process. This module aims to ensure hierarchical differentiation and conflict coordination of control actions, avoid confusion between short-term fluctuations and long-term trends, improve control accuracy and stability, and achieve refined management of the production process. This module can be a hierarchical control structure composed of advanced controllers, PLCs, or servo drives, and its logic for real-time stability fine-tuning, strategic parameter resetting, and collaborative decision-making is implemented through programming.

[0056] This integrated control system for high-pressure fiberglass pipeline production achieves comprehensive perception, intelligent evaluation, and hierarchical control of complex production processes through modular design. Its overall operational logic is as follows: The data acquisition and preprocessing module, acting as the system's perception layer, continuously and in real-time acquires multi-dimensional, high-frequency key process parameter data from the production site. It performs preliminary quality control on this time-series data, effectively removing noise and occasional interference through outlier removal and moving average filtering, ensuring that the data input to subsequent evaluation stages is accurate, reliable, and representative. This preprocessing stage is fundamental to the stable operation of the entire system, preventing misjudgments and erroneous control due to raw data quality issues. Next, the preprocessed clean data is sent to the state evaluation index calculation module. This module is the core of the system's intelligent analysis; through complex algorithms and models, it comprehensively calculates and outputs the fiber-resin system health index and the winding process dynamic performance index. The calculation of these two indices abstracts discrete, variable low-level data into high-level, easily understood, and decision-making comprehensive indicators, thereby overcoming the fragmented state perception problem in traditional control and providing a comprehensive and in-depth basis for subsequent control decisions. Finally, the hierarchical control execution module executes hierarchical adaptive control based on the fiber-resin system health index and winding process dynamic performance index output by the state assessment index calculation module. This module's control strategy is not rigid or singular, but rather adopts different levels of response based on the nature and severity of the problem: for short-term, high-frequency process fluctuations, it performs real-time stability fine-tuning to quickly suppress fluctuations; when the fiber-resin system health index or winding process dynamic performance index continues to deteriorate, indicating a deeper problem, it triggers a more impactful strategic parameter reset to fundamentally correct the deviation; to avoid potential conflicts between different control levels or strategies, this module also possesses an intelligent collaborative judgment architecture to ensure the coordination and effectiveness of control actions. Through this integrated system design, data acquisition, state assessment, and hierarchical control are organically combined, achieving closed-loop management from bottom-level data to top-level decision-making. The data acquisition and preprocessing module provides an accurate foundation for assessment, the state assessment index calculation module transforms complex data into decision-making indicators, and the hierarchical control execution module then performs precise and coordinated interventions based on these indicators. This close integration and collaborative operation enables the system to respond in real time to various disturbances and changes in the production process, effectively solving the problems of low execution efficiency, insufficient real-time performance, and poor coordination in traditional methods, and significantly improving the process stability, product consistency, and overall production efficiency of high-pressure glass fiber pipeline production.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An integrated control method for the production of high-pressure glass fiber pipelines, characterized in that, The method includes: S1: Real-time acquisition of fiber tension, resin viscosity, resin temperature, mandrel rotation speed, actual and set values ​​of winding angle, and overall tension feedback and set values; and removal of outliers and performance of moving average filtering on the time-series data of each acquired parameter. S2: Based on the preprocessed data in S1, calculate and output the health index of the fiber-resin system and the dynamic performance index of the winding process; S3: Based on the fiber-resin system health index and winding process dynamic performance index output by S2, execute the hierarchical adaptive control process, including real-time stability fine-tuning, strategic parameter reset and collaborative judgment.

2. The integrated control method for high-pressure glass fiber pipeline production according to claim 1, characterized in that, The calculation of the health index of the fiber-resin system in S2 specifically includes: S21: For the preprocessed data in S1, calculate the coefficient of variation of fiber tension, the percentage deviation of resin viscosity relative to the set value, the percentage deviation of resin temperature relative to the set value, and the yarn breakage rate per unit time; wherein, the yarn breakage rate per unit time is calculated by dividing the number of yarn breaks detected in a specific time period by the theoretical number of fiber strands or lengths that should be wound in that time period; yarn breakage detection is achieved through a preset sensor. S22: Map the parameters calculated in S21 to a unified degradation range of zero to one: The degree of deterioration of the fiber tension variation coefficient is the smaller value between its division by the preset upper limit threshold and 1; The degree of deterioration of the percentage deviation of resin viscosity is the smaller value between its absolute value divided by the preset upper limit threshold and 1; The degree of degradation of the resin temperature percentage deviation is the smaller value between its absolute value divided by the preset upper limit threshold and 1. The degree of deterioration of the yarn breakage rate per unit time is the smaller value between its division by the preset upper limit threshold and 1; S23: Calculate the fiber-resin system health index using the weighted arithmetic mean method: Multiply the four deterioration degrees obtained in S22 by their respective preset weight coefficients and sum them up. The sum of the four weight coefficients is one. Subtract the sum from one to obtain the fiber-resin system health index. The closer the value is to one, the healthier the system is.

3. The integrated control method for high-pressure glass fiber pipeline production according to claim 2, characterized in that, The weighting coefficients in S23 satisfy the following conditions: the weight corresponding to the resin viscosity deviation is greater than the weight corresponding to the fiber tension variation coefficient, the weight corresponding to the fiber tension variation coefficient is greater than the weight corresponding to the yarn breakage rate, and the weight corresponding to the yarn breakage rate is greater than the weight corresponding to the resin temperature deviation.

4. The integrated control method for high-pressure glass fiber pipeline production according to claim 1, characterized in that, The calculation of the dynamic performance index of the winding process in S2 specifically includes: S24: Based on the preprocessed data in S1, calculate the absolute deviation between the winding angle set value and the actual value, and the absolute deviation between the overall tension set value and the feedback value; at the same time, calculate the standard deviation of the fiber tension within the set sliding time window, and the standard deviation of the mandrel rotation speed within the same window. S25: Map the parameters calculated in S24 and the fiber-resin system health index output in S2 to a unified performance degradation range of zero to one: The performance degradation of the absolute deviation of the winding angle is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation degree of the overall tension absolute deviation is the smaller value between its division by the preset upper limit threshold and 1; The performance degradation of the fiber tension standard deviation is the smaller value between its division by the preset upper threshold and 1; The performance degradation of the standard deviation of the core mold rotation speed is the smaller value between its division by the preset upper limit threshold and 1; The degree of influence of material health status on performance degradation is 1 minus the value of the fiber-resin system health index; S26: The dynamic performance index of the winding process is calculated using a weighted synthesis method based on exponential decay: the five degradation degrees obtained in S25 are multiplied by preset influence weights and then summed to obtain the comprehensive degradation sum; the dynamic performance index of the winding process is calculated; the closer the dynamic performance index of the winding process is to a certain value, the better the dynamic performance of the process; the specific expression of the dynamic performance index of the winding process is as follows: , This represents the dynamic performance index of the winding process, where k is a preset sensitivity coefficient. This indicates overall degradation.

5. The integrated control method for high-pressure glass fiber pipeline production according to claim 4, characterized in that, The influence weights in S26 satisfy the following: the sum of the weights corresponding to the winding angle deviation and the overall tension deviation is greater than the sum of the weights corresponding to the fiber tension standard deviation and the mandrel rotation speed standard deviation, and the sum of the latter two weights is greater than the weight corresponding to the influence degree of material health status.

6. The integrated control method for high-pressure glass fiber pipeline production according to claim 5, characterized in that, The sensitivity coefficient k in S26 has a value range of [2, 5].

7. The method according to claim 1, characterized in that, S3 specifically includes: S31: When the standard deviation of fiber tension or the standard deviation of mandrel rotation speed exceeds their respective first threshold, the judgment process experiences short-term fluctuations. Based on the main frequency characteristics of the fluctuations, the proportional-integral-derivative parameters of the tension controller are adaptively adjusted within a preset range. S32: When the fiber-resin system health index is continuously lower than the second threshold for the first time period, or the winding process dynamic performance index is continuously lower than the third threshold for the second time period, according to the dominant factors of the deterioration of the fiber-resin system health index and the winding process dynamic performance index, the pre-stored strategy table is queried, and the basic tension setting value or resin temperature setting value is adjusted in a stepwise manner. S33: If the fine-tuning instruction of S31 and the reset instruction of S32 act on the same actuator and conflict in direction, the instruction of S32 shall be executed first, and the fine-tuning of S31 shall be paused for one adjustment cycle. The pre-stored strategy table is a two-dimensional lookup table indexed by the fiber-resin system health index and the winding process dynamic performance index. Its contents are optimized through correlation analysis between historical production data and final product quality.

8. An integrated control system for the production of high-pressure glass fiber pipelines, the system being used to implement the integrated control method for the production of high-pressure glass fiber pipelines as described in any one of claims 1 to 7, characterized in that, The system includes: The data preprocessing module is used to collect fiber tension, resin viscosity, resin temperature, mandrel speed, actual and set values ​​of winding angle, and overall tension feedback and set values ​​in real time. It also removes outliers and performs moving average filtering on the collected time-series data of each parameter. The condition assessment module is used to calculate and output the health index of the fiber-resin system and the dynamic performance index of the winding process based on the data preprocessed by the data processing module. The hierarchical control execution module is used to execute a hierarchical adaptive control process based on the fiber-resin system health index and winding process dynamic performance index output by the state assessment module, including real-time stability fine-tuning, strategic parameter reset, and collaborative judgment.

Images