Intelligent control system and method for on-line yarn control based on industrial visualization and PLC closed-loop control

Abstract

The application discloses a pultrusion process yarn feeding intelligent control system and method based on industrial visualization and PLC closed-loop control, which comprises a hardware layer, a PLC closed-loop control layer, a software management layer and a visualization interaction layer which are sequentially communicated and connected; the hardware layer is used for collecting physical parameter data in the yarn feeding process and executing control instructions; the PLC closed-loop control layer is communicated and connected with the hardware layer by using a modular PLC controller, and real-time receives the physical parameter data collected and transmitted by the hardware layer; the software management layer is used for realizing hierarchical permission management of administrators and operators, setting, storing and calling a preset process parameter library of different product models, carrying out fault diagnosis and generating fault treatment suggestions; the visualization interaction layer is communicated with the upper computer monitoring software through an industrial touch screen, and is used for graphical real-time monitoring of yarn feeding process parameters, multi-state fault color code warning, process parameter configuration and issuing, historical data trend analysis and intelligent report automatic generation.

Description

Technical Field

[0001] This invention belongs to the field of composite material pultrusion molding process control technology, specifically relating to an intelligent control system and method for yarn feeding in pultrusion processes based on industrial visualization and PLC closed-loop control. Background Technology

[0002] Pultrusion is one of the core processes in composite material molding. Yarn loading, as the initial step in pultrusion production, directly determines the fiber content, uniformity of yarn arrangement, and mechanical properties of the product. Traditional yarn loading relies heavily on manual operation, requiring workers to adjust key parameters such as yarn tension, yarn unwinding speed, and yarn arrangement position based on experience. This method has the following technical drawbacks: 1. Low precision of manual control: Large fluctuations in yarn tension (error usually exceeds ±15%), which can easily lead to yarn breakage or accumulation, affecting the uniformity of product wall thickness; disordered yarn arrangement, overlapping, and offset problems, reducing the stability of product mechanical properties.

[0003] 2. Lack of real-time monitoring: It is impossible to intuitively obtain key information such as yarn running status, tension changes, and remaining yarn spools, resulting in delayed fault detection and a high production scrap rate (usually 8%-12%).

[0004] 3. Low production efficiency: Manual parameter adjustment is time-consuming, and readjustment is required when changing production lines. In addition, each production line requires 1-2 full-time yarn-feeding workers, resulting in high labor costs.

[0005] 4. Data cannot be traced: There are no records of parameter adjustments and malfunctions during the yarn feeding process, which is not conducive to production quality analysis and process optimization.

[0006] In the existing technology, some pultrusion equipment attempts to use a single PLC for simple tension control, but without combining visualization technology and dedicated software systems, there are problems such as unintuitive parameter settings, fixed control logic, and inability to achieve multi-parameter linkage adjustment, which makes it difficult to meet the needs of high-precision and intelligent production. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent control system and method for pultrusion yarn loading based on industrial visualization and PLC closed-loop control. This solves the problems of low tension control accuracy, lack of monitoring means, poor speed synchronization, passive fault response and weak data management in the prior art, and achieves high-precision, high-stability and intelligent control of the yarn loading process.

[0008] To achieve the above objectives, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides an intelligent control system for yarn loading in pultrusion process based on industrial visualization and PLC closed-loop control, comprising a hardware layer, a PLC closed-loop control layer, a software management layer and a visualization interaction layer that are sequentially connected in communication. The hardware layer is used to collect physical parameter data during the yarn feeding process, including yarn tension value, yarn feeding speed, yarn spool state, and yarn arrangement image, and to execute control commands to adjust the yarn feeding speed and tension. The PLC closed-loop control layer adopts a modular PLC controller, which communicates with the hardware layer through a high-speed I / O module. It receives physical parameter data collected and transmitted by the hardware layer in real time, and sends control commands to the hardware layer and real-time fault signals to the software management layer based on the analysis results of the physical parameter data. The software management layer, developed based on an industrial control software platform, is used to implement hierarchical permission management for administrators and operators, set, store and call preset process parameter libraries for different product models, parse, filter and store real-time data uploaded by the PLC closed-loop control layer, and perform fault diagnosis and generate fault handling suggestions based on historical data and real-time fault signals. The visualization and interaction layer, through an industrial touch screen and host computer monitoring software, is used for graphical real-time monitoring of yarn loading process parameters, multi-state fault color-coded warnings, process parameter configuration and distribution, historical data trend analysis, and automatic generation of intelligent reports.

[0009] Furthermore, the hardware layer includes a tension detection module, a speed detection module, a yarn spool state detection module, a control execution module, and an image acquisition module; The tension detection module uses a high-precision tension sensor with a measurement accuracy of ±0.5%FS, which is installed on the yarn feeding path of each yarn to collect the tension value of a single yarn in real time. The speed detection module uses an encoder coaxially connected to the yarn pay-off roller to collect the pay-off speed and cumulative pay-off length. The yarn bundle status detection module includes an infrared sensor and a weight sensor, which are used to detect the yarn bundle breakage status and the remaining weight of the yarn bundle, respectively, and trigger a yarn replenishment reminder when the remaining weight is lower than the threshold set in the preset process parameter library. The control execution module includes a servo motor with a control accuracy of ±0.1 rpm and an electromagnetic brake. The servo motor is used to drive the wire feeding roller to adjust the wire feeding speed, and the electromagnetic brake is used to adjust the braking torque according to the tension feedback to regulate the tension closed-loop control. The image acquisition module uses an industrial camera with a resolution of ≥1920×1080 and a frame rate of ≥30fps, which is installed in the yarn confluence area to acquire images of the yarn arrangement in real time.

[0010] Furthermore, the PLC closed-loop control layer includes a data receiving module, a closed-loop control module, a linkage control module, and a fault early warning and processing module; The data receiving module receives the tension value, unwinding speed, yarn weight, yarn breakage signal, and yarn arrangement image data of the hardware layer in real time at a sampling frequency of ≥100Hz. The closed-loop control module, based on the preset tension threshold in the preset process parameter library and the real-time collected tension data, adjusts the servo motor speed and the electromagnetic brake torque through a configurable PID algorithm, so that the yarn tension is stabilized within the set range of the preset process parameter library and the fluctuation range is ≤±3%. The linkage control module automatically matches the yarn feeding speed according to the traction speed of the pultrusion host to synchronize the yarn feeding amount with the traction speed; The fault early warning processing module performs abnormal detection of physical parameter data. When it detects a broken yarn signal, insufficient yarn ball balance, or tension value exceeding a set threshold, it sends a real-time fault alarm signal to the software management layer and executes preset protection actions.

[0011] Furthermore, the software management layer includes a preset parameter module, a data processing and storage module, an image recognition module, a fault diagnosis module, and a permission management module; The preset parameter module is used to preset process parameter libraries for different product models, and to set, store and call tension target values, yarn feeding speed limits, yarn spool balance thresholds and alarm thresholds. The data processing and storage module parses and filters the real-time data transmitted by the PLC closed-loop control layer, and stores key parameter data with a storage period of ≥90 days, which is exported in Excel and CSV formats. The image recognition module analyzes the yarn arrangement image through image recognition algorithm to determine the yarn overlap and offset status, and automatically adjusts the position of the pay-off roller or sends a manual adjustment reminder when the deviation exceeds the set threshold of the preset process parameter library. The fault diagnosis module performs fault diagnosis based on historical data and real-time fault signals, generates fault handling suggestions, and sends them to the visual interaction layer. The permission management module is used for hierarchical permission management for administrators and operators.

[0012] Furthermore, the visualization interaction layer includes a graphics display module, a parameter input module, a data query module, an alarm display module, and a report generation module; The graphic display module displays the tension value of a single yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time in a graphical manner, using green / yellow / red to indicate the normal / warning / alarm status of the parameters; The parameter input module is used to manually input or select preset parameters from the process parameter library and send them to the PLC control layer in real time. The data query module displays the tension change trend and the wire release speed change trend in the form of curves, as well as historical data query and comparison. The alarm display module displays the fault type, occurrence time, and handling suggestions in real time through audible and visual alarms and text prompts. The report generation module automatically generates daily production reports and quality analysis reports that include information such as production duration, product model, parameter fluctuation range, and number of failures.

[0013] Furthermore, the tension closed-loop control of the control execution module adopts a PID algorithm, and its proportional coefficient, integral coefficient, and derivative coefficient are configured online through the intelligent algorithm software layer and adaptively adjusted according to the tension fluctuation to achieve stable control with a tension fluctuation range of ≤±3%. The image recognition module uses the OpenCV open-source library for image recognition algorithm. Through contour detection and position matching algorithms, it identifies the yarn arrangement offset with an accuracy of ≤0.1mm. It calculates the yarn spacing and center line position in real time. When the yarn offset exceeds 2mm or there is overlap, it triggers automatic position adjustment or manual intervention reminder. The linkage control module receives the traction line speed signal v of the pultrusion host in real time through the communication interface. t According to the speed linkage formula v f =k*v t Automatically calculate the target laying speed v f Where k is the linkage matching coefficient; the encoder provides real-time feedback of the actual wire feeding speed, compares it with the target speed to form a deviation, and corrects the servo output through closed-loop adjustment to achieve high-precision synchronous matching between the wire feeding speed and the traction speed.

[0014] The fault early warning and processing module adopts a graded response mechanism: when a broken yarn or severe tension loss is detected, an emergency shutdown protection is immediately executed; when insufficient yarn stock or slight tension fluctuation is detected, an early warning prompt is triggered and the pultrusion host is linked to reduce its speed.

[0015] In a second aspect, the present invention provides a control method for an intelligent control system for yarn feeding in a pultrusion process based on industrial visualization and PLC closed-loop control as described in any of the first aspects, comprising the following steps: Step S1: After the system is powered on, the visual interaction layer starts the industrial touch screen and the host computer monitoring software. The operator logs into the system and selects the corresponding product model's process parameters from the preset process parameter library in the software management layer according to the current production task.

[0016] The process parameters include target tension value, pay-off speed limit, yarn spool balance threshold, and alarm threshold.

[0017] The software management layer sends the parameters to the PLC closed-loop control layer and completes the initialization configuration. At the same time, it performs communication detection and status self-check on the tension detection module, speed detection module, yarn spool status detection module, and image acquisition module to ensure that each hardware module is in normal working condition.

[0018] Step S2: During the yarn feeding process, the hardware layer collects the tension value of each yarn in real time through the tension detection module, collects the yarn feeding speed and cumulative yarn feeding length in real time through the encoder, detects the yarn breakage signal through the infrared sensor and detects the remaining weight of the yarn bundle through the weight sensor. At the same time, the image acquisition module continuously collects the yarn arrangement image in the yarn confluence area. The above-mentioned data are transmitted to the PLC closed-loop control layer in real time at the set sampling frequency for unified processing.

[0019] Step S3: The PLC closed-loop control layer receives data such as tension value, yarn feeding speed, yarn weight, yarn breakage signal and yarn arrangement image transmitted from the hardware layer through the high-speed I / O module, and performs time synchronization and preliminary filtering processing on the received data.

[0020] The processed data is used for closed-loop control calculations on one hand, and uploaded in real time to the software management layer for further data analysis, storage, and historical record management on the other hand.

[0021] Step S4: The PLC closed-loop control layer calculates the tension deviation based on the target tension value in the preset process parameter library and the real-time acquired tension value. It then adjusts the servo motor speed and the electromagnetic brake torque using a PID control algorithm to achieve closed-loop stable control of the yarn tension. Simultaneously, the linkage control module automatically calculates the target unwinding speed based on the pultrusion host traction speed signal and forms a speed closed-loop adjustment through encoder feedback, ensuring that the unwinding speed and traction speed remain synchronized, thereby guaranteeing stable yarn feeding.

[0022] Step S5: The software management layer calls the image recognition module to analyze the yarn arrangement image captured by the industrial camera. It calculates the yarn spacing and center line offset through contour detection and position matching algorithms to determine whether there is yarn overlap or arrangement misalignment. When the yarn offset exceeds the allowable range in the preset process parameter library, the system automatically adjusts the position of the pay-off roller or issues a manual adjustment prompt to the operator to ensure that the yarn arrangement is neat and stable.

[0023] Step S6: The fault early warning processing module of the PLC closed-loop control layer performs real-time anomaly detection on the collected physical parameters. When a broken yarn signal, yarn ball balance below the threshold, or tension value exceeding the set range is detected, a fault alarm signal is immediately sent to the software management layer. The software management layer performs fault diagnosis based on real-time data and historical data and generates processing suggestions. At the same time, it executes corresponding protection measures according to the fault level, including emergency shutdown protection or linkage with the extrusion host to reduce speed.

[0024] Step S7: The visual interaction layer displays the tension value of each yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time, and uses different colors to indicate the parameter status. The system synchronously records key parameter data in the production process and stores it in the database, supporting historical data query, trend curve analysis, and automatic generation of daily production reports and quality analysis reports, thereby realizing intelligent monitoring and management of the pultrusion yarn loading process.

[0025] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: The intelligent control system and method for yarn feeding in pultrusion process based on industrial visualization and PLC closed-loop control provided by the present invention include a hardware layer, a PLC control layer, an intelligent algorithm software layer and an industrial-grade visualization interaction layer connected in sequence. Each layer works together to realize intelligent control of the entire yarn feeding process. The PLC communicates with the pultrusion host in real time and automatically matches the yarn feeding speed according to the traction speed. The linkage response time is <100ms, which completely eliminates the problem of yarn slack or breakage caused by speed loss. Employing a high-precision tension sensor in conjunction with a dual actuator consisting of a servo motor and an electromagnetic brake, and utilizing PID closed-loop control, tension fluctuations are kept within ±3%. Through multi-sensor fusion detection and intelligent diagnostic algorithms, early warnings and tiered handling of faults such as yarn breakage and insufficient allowance are achieved, reducing downtime and significantly improving production efficiency. An industrial touchscreen and host computer software provide real-time graphical and color-coded displays of multi-yarn status, allowing operators to intuitively grasp the overall production situation and shortening anomaly identification time. Furthermore, complete production data recording and in-depth analysis provide data support for process parameter optimization, quality improvement, and equipment maintenance, driving the digital and intelligent transformation of pultrusion production. Attached Figure Description

[0026] Figure 1 This is a block diagram of an intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control, provided as an embodiment of the present invention.

[0027] Figure 2 The flowchart illustrates an intelligent control method for yarn feeding in a pultrusion process based on industrial visualization and PLC closed-loop control, as provided in this embodiment of the invention. Detailed Implementation

[0028] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0029] like Figure 1 As shown, this embodiment of the invention provides an intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control, including a hardware layer, a PLC closed-loop control layer, a software management layer, and a visualization interaction layer that are sequentially connected in communication.

[0030] The hardware layer is used to collect physical parameter data during the yarn feeding process, including yarn tension value, yarn release speed, yarn spool state, and yarn arrangement image, and to execute control commands to adjust the yarn release speed and tension.

[0031] The PLC closed-loop control layer adopts a modular PLC controller, which communicates with the hardware layer through a high-speed I / O module. It receives physical parameter data collected and transmitted by the hardware layer in real time, and sends control commands to the hardware layer and real-time fault signals to the software management layer based on the analysis results of the physical parameter data.

[0032] In this embodiment, the PLC closed-loop control layer adopts a modular PLC controller, wherein the PLC controller is a Siemens S7-1500 series PLC, equipped with a high-speed I / O module and an Ethernet communication module, and supports PID control function.

[0033] The PLC programming software uses TIAPortalV16 to write programs for data acquisition, PID control, and linkage control; the host computer software is developed based on C# and LabVIEW, integrating functions such as parameter setting, data storage, image recognition, and report generation.

[0034] The software management layer, developed based on an industrial control software platform, is used to implement hierarchical permission management for administrators and operators, set, store and call preset process parameter libraries for different product models, parse, filter and store real-time data uploaded by the PLC closed-loop control layer, and perform fault diagnosis and generate fault handling suggestions based on historical data and real-time fault signals.

[0035] The visualization and interaction layer, through an industrial touch screen and host computer monitoring software, is used for graphical real-time monitoring of yarn loading process parameters, multi-state fault color-coded warnings, process parameter configuration and distribution, historical data trend analysis, and automatic generation of intelligent reports.

[0036] In this embodiment, the hardware layer includes a tension detection module, a speed detection module, a yarn spool state detection module, a control execution module, and an image acquisition module.

[0037] The tension detection module employs a high-precision tension sensor with a measurement accuracy of ±0.5%FS, installed on the unwinding path of each yarn, for real-time acquisition of the tension value of a single yarn. In this embodiment, the tension sensor selected is the HBMU9B series tension sensor, with a measurement range of 0-50N and an accuracy of ±0.2%FS.

[0038] The speed detection module uses an encoder coaxially connected to the yarn feeding roller to collect the feeding speed and cumulative feeding length.

[0039] The yarn bundle status detection module includes an infrared sensor and a weight sensor, which are used to detect the yarn bundle breakage status and the remaining weight of the yarn bundle, respectively, and trigger a yarn replenishment reminder when the remaining weight is lower than the threshold set in the preset process parameter library.

[0040] The control execution module includes a servo motor with a control accuracy of ±0.1 rpm and an electromagnetic brake. The servo motor drives the wire feeding roller to adjust the wire feeding speed, and the electromagnetic brake adjusts the braking torque based on tension feedback to regulate the tension closed-loop control. In this embodiment, a Panasonic MSMF series servo motor with a rated speed of 3000 rpm and a control accuracy of ±0.1 rpm is selected.

[0041] The image acquisition module uses an industrial camera with a resolution ≥1920×1080 and a frame rate ≥30fps, which is installed in the yarn confluence area to acquire real-time images of the yarn arrangement. In this embodiment, the industrial camera selected is a Hikvision MV-CA050-10GC industrial camera with a resolution of 2448×2048 and a frame rate of 30fps.

[0042] In this embodiment, the PLC closed-loop control layer includes a data receiving module, a closed-loop control module, a linkage control module, and a fault early warning and processing module.

[0043] The data receiving module receives the tension value, unwinding speed, yarn weight, yarn breakage signal, and yarn arrangement image data of the hardware layer in real time at a sampling frequency of ≥100Hz.

[0044] The closed-loop control module adjusts the servo motor speed and electromagnetic brake torque through a configurable PID algorithm based on the preset tension threshold in the preset process parameter library and the real-time collected tension value data, so that the yarn tension is stabilized within the set range of the preset process parameter library and the fluctuation range is ≤±3%.

[0045] The linkage control module automatically matches the yarn feeding speed with the traction speed of the pultrusion host to synchronize the yarn feeding amount with the traction speed.

[0046] The fault early warning processing module performs abnormal detection of physical parameter data. When it detects a broken yarn signal, insufficient yarn ball balance, or tension value exceeding a set threshold, it sends a real-time fault alarm signal to the software management layer and executes preset protection actions.

[0047] Through PLC closed-loop control and PID algorithm, the yarn tension fluctuation range is ≤±3%, and the synchronization accuracy of the yarn feeding speed and traction speed is ≤±0.5%, which effectively solves the problems of unstable tension and disordered yarn arrangement in traditional manual control, and improves the product mechanical performance qualification rate by more than 15%.

[0048] In this embodiment, the software management layer includes a preset parameter module, a data processing and storage module, an image recognition module, a fault diagnosis module, and a permission management module; The preset parameter module is used to preset process parameter libraries for different product models, set, store and call tension target values, yarn feeding speed limits, yarn spool balance thresholds and alarm thresholds; it supports yarn feeding control of different specifications and materials, the process parameter library can be flexibly expanded, adapt to the production needs of various pultrusion products, and has a wide range of applications.

[0049] The data processing and storage module parses and filters the real-time data transmitted by the PLC closed-loop control layer, and stores key parameter data with a storage period of ≥90 days, exporting it in Excel and CSV formats.

[0050] The image recognition module analyzes the yarn arrangement image using an image recognition algorithm to determine the yarn overlap and offset status. When the deviation exceeds a preset threshold in the process parameter library, it automatically adjusts the position of the pay-off roller or sends a manual adjustment reminder. The image recognition algorithm uses the OpenCV open-source library, employing contour detection and position matching algorithms to identify the yarn arrangement offset with an accuracy ≤0.1mm.

[0051] The fault diagnosis module performs fault diagnosis based on historical data and real-time fault signals, generates fault handling suggestions, and sends them to the visual interaction layer.

[0052] The permission management module is used for hierarchical permission management for administrators and operators.

[0053] In this embodiment, the visualization interaction layer includes a graphics display module, a parameter input module, a data query module, an alarm display module, and a report generation module.

[0054] The graphic display module displays the tension value of a single yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time in a graphical manner, using green / yellow / red to indicate the normal / warning / alarm status of the parameters.

[0055] The parameter input module is used to manually input or select preset parameters from the process parameter library and send them to the PLC control layer in real time.

[0056] The data query module displays the trends of tension and unwinding speed in the form of curves, as well as historical data query and comparison; key parameters and fault records in the production process are stored for a long time, supporting query and analysis, providing data support for process optimization, and helping to continuously improve product quality.

[0057] The alarm display module displays the fault type, occurrence time, and handling suggestions in real time through audible and visual alarms and text prompts.

[0058] The report generation module automatically generates daily production reports and quality analysis reports that include information such as production duration, product model, parameter fluctuation range, and number of failures.

[0059] The industrial touchscreen used is the Weintek MT8150iE touchscreen, 15 inches, with a resolution of 1920×1080.

[0060] By using industrial cameras and real-time data display, the entire yarn loading process can be monitored visually. Operators can intuitively grasp the yarn tension, yarn spool status, and arrangement, reducing fault detection time to within 1 second and minimizing the risk of escalation. The system enables automatic parameter adjustment, automatic fault alarms and protection, automatic data storage and report generation, reducing the need for one dedicated operator per production line and shortening changeover and debugging time by more than 60%.

[0061] In this embodiment, the tension closed-loop control of the control execution module adopts the PID algorithm. Its proportional coefficient, integral coefficient, and derivative coefficient are configured online through the intelligent algorithm software layer and are adaptively adjusted according to the tension fluctuation to achieve stable control with a tension fluctuation range of ≤±3%.

[0062] The image recognition module uses the OpenCV open-source library for image recognition algorithm. Through contour detection and position matching algorithms, it identifies the yarn arrangement offset with an accuracy of ≤0.1mm. It calculates the yarn spacing and center line position in real time. When the yarn offset exceeds 2mm or there is overlap, it triggers automatic position adjustment or manual intervention reminder.

[0063] The linkage control module receives the traction line speed signal v of the pultrusion host in real time through the communication interface. t According to the speed linkage formula v f =k*v t Automatically calculate the target laying speed v fWhere k is the linkage matching coefficient; the encoder provides real-time feedback of the actual wire feeding speed, compares it with the target speed to form a deviation, and corrects the servo output through closed-loop adjustment to achieve high-precision synchronous matching between the wire feeding speed and the traction speed.

[0064] The fault early warning and processing module adopts a graded response mechanism: when a broken yarn or severe tension loss is detected, an emergency shutdown protection is immediately executed; when insufficient yarn stock or slight tension fluctuation is detected, an early warning prompt is triggered and the pultrusion host is linked to reduce its speed.

[0065] In this embodiment, taking the production of glass fiber pultruded profiles as an example, the target tension value is set to 15N, the allowable tension fluctuation range is 14.5-15.5N, the traction speed of the pultrusion host is 0.5m / min, the PLC adjusts the servo motor speed to 30rpm through the PID algorithm, and the braking torque of the electromagnetic brake is 2N·m, so as to achieve stable control of yarn tension, the yarn arrangement offset is ≤0.2mm, there is no broken or piled-up phenomenon during the production process, the product wall thickness uniformity error is ≤±0.3mm, and the scrap rate is reduced by 10% compared with the traditional manual yarn loading process.

[0066] In a second aspect, the present invention provides a control method for an intelligent control system for yarn feeding in a pultrusion process based on industrial visualization and PLC closed-loop control as described in any of the first aspects, comprising the following steps: Step S1: After the system is powered on, the visual interaction layer starts the industrial touch screen and the host computer monitoring software. The operator logs into the system and selects the corresponding product model's process parameters from the preset process parameter library in the software management layer according to the current production task.

[0067] The process parameters include target tension value, pay-off speed limit, yarn spool balance threshold, and alarm threshold.

[0068] The software management layer sends the parameters to the PLC closed-loop control layer and completes the initialization configuration. At the same time, it performs communication detection and status self-check on the tension detection module, speed detection module, yarn spool status detection module, and image acquisition module to ensure that each hardware module is in normal working condition.

[0069] Step S2: During the yarn feeding process, the hardware layer collects the tension value of each yarn in real time through the tension detection module, collects the yarn feeding speed and cumulative yarn feeding length in real time through the encoder, detects the yarn breakage signal through the infrared sensor and detects the remaining weight of the yarn bundle through the weight sensor. At the same time, the image acquisition module continuously collects the yarn arrangement image in the yarn confluence area. The above-mentioned data are transmitted to the PLC closed-loop control layer in real time at the set sampling frequency for unified processing.

[0070] Step S3: The PLC closed-loop control layer receives data such as tension value, yarn feeding speed, yarn weight, yarn breakage signal and yarn arrangement image transmitted from the hardware layer through the high-speed I / O module, and performs time synchronization and preliminary filtering processing on the received data.

[0071] The processed data is used for closed-loop control calculations on one hand, and uploaded in real time to the software management layer for further data analysis, storage, and historical record management on the other hand.

[0072] Step S4: The PLC closed-loop control layer calculates the tension deviation based on the target tension value in the preset process parameter library and the real-time acquired tension value. It then adjusts the servo motor speed and the electromagnetic brake torque using a PID control algorithm to achieve closed-loop stable control of the yarn tension. Simultaneously, the linkage control module automatically calculates the target unwinding speed based on the pultrusion host traction speed signal and forms a speed closed-loop adjustment through encoder feedback, ensuring that the unwinding speed and traction speed remain synchronized, thereby guaranteeing stable yarn feeding.

[0073] Step S5: The software management layer calls the image recognition module to analyze the yarn arrangement image captured by the industrial camera. It calculates the yarn spacing and center line offset through contour detection and position matching algorithms to determine whether there is yarn overlap or arrangement misalignment. When the yarn offset exceeds the allowable range in the preset process parameter library, the system automatically adjusts the position of the pay-off roller or issues a manual adjustment prompt to the operator to ensure that the yarn arrangement is neat and stable.

[0074] Step S6: The fault early warning processing module of the PLC closed-loop control layer performs real-time anomaly detection on the collected physical parameters. When a broken yarn signal, yarn ball balance below the threshold, or tension value exceeding the set range is detected, a fault alarm signal is immediately sent to the software management layer. The software management layer performs fault diagnosis based on real-time data and historical data and generates processing suggestions. At the same time, it executes corresponding protection measures according to the fault level, including emergency shutdown protection or linkage with the extrusion host to reduce speed.

[0075] Step S7: The visual interaction layer displays the tension value of each yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time, and uses different colors to indicate the parameter status. The system synchronously records key parameter data in the production process and stores it in the database, supporting historical data query, trend curve analysis, and automatic generation of daily production reports and quality analysis reports, thereby realizing intelligent monitoring and management of the pultrusion yarn loading process.

[0076] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control, characterized in that, It includes a hardware layer, a PLC closed-loop control layer, a software management layer, and a visual interaction layer that are connected in sequence. The hardware layer is used to collect physical parameter data during the yarn feeding process, including yarn tension value, yarn feeding speed, yarn spool state, and yarn arrangement image, and to execute control commands to adjust the yarn feeding speed and tension. The PLC closed-loop control layer adopts a modular PLC controller, which communicates with the hardware layer through a high-speed I / O module. It receives physical parameter data collected and transmitted by the hardware layer in real time, and sends control commands to the hardware layer and real-time fault signals to the software management layer based on the analysis results of the physical parameter data. The software management layer, developed based on an industrial control software platform, is used to implement hierarchical permission management for administrators and operators, set, store and call preset process parameter libraries for different product models, parse, filter and store real-time data uploaded by the PLC closed-loop control layer, and perform fault diagnosis and generate fault handling suggestions based on historical data and real-time fault signals. The visualization and interaction layer, through an industrial touch screen and host computer monitoring software, is used for graphical real-time monitoring of yarn loading process parameters, multi-state fault color-coded warnings, process parameter configuration and distribution, historical data trend analysis, and automatic generation of intelligent reports.

2. The intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control according to claim 1, characterized in that, The hardware layer includes a tension detection module, a speed detection module, a yarn bundle state detection module, a control execution module, and an image acquisition module; The tension detection module uses a high-precision tension sensor with a measurement accuracy of ±0.5%FS, which is installed on the yarn feeding path of each yarn to collect the tension value of a single yarn in real time. The speed detection module uses an encoder coaxially connected to the yarn pay-off roller to collect the pay-off speed and cumulative pay-off length. The yarn bundle status detection module includes an infrared sensor and a weight sensor, which are used to detect the yarn bundle breakage status and the remaining weight of the yarn bundle, respectively, and trigger a yarn replenishment reminder when the remaining weight is lower than the threshold set in the preset process parameter library. The control execution module includes a servo motor with a control accuracy of ±0.1 rpm and an electromagnetic brake. The servo motor is used to drive the wire feeding roller to adjust the wire feeding speed, and the electromagnetic brake is used to adjust the braking torque according to the tension feedback to regulate the tension closed-loop control. The image acquisition module uses an industrial camera with a resolution of ≥1920×1080 and a frame rate of ≥30fps, which is installed in the yarn confluence area to acquire images of the yarn arrangement in real time.

3. The intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control according to claim 1, characterized in that, The PLC closed-loop control layer includes a data receiving module, a closed-loop control module, a linkage control module, and a fault early warning and processing module. The data receiving module receives the tension value, unwinding speed, yarn weight, yarn breakage signal, and yarn arrangement image data of the hardware layer in real time at a sampling frequency of ≥100Hz. The closed-loop control module, based on the preset tension threshold in the preset process parameter library and the real-time collected tension data, adjusts the servo motor speed and the electromagnetic brake torque through a configurable PID algorithm, so that the yarn tension is stabilized within the set range of the preset process parameter library and the fluctuation range is ≤±3%. The linkage control module automatically matches the yarn feeding speed according to the traction speed of the pultrusion host to synchronize the yarn feeding amount with the traction speed; The fault early warning processing module performs abnormal detection of physical parameter data. When it detects a broken yarn signal, insufficient yarn ball balance, or tension value exceeding a set threshold, it sends a real-time fault alarm signal to the software management layer and executes preset protection actions.

4. The intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control according to claim 1, characterized in that, The software management layer includes a preset parameter module, a data processing and storage module, an image recognition module, a fault diagnosis module, and a permission management module; The preset parameter module is used to preset process parameter libraries for different product models, and to set, store and call tension target values, yarn feeding speed limits, yarn spool balance thresholds and alarm thresholds. The data processing and storage module parses and filters the real-time data transmitted by the PLC closed-loop control layer, and stores key parameter data with a storage period of ≥90 days, which is exported in Excel and CSV formats. The image recognition module analyzes the yarn arrangement image through image recognition algorithm to determine the yarn overlap and offset status, and automatically adjusts the position of the pay-off roller or sends a manual adjustment reminder when the deviation exceeds the set threshold of the preset process parameter library. The fault diagnosis module performs fault diagnosis based on historical data and real-time fault signals, generates fault handling suggestions, and sends them to the visual interaction layer. The permission management module is used for hierarchical permission management for administrators and operators.

5. The intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control according to claim 1, characterized in that, The visualization interaction layer includes a graphics display module, a parameter input module, a data query module, an alarm display module, and a report generation module; The graphic display module displays the tension value of a single yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time in a graphical manner, using green / yellow / red to indicate the normal / warning / alarm status of the parameters; The parameter input module is used to manually input or select preset parameters from the process parameter library and send them to the PLC control layer in real time. The data query module displays the tension change trend and the wire release speed change trend in the form of curves, as well as historical data query and comparison. The alarm display module displays the fault type, occurrence time, and handling suggestions in real time through audible and visual alarms and text prompts. The report generation module automatically generates daily production reports and quality analysis reports that include information such as production duration, product model, parameter fluctuation range, and number of failures.

6. The intelligent control system for pultrusion yarn feeding based on industrial visualization and PLC closed-loop control according to claim 1, characterized in that, The tension closed-loop control of the control execution module adopts the PID algorithm. Its proportional coefficient, integral coefficient, and derivative coefficient are configured online through the intelligent algorithm software layer and are adaptively adjusted according to the tension fluctuation to achieve stable control with a tension fluctuation range of ≤±3%. The image recognition module uses the OpenCV open-source library for image recognition algorithm. Through contour detection and position matching algorithms, it identifies the yarn arrangement offset with an accuracy of ≤0.1mm. It calculates the yarn spacing and center line position in real time. When the yarn offset exceeds 2mm or there is overlap, it triggers automatic position adjustment or manual intervention reminder. The linkage control module receives the traction line speed signal v of the extrusion host in real time via a communication interface. t According to the speed linkage formula v f =k*v t Automatically calculate the target laying speed v f , where k is the linkage matching coefficient; the encoder provides real-time feedback of the actual wire feeding speed, compares it with the target speed to form a deviation, and corrects the servo output through closed-loop adjustment to achieve high-precision synchronous matching between the wire feeding speed and the traction speed; The fault early warning and processing module adopts a graded response mechanism: when a broken yarn or severe tension loss is detected, an emergency shutdown protection is immediately executed; when insufficient yarn stock or slight tension fluctuation is detected, an early warning prompt is triggered and the pultrusion host is linked to reduce its speed.

7. A control method for an intelligent control system for yarn feeding in a pultrusion process based on industrial visualization and PLC closed-loop control as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step S1: After the system is powered on, the visual interaction layer starts the industrial touch screen and the host computer monitoring software. The operator logs into the system and selects the corresponding product model's process parameters from the preset process parameter library in the software management layer according to the current production task. The process parameters include target tension value, pay-off speed limit, yarn spool balance threshold, and alarm threshold. The software management layer sends the parameters to the PLC closed-loop control layer and completes the initialization configuration. At the same time, it performs communication detection and status self-check on the tension detection module, speed detection module, yarn status detection module and image acquisition module to ensure that each hardware module is in normal working condition. Step S2: During the yarn feeding process, the hardware layer collects the tension value of each yarn in real time through the tension detection module, collects the yarn feeding speed and cumulative yarn feeding length in real time through the encoder, detects the yarn breakage signal through the infrared sensor and detects the remaining weight of the yarn bundle through the weight sensor. At the same time, the image acquisition module continuously collects the yarn arrangement image in the yarn confluence area. The above collected data is transmitted to the PLC closed-loop control layer in real time at the set sampling frequency for unified processing. Step S3: The PLC closed-loop control layer receives data such as tension value, yarn feeding speed, yarn weight, yarn breakage signal and yarn arrangement image transmitted from the hardware layer through the high-speed I / O module, and performs time synchronization and preliminary filtering processing on the received data. The processed data is used for closed-loop control calculations on one hand, and uploaded in real time to the software management layer for further data analysis, storage and historical record management on the other hand. Step S4: The PLC closed-loop control layer calculates the tension deviation based on the target tension value in the preset process parameter library and the real-time acquired tension value. It then adjusts the servo motor speed and the electromagnetic brake torque using a PID control algorithm to achieve closed-loop stable control of the yarn tension. Simultaneously, the linkage control module automatically calculates the target unwinding speed based on the pultrusion host traction speed signal and forms a speed closed-loop adjustment through encoder feedback, ensuring that the unwinding speed and traction speed remain synchronized, thereby guaranteeing stable yarn feeding. Step S5: The software management layer calls the image recognition module to analyze the yarn arrangement image captured by the industrial camera. It calculates the yarn spacing and center line offset through contour detection and position matching algorithms to determine whether there is yarn overlap or arrangement misalignment. When the yarn offset exceeds the allowable range in the preset process parameter library, the system automatically adjusts the position of the pay-off roller or issues a manual adjustment prompt to the operator to ensure that the yarn arrangement is neat and stable. Step S6: The fault early warning processing module of the PLC closed-loop control layer performs real-time anomaly detection on the collected physical parameters. When a broken yarn signal, yarn ball balance below the threshold, or tension value exceeding the set range is detected, a fault alarm signal is immediately sent to the software management layer. The software management layer performs fault diagnosis based on real-time data and historical data and generates processing suggestions. At the same time, it executes corresponding protection measures according to the fault level, including emergency shutdown protection or linkage with the pultrusion host to reduce speed. Step S7: The visual interaction layer displays the tension value of each yarn, the unwinding speed, the remaining weight of the yarn bundle, and the yarn arrangement image in real time, and uses different colors to indicate the parameter status. The system synchronously records key parameter data in the production process and stores it in the database, supporting historical data query, trend curve analysis, and automatic generation of daily production reports and quality analysis reports, thereby realizing intelligent monitoring and management of the pultrusion yarn loading process.

Images