Braiding control method and system based on intelligent learning and multi-sensor fusion

Abstract

The application discloses a kind of based on intelligent learning and the weaving tape weaving control method and system of multi-sensor fusion, method includes using the preset multiple sensor units to construct weaving data sensing mechanism;Extract model training number, and construct parameter identification control model;According to the target weaving process parameter of weaving tape weaving request instruction acquisition;Determine the target weaving material trace of composite weaving material;According to target weaving material trace, generate weaving tape identification unit;With each weaving tape identification unit as basic unit, the distance between target weaving material trace and current weaving material trace is calculated, obtain several trace deviation information;When identifying that certain trace deviation information meets the preset trace deviation condition, obtain the adjustment parameter information of weaving tape weaving equipment;Parameter adjustment is carried out to weaving tape weaving equipment based on adjustment parameter information;The application has reduced the quality control difficulty of weaving control system, realizes the effect of accurate weaving control to composite weaving tape weaving.

Description

Technical Field

[0001] This application relates to the field of ribbon weaving technology, and in particular to a ribbon weaving control method and system based on intelligent learning and multi-sensor fusion. Background Technology

[0002] In the field of textile machinery automation control and intelligent manufacturing, composite webbing weaving technology has always been a research hotspot, with enterprises placing higher demands on the flexibility and intelligence level of production lines. In recent years, with the continuous advancement of sensor technology, composite webbing weaving control methods have seen significant development; however, despite the achievements in related technologies, many challenges and problems remain to be solved.

[0003] First, traditional composite webbing weaving control methods often rely on fixed control parameters, which lack the ability to self-optimize and adjust during production. Second, existing weaving control systems have significant shortcomings in data collection and analysis. Although traditional technologies can achieve a certain degree of automated production, the lack of comprehensive data support makes precise control a challenging task. This not only limits the transparency of the production process but also increases the difficulty of quality control, thus requiring urgent improvement. Summary of the Invention

[0004] To reduce the difficulty of quality control in weaving control systems and achieve precise weaving control of composite webbing, this application provides a webbing weaving control method and system based on intelligent learning and multi-sensor fusion.

[0005] Firstly, the objective of this invention is achieved through the following technical solution:

[0006] A webbing weaving control method based on intelligent learning and multi-sensor fusion includes:

[0007] A weaving data sensing mechanism is constructed using multiple sensor units pre-installed on the weaving equipment; the weaving data sensing mechanism is used to collect weaving data of the composite weaving during the weaving process; model training data of the historical weaving dataset of the weaving data sensing mechanism is extracted; a parameter recognition and control model corresponding to the weaving equipment is constructed through a deep learning model and the model training data;

[0008] Identify the webbing weaving request instruction, and obtain the target weaving process parameters according to the webbing weaving request instruction; in the parameter identification and control model, determine the target weaving trace line of the composite weaving fabric based on the target weaving process parameters;

[0009] Several webbing identification units are generated during the composite webbing weaving process based on the target woven fabric trace and the target weaving process parameters.

[0010] Using each webbing identification unit as a basic unit, the current weaving sensing data set is acquired, and the current weaving trace line is determined; the distance between the target weaving trace line of the webbing identification unit and the corresponding current weaving trace line is calculated to obtain several trace line deviation information.

[0011] When a trace deviation is detected that matches a preset trace deviation condition, a sensor information collection command is triggered.

[0012] According to the sensor information collection instruction, the adjustment parameter information of the ribbon weaving equipment is obtained in the parameter identification and control model; the parameters of the ribbon weaving equipment are adjusted based on the adjustment parameter information.

[0013] By adopting the above technical solution, multiple sensor units, such as tension sensors, vision sensors, and yarn metering sensors, are pre-installed on the webbing equipment to detect the webbing tension, yarn length, weaving pattern, and size during the composite webbing weaving process. This facilitates the transparency of control during the production process of composite woven webbing and provides data support and reference for subsequent precise production of woven webbing and adjustment of equipment parameters. The parameter identification control model of this application is obtained through deep learning and multiple training sessions based on a deep learning model and model training data selected from historical weaving datasets. The model training data consists of high-quality weaving data selected based on the composite target weaving process requirements during historical weaving processes, which helps improve the accuracy of the model calculation results of the parameter identification control model and indirectly improves the control efficiency of composite webbing weaving control.

[0014] Specifically, the parameter recognition control model can identify and calibrate data for abnormal situations during the weaving process of the composite webbing. When a webbing weaving request instruction is detected, the webbing weaving equipment begins weaving. At this time, the weaving data sensing mechanism acquires the current weaving sensing data and obtains the corresponding target weaving process parameters through the parameter recognition control model. The target weaving process parameters are the weaving data of the finished composite webbing that meets the relevant process requirements. To judge the weaving effect of the webbing weaving equipment, a target woven trace line is set for preliminary comparison of the weaving effect. At the same time, to facilitate the monitoring of the finished quality of the composite webbing, this application generates several webbing recognition units during the weaving process for detecting and evaluating the weaving quality of the composite webbing. The target woven trace line of each webbing recognition unit is calculated to be the same as the corresponding currently detected woven trace line. The distance is used to obtain several corresponding trace line deviation information. The trace line deviation information is compared with the preset trace line deviation conditions. When a trace line deviation information is found to meet the trace line deviation conditions, it indicates that the current woven fabric trace line does not meet the process requirements, and the webbing weaving equipment may have a weaving failure. This triggers a sensor information collection command to prompt relevant personnel to take corresponding adjustment measures in a timely manner. At the same time, adjustment parameter information can also be obtained to adjust the control parameters of the weaving equipment in a timely manner when the woven fabric trace line does not meet the process requirements, thereby improving the weaving control accuracy of the webbing weaving equipment. This application uses intelligent learning and multi-sensor unit fusion to calculate and obtain corresponding adjustment parameter information when the webbing weaving equipment has a weaving abnormality, thereby reducing the quality control difficulty of the weaving control system and achieving the goal of precise weaving control of composite webbing.

[0015] In a preferred embodiment of this application: the identification of the webbing weaving request instruction, and the acquisition of target weaving process parameters based on the webbing weaving request instruction, include:

[0016] The webbing weaving equipment identifies webbing request commands and obtains weaving numerical information based on the webbing weaving request commands; the weaving numerical information includes weaving speed, tension force, fabric heating temperature, and webbing leveling time;

[0017] Obtain yarn attribute selection parameters; in the parameter recognition and control model, obtain target weaving process parameters based on the yarn attribute selection parameters and equipment weaving numerical information.

[0018] Based on the target weaving process parameters and the equipment weaving data information, a webbing weaving command is triggered.

[0019] By adopting the above technical solution, when the ribbon weaving equipment receives a ribbon weaving request instruction from the user or automatically triggered, it acquires equipment weaving data information that meets the relevant process requirements. In order to improve the weaving quality and yield rate of the composite ribbon after weaving, parameters are selected based on the equipment weaving data information and the selected yarn attributes. The yarn attribute selection parameters include information such as yarn type, yarn color, and yarn size. Then, the equipment weaving data information is corrected and optimized to obtain the target weaving process parameters. The target weaving process parameters are detected and verified on the control parameters and control process of the ribbon weaving equipment before the composite ribbon weaving begins, which is conducive to achieving the effect of precise weaving control of composite ribbon weaving.

[0020] In a preferred embodiment of this application: the target weaving process parameters include weaving speed and webbing attribute parameters; the sensor unit includes a visual sensing unit; the generation of several webbing recognition units during the composite webbing weaving process based on the target woven material trace and the target weaving process parameters includes:

[0021] Based on the distance change information between two adjacent traces of the target woven fabric trace during the preset weaving process, the target trace change information is obtained;

[0022] Based on the process parameter change information of the target weaving process parameters during the preset weaving process, the change value of the target weaving working parameter is compared with a preset change threshold. When the change value of the target weaving working parameter is greater than the preset change threshold, the target weaving parameter change information is output. The target weaving parameter change information includes weaving speed change information and webbing attribute change information.

[0023] Based on the target woven trace, within the weaving time of the composite webbing weaving process, the visual sensing unit is triggered to acquire the target trace image based on a fixed time interval to obtain the first trace image.

[0024] or,

[0025] When the weaving speed change information or the webbing attribute change information is obtained, the visual sensing unit is triggered to acquire the target trace line image to obtain the second trace line image.

[0026] The first trace line image and the second trace line image are corrected based on the target woven fabric trace line to form a webbing recognition unit.

[0027] By adopting the above technical solutions, composite woven webbing typically exhibits characteristics such as high tensile strength, crisp fabric surface, and difficulty in controlling woven deformation. When judging the weaving effect of composite woven webbing, the detection and measurement method of the weaving trace is often used to measure its woven performance. However, to improve the efficiency of judging the weaving effect of composite woven webbing, this application utilizes a parameter recognition and control model based on machine intelligence learning. After obtaining the target weaving process parameters, it predicts the entire weaving process using historical weaving processes and data, and extracts information on changes in the target trace and target weaving traces exceeding a change threshold throughout the entire weaving process. The system first extracts information on weaving parameter changes; then, based on the entire weaving time of the composite webbing process, several first trace line images are extracted at fixed time intervals; and second trace line images are obtained when significant changes occur in weaving speed and webbing attributes (webbing attributes include the number of webbings involved in weaving, weaving angle, and the type of yarn involved in weaving). After image processing and correction, including image feature extraction, image binarization, image denoising, and region normalization, a webbing recognition unit with higher trace line recognition efficiency is obtained. This improves the efficiency and accuracy of trace line recognition. By dividing the system into several webbing recognition units, the quality control difficulty of the weaving control system is reduced, and the accuracy of evaluating the weaving effect of the webbing weaving equipment is improved.

[0028] In a preferred embodiment of this application, the webbing weaving equipment includes a spring pressing mechanism for pressing and straightening the woven webbing;

[0029] After identifying and obtaining a webbing weaving request instruction through the webbing weaving device, and obtaining the device weaving numerical information based on the webbing weaving request instruction, the method further includes:

[0030] Based on the target weaving process parameters and the equipment weaving data information, the pressure threshold range corresponding to the spring pressure mechanism is obtained;

[0031] The pressing force data of the spring pressing mechanism is acquired in real time, and it is determined whether the pressing force data is within the corresponding pressing force threshold range to obtain the pressing force judgment result.

[0032] When a certain pressure data is not within the corresponding pressure threshold range, an abnormal prompt message is generated and sent to the preset user control terminal.

[0033] By adopting the above technical solution, after the composite braided webbing is initially woven, it still needs to undergo a pressing and straightening process to improve the flatness of the composite webbing. The spring pressing and straightening mechanism is used to press and straighten the initially woven composite braided webbing. In order to monitor the pressing and straightening effect of the spring pressing and straightening mechanism on the composite braided webbing, this application compares the pressing force data of the spring pressing and straightening mechanism acquired in real time with the pressing force threshold range that meets the pressing and straightening process requirements, and judges whether the pressing force data is within the corresponding pressing force threshold range to obtain the pressing force judgment result. When a certain pressing force data is not within the corresponding pressing force threshold range, it indicates that the pressing force of the spring pressing and straightening mechanism is too large or too small, and the flatness of the pressed composite webbing is poor. It is necessary to adjust the pressing force of the spring pressing and straightening mechanism in time so that its pressing force data returns to the corresponding pressing force threshold range. At this time, an abnormal prompt message is generated to prompt the user that the spring pressing and straightening mechanism has an abnormality and needs to take handling measures. This facilitates the weaving control of each production step of the webbing weaving equipment, which is conducive to quality control and improves the weaving effect of composite webbing.

[0034] In a preferred embodiment of this application: the spring-pressing mechanism includes a pressing elastic element and an adjusting mechanism for adjusting the pressing force of the pressing elastic element; after generating an abnormal prompt message and sending it to a preset user control terminal when a certain pressing force data is not within the corresponding pressing force threshold range, the method further includes:

[0035] Obtain the current clamping force data of the spring clamping mechanism, and calculate the minimum clamping force difference required for the current clamping force data to reach the corresponding clamping force threshold range;

[0036] The minimum pressure difference is compared with a preset critical difference and a preset limit difference, and the preset critical difference is less than the preset limit difference.

[0037] When the minimum compression force difference is less than the critical difference, an automatic compression command is sent to the regulating mechanism;

[0038] When the minimum pressure difference is greater than or equal to the critical difference, a text box for parameter input is sent to the user control terminal, and the pressure adjustment parameter input in the text box is obtained. Based on the pressure adjustment parameter, a compression command is sent to the adjustment mechanism.

[0039] When the minimum pressure difference is greater than or equal to the limit difference, a replacement prompt message is sent to the user control terminal.

[0040] By adopting the above technical solution, the method for adjusting the spring-loaded adjusting mechanism is determined by comparing the minimum pressure difference with the critical difference and the limit difference, so as to improve the applicability of the spring-loaded adjusting mechanism and effectively ensure the flatness of the composite webbing. When the minimum pressure difference is less than the critical difference, it indicates that the pressure data of the spring-loaded adjusting mechanism has decreased slightly, and the elastic modulus of the clamping elastic element has decreased slightly. At this time, the clamping elastic element is automatically adjusted by the adjusting mechanism to complete the adjustment of the clamping force of the spring-loaded adjusting mechanism. When the minimum pressure difference is greater than or equal to the critical difference, it indicates that the spring-loaded adjusting mechanism has decreased slightly. If the compressive force data drops significantly, the elastic modulus of the clamping elastic element will also decrease significantly. Information prompts will be sent to the user control terminal. Staff will need to determine whether the clamping elastic element needs replacement based on actual usage requirements. If replacement is not necessary, adjustment parameters will be input to bring the compressive force data back into the corresponding compressive force threshold range. When the minimum compressive force difference is greater than or equal to the limit difference, it indicates a severe decrease in the elastic modulus of the clamping elastic element, resulting in a short service life and requiring timely replacement. Replacement prompts will be sent to the user to remind them to replace the clamping elastic element promptly.

[0041] In a preferred embodiment of this application: the plurality of sensor units further includes a yarn measuring unit for measuring the length of the yarns involved in weaving; the step of obtaining yarn attribute selection parameters, after obtaining the target weaving process parameters in the parameter identification and control model based on the yarn attribute selection parameters and the equipment weaving numerical information, further includes:

[0042] Obtain the total length of yarn in multiple spools of the ribbon weaving equipment, the historical yarn consumption length of the corresponding yarn metering unit, and the associated spool identifier for each spool;

[0043] In the parameter identification and control model, the length of the yarn to be woven corresponding to the webbing weaving request instruction is calculated based on the equipment weaving numerical information and the target weaving process parameters, and the corresponding yarn length data group to be consumed is obtained.

[0044] Based on the total yarn length of multiple bobbins, the historical yarn consumption length, and the corresponding yarn length to be consumed data set, the yarn consumption judgment result is determined.

[0045] During the composite weaving process, based on the yarn consumption judgment result, the bobbin identification and the corresponding weaving speed, the time node for outputting the yarn replacement prompt information is determined, and the yarn replacement prompt command is triggered at the time node.

[0046] By adopting the above technical solution, the yarn usage of multiple bobbins in the ribbon weaving equipment is monitored in real time. Based on the required yarn length and actual weaving speed of the composite ribbon being woven, the timing of yarn replacement prompts is calculated in advance. This timing is earlier than the yarn bobbin's yarn depletion time, thus allowing workers time to replace the yarn. This achieves automated and intelligent production of the ribbon weaving equipment and improves the control effect of the equipment.

[0047] Secondly, the objective of this invention is achieved through the following technical solution:

[0048] A webbing weaving control system based on intelligent learning and multi-sensor fusion, comprising:

[0049] A weaving data sensing mechanism is constructed using multiple sensor units pre-installed on the weaving equipment to collect weaving data of the composite weaving during the weaving process.

[0050] An intelligent control mechanism is used to extract model training data from the historical weaving dataset of the weaving data sensing mechanism; and to construct a parameter recognition and control model corresponding to the webbing weaving equipment through a deep learning model and the model training data.

[0051] The intelligent control mechanism also includes:

[0052] The request instruction recognition module is used to recognize the webbing weaving request instruction, obtain the target weaving process parameters according to the webbing weaving request instruction, and determine the target weaving trace line of the composite weaving fabric based on the target weaving process parameters in the parameter recognition control model.

[0053] The webbing identification area division module is used to generate several webbing identification units during the composite webbing weaving process based on the target woven fabric trace line and the target weaving process parameters.

[0054] The trace line deviation calculation module is used to acquire the current weaving sensing data group with each webbing identification unit as the basic unit, and determine the current weaving trace line; calculate the distance between the target weaving trace line of the webbing identification unit and the corresponding current weaving trace line to obtain several trace line deviation information.

[0055] The deviation information collection module is used to trigger a sensor information collection command when it is detected that a certain trace line deviation information meets the preset trace line deviation conditions.

[0056] The equipment parameter adjustment module is used to obtain adjustment parameter information of the webbing weaving equipment in the parameter identification and control model according to the sensor information collection instruction; and to adjust the parameters of the webbing weaving equipment based on the adjustment parameter information.

[0057] By adopting the above technical solution, the parameter recognition control model can identify and calibrate data for abnormal situations during the weaving process of the composite webbing. When a webbing weaving request instruction is detected, the webbing weaving equipment starts weaving. At this time, the weaving data sensing mechanism acquires the current weaving sensing data and obtains the corresponding target weaving process parameters through the parameter recognition control model. The target weaving process parameters are the weaving data of the finished composite webbing that meets the relevant process requirements. To judge the weaving effect of the weaving equipment, a target woven trace line is set for preliminary comparison of the weaving effect. At the same time, to facilitate the monitoring of the finished quality of the composite webbing, this application generates several webbing identification units during the weaving process for detecting and evaluating the weaving quality of the composite webbing. The target woven trace line of each webbing identification unit is calculated to match the corresponding actual detected current woven trace line. The distance between them is used to obtain several corresponding trace line deviation information. The trace line deviation information is compared with the preset trace line deviation conditions. When a trace line deviation information is found to meet the trace line deviation conditions, it indicates that the current woven fabric trace line does not meet the process requirements, and the webbing weaving equipment may have a weaving failure. This triggers a sensor information collection command to prompt relevant personnel to take corresponding adjustment measures in a timely manner. At the same time, adjustment parameter information can also be obtained to adjust the control parameters of the weaving equipment in a timely manner when the woven fabric trace line does not meet the process requirements, thereby improving the weaving control accuracy of the webbing weaving equipment. This application uses intelligent learning and multi-sensor unit fusion to calculate and obtain corresponding adjustment parameter information when the webbing weaving equipment has a weaving abnormality, thereby reducing the quality control difficulty of the weaving control system and achieving the goal of precise weaving control of composite webbing.

[0058] In a preferred embodiment of this application, the request instruction recognition module includes:

[0059] The equipment weaving value acquisition submodule is used to identify the weaving request command through the weaving equipment and acquire the equipment weaving value information based on the weaving request command; the equipment weaving value information includes weaving speed, tension force, fabric heating temperature, and weaving leveling time.

[0060] The target weaving parameter acquisition submodule is used to acquire yarn attribute selection parameters. In the parameter recognition and control model, the target weaving process parameters are acquired based on the yarn attribute selection parameters and the equipment weaving numerical information.

[0061] The weaving instruction triggering submodule is used to trigger a webbing weaving instruction based on the target weaving process parameters and the equipment weaving data information.

[0062] By adopting the above technical solution, when the ribbon weaving equipment receives a ribbon weaving request instruction from a user or automatically triggered instruction, it acquires equipment weaving data information that meets the relevant process requirements. To improve the weaving quality and yield rate of the composite ribbon after weaving, parameters are selected based on the equipment weaving data information and the selected yarn attributes. These yarn attribute selection parameters include information such as yarn type, yarn color, and yarn size. The equipment weaving data information is then corrected and optimized to obtain the target weaving process parameters. The target weaving process parameters are used to detect and verify the control parameters and control process of the ribbon weaving equipment before the composite ribbon weaving begins, which is beneficial to achieving precise weaving control of composite ribbon weaving.

[0063] In summary, this application includes at least one of the following beneficial technical effects:

[0064] 1. The parameter recognition control model can identify and calibrate data for abnormal situations during the weaving process of composite webbing. When a webbing weaving request instruction is detected, the webbing weaving equipment starts weaving. At this time, the weaving data sensing mechanism acquires the current weaving sensing data and obtains the corresponding target weaving process parameters through the parameter recognition control model. The target weaving process parameters are the weaving data of the finished composite webbing that meets the relevant process requirements. To judge the weaving effect of the webbing weaving equipment, a target woven trace line is set for preliminary comparison of the weaving effect. At the same time, to facilitate the monitoring of the finished quality of the composite webbing, this application generates several webbing recognition units during the weaving process for detecting and evaluating the weaving quality of the composite webbing. The distance between the target woven trace line of each webbing recognition unit and the corresponding actual detected current woven trace line is calculated. The system obtains several corresponding trace line deviation information, compares the trace line deviation information with preset trace line deviation conditions, and when a trace line deviation information is found to meet the trace line deviation conditions, it indicates that the current woven fabric trace line does not meet the process requirements, and the webbing weaving equipment may have a weaving failure. This triggers a sensor information collection command to prompt relevant personnel to take corresponding adjustment measures in a timely manner. At the same time, it can also obtain adjustment parameter information to adjust the control parameters of the weaving equipment in a timely manner when the woven fabric trace line does not meet the process requirements, thereby improving the weaving control accuracy of the webbing weaving equipment. This application, through intelligent learning and multi-sensor unit fusion, can also calculate and obtain corresponding adjustment parameter information when the webbing weaving equipment has a weaving abnormality, thereby reducing the quality control difficulty of the weaving control system and achieving the goal of precise weaving control of composite webbing.

[0065] 2. When the ribbon weaving equipment receives a ribbon weaving request instruction from the user or automatically triggered, it acquires equipment weaving data information that meets the relevant process requirements. To improve the weaving quality and yield rate of the composite ribbon after weaving, it selects parameters based on the equipment weaving data information and the selected yarn attributes. The yarn attribute selection parameters include information such as yarn type, yarn color, and yarn size. Then, it corrects and optimizes the equipment weaving data information to obtain the target weaving process parameters. The target weaving process parameters are detected and verified before the composite ribbon weaving begins, which is conducive to achieving precise weaving control of composite ribbon weaving.

[0066] 3. Composite woven webbing typically exhibits high tensile strength, a crisp fabric surface, and difficulty in controlling woven deformation. When assessing the weaving effect, trace line detection is often used to measure the weaving performance. To improve the efficiency of judging the weaving effect of composite woven webbing, this application utilizes a parameter recognition and control model based on machine intelligence learning. After acquiring the target weaving process parameters, it predicts the entire weaving process using historical weaving processes and data, extracting target trace line changes and target weaving parameter changes exceeding a threshold throughout the process. Then, based on the entire weaving duration of the composite webbing process, several first trace line images are extracted at fixed time intervals. Second trace line images are obtained when significant changes occur in weaving speed and webbing attributes (including the number of webbing components, weaving angle, and yarn type). Finally, image processing and correction are performed on the first and second trace line images, including image feature extraction, image binarization, image denoising, and region normalization, resulting in a webbing recognition unit with higher trace line recognition efficiency. It helps to improve the efficiency and accuracy of trace line recognition. By dividing the webbing into several webbing recognition units, it helps to reduce the difficulty of quality control in the weaving control system and improve the accuracy of evaluating the weaving effect of the weaving equipment. Attached Figure Description

[0067] Figure 1 This is a flowchart of a ribbon weaving control method based on intelligent learning and multi-sensor fusion in one embodiment of this application;

[0068] Figure 2 This is a flowchart of step S2 in a ribbon weaving control method based on intelligent learning and multi-sensor fusion in one embodiment of this application;

[0069] Figure 3 This is a flowchart following step S21 in a ribbon weaving control method based on intelligent learning and multi-sensor fusion in one embodiment of this application;

[0070] Figure 4 This is a flowchart following step S213 in a ribbon weaving control method based on intelligent learning and multi-sensor fusion in one embodiment of this application. Detailed Implementation

[0071] The present application will be further described in detail below with reference to the accompanying drawings.

[0072] In one embodiment, such as Figure 1 As shown, this application discloses a webbing weaving control method based on intelligent learning and multi-sensor fusion, which specifically includes the following steps:

[0073] S1: Construct a weaving data sensing mechanism using multiple sensor units pre-installed on the weaving equipment; the weaving data sensing mechanism is used to collect weaving data of the composite weaving during the weaving process; extract model training data from the historical weaving dataset of the weaving data sensing mechanism; and construct a parameter recognition and control model for the corresponding weaving equipment through a deep learning model and the model training data.

[0074] In this embodiment, multiple sensor units, including tension sensors, vision sensors, and yarn metering sensors, are used to detect the tension of the composite webbing, yarn length, weaving pattern, and size during the weaving process through a weaving data sensing mechanism. This facilitates the control transparency of the composite weaving webbing during production. The historical weaving dataset is a collection of historical data collected using the weaving data sensing mechanism. The model training data consists of high-quality weaving data selected based on the composite target weaving process requirements during historical weaving processes, which helps improve the accuracy of the model calculation results of the parameter identification and control model.

[0075] Specifically, the parameter recognition control model of this application is obtained by deep learning and multiple training based on a deep learning model and model training data selected from historical weaving datasets. The parameter recognition control model uses a CNN network object detection method to perform image feature recognition on the images obtained by the visual sensor in order to identify the precise position of the trace line of the weave in the image. The parameter recognition control model can also identify and calibrate the data for abnormalities in the weaving process of the composite weave.

[0076] S2: Identify the webbing weaving request instruction and obtain the target weaving process parameters based on the webbing weaving request instruction; in the parameter identification and control model, determine the target weaving trace line of the composite weaving fabric based on the target weaving process parameters.

[0077] In this embodiment, when a webbing weaving request instruction is detected, the target weaving process parameters are first obtained, and the webbing weaving equipment is triggered to perform composite webbing weaving based on the target weaving process parameters; wherein the target weaving process parameters are the weaving data and weaving control data of the finished composite webbing that meet the relevant process requirements; the target woven trace line is the predicted trace line of the finished composite webbing that meets the relevant process requirements.

[0078] S3: Generate several webbing identification units during the composite webbing weaving process based on the target woven fabric trace lines and target weaving process parameters.

[0079] In this embodiment, the webbing recognition unit is a webbing weaving effect evaluation unit divided on the composite webbing after weaving. During the composite webbing weaving process, a visual sensor is set at the position of the webbing corresponding to the completed composite webbing weaving on the webbing weaving equipment to obtain the weaving data of the webbing recognition unit; at the same time, multiple composite webbing areas are cut out as webbing recognition units during the entire composite webbing weaving process.

[0080] S4: Using each webbing identification unit as a basic unit, acquire the current weaving sensing data set and determine the current weaving trace line; calculate the distance between the target weaving trace line of the webbing identification unit and the corresponding current weaving trace line to obtain several trace line deviation information.

[0081] In this embodiment, in the parameter recognition control model, the deviation distance between the target woven fabric trace line and the actual current woven fabric trace line of the same webbing recognition unit is identified and calculated to obtain the corresponding trace line deviation information.

[0082] Specifically, a webbing unit includes at least three or more trace lines.

[0083] S5: When a trace deviation is detected that meets the preset trace deviation conditions, a sensor information collection command is triggered.

[0084] Specifically, when a certain trace deviation information is identified that meets the trace deviation condition, it indicates that the current woven fabric trace does not meet the process requirements. At this time, a sensor information collection command is triggered, which enables the weaving data sensing mechanism to acquire various types of weaving data of the current weaving equipment in real time, so as to analyze the abnormal components of the weaving equipment and prompt relevant personnel to take corresponding adjustment measures in a timely manner.

[0085] S6: Based on the sensor information collection instructions, obtain the adjustment parameter information of the webbing weaving equipment in the parameter identification and control model; adjust the parameters of the webbing weaving equipment based on the adjustment parameter information.

[0086] In this embodiment, the parameter identification and control model can also calculate the adjustment parameters required by the webbing weaving equipment based on the real-time weaving data obtained from the sensor information collection instructions. Then, based on the adjustment parameters, the control parameters of the weaving equipment can be adjusted in a timely manner when the weaving trace does not meet the process requirements, thereby improving the weaving control accuracy of the webbing weaving equipment. This application, through intelligent learning and multi-sensor unit fusion, can also calculate and obtain the corresponding adjustment parameters when weaving abnormalities occur in the webbing weaving equipment, thereby reducing the quality control difficulty of the weaving control system and achieving the goal of precise weaving control of composite webbing.

[0087] In one embodiment, such as Figure 2 As shown, in step S2, the webbing weaving request instruction is identified, and the target weaving process parameters are obtained according to the webbing weaving request instruction, including:

[0088] S21: Identify the webbing request command through the webbing weaving equipment, and obtain the equipment weaving numerical information based on the webbing weaving request command; the equipment weaving numerical information includes weaving speed, tension force, fabric heating temperature, fabric pressing force, and webbing leveling time.

[0089] In this embodiment, when the webbing weaving equipment receives a webbing weaving request instruction from a user or automatically triggered instruction, it first acquires equipment weaving data information that meets the relevant process requirements; the weaving speed information is the yarn weaving speed value of each yarn bobbin participating in the weaving on the webbing weaving equipment; the tension force is the tension and pulling force value of the weaving tensioning mechanism of the webbing weaving equipment; the weaving heating temperature is the heating temperature value when the composite webbing is initially woven and then pressed; the weaving pressing force is the pressing force value when the webbing is pressed; and the webbing leveling time is the duration of webbing pressing per unit area.

[0090] S22: Obtain yarn attribute selection parameters. In the parameter recognition and control model, obtain the target weaving process parameters based on the yarn attribute selection parameters and equipment weaving numerical information.

[0091] In this embodiment, in order to improve the weaving quality and increase the yield rate of the conformal webbing after weaving, the selection parameters are based on the equipment weaving data information and the selected yarn attributes, wherein the yarn attribute selection parameters include yarn material type, yarn color, and yarn size.

[0092] S23: Trigger the webbing weaving command based on the target weaving process parameters and equipment weaving data information.

[0093] In this embodiment, the webbing weaving equipment starts to execute the webbing weaving operation based on the webbing weaving instruction; the equipment weaving data information is corrected and optimized according to the equipment weaving data information to obtain the target weaving process parameters. The target weaving process parameters are detected and verified before the webbing weaving begins, which is conducive to achieving the effect of precise weaving control of composite webbing weaving.

[0094] In one embodiment, in step S3, the target weaving process parameters include weaving speed and webbing attribute parameters; the sensor unit includes a visual sensing unit; and several webbing recognition units are generated during the composite webbing weaving process based on the target woven trace and the target weaving process parameters, specifically including:

[0095] S31: Obtain the target trace line change information based on the distance change information between two adjacent trace lines during the preset weaving process.

[0096] In this embodiment, since composite woven webbing usually has characteristics such as high tensile strength, stiff fabric surface, and difficulty in controlling the deformation of the woven fabric, the performance of the woven fabric is often measured by the detection and measurement method of the trace line when judging the weaving effect of the woven webbing.

[0097] Specifically, based on a pre-set trace line distance threshold, when the distance change information between two adjacent trace lines is greater than or equal to the trace line distance threshold, the corresponding target trace line change information is obtained.

[0098] S32: Based on the process parameter change information of the target weaving process parameters during the preset weaving process, the change value of the target weaving working parameters is compared with the preset change threshold. When the change value of the target weaving working parameters is greater than the preset change threshold, the target weaving parameter change information is output. The target weaving parameter change information includes weaving speed change information and webbing attribute change information.

[0099] Specifically, the preset change threshold is predefined based on different types of weaving data; the webbing attributes include the number of webbings involved in weaving, the weaving angle, and the type of yarn involved in weaving; the change threshold is preset based on the weaving speed information and the webbing attribute information, and when the weaving speed information is greater than the corresponding change threshold, the weaving speed change information and the webbing attribute change information are output.

[0100] S33: Based on the target woven trace, within the weaving time of the composite webbing weaving process, the visual sensing unit is triggered to acquire the target trace image based on a fixed time interval to obtain the first trace image.

[0101] In this embodiment, the weaving time is the predicted total time required to weave the composite webbing according to the target weaving process parameters; based on the fixed time interval, the visual sensing unit is triggered to extract the webbing area image at the position where the composite webbing has just finished weaving, as the first trace line image; the target trace line image is the actual webbing area image during the weaving process of the composite webbing.

[0102] or,

[0103] S34: When information about changes in weaving speed or changes in webbing properties is obtained, the visual sensing unit is triggered to acquire the target trace image to obtain the second trace image.

[0104] Specifically, based on the significant changes in weaving speed and webbing properties (webbing properties include the number of webbings involved in weaving, weaving angle, type of yarn involved in weaving, etc.), the area image of the weaving webbing in the image area is used as the second trace line image.

[0105] S35: Correct the first trace line image and the second trace line image according to the trace line of the target woven fabric to form a webbing recognition unit.

[0106] In this embodiment, after image processing and correction such as image feature information extraction, image binarization, image denoising, and region normalization are performed on the obtained first and second trace line images, a webbing recognition unit with higher trace line recognition efficiency is obtained. This is beneficial to improving the trace line recognition efficiency and accuracy. By dividing the webbing into several webbing recognition units, it is beneficial to reduce the quality control difficulty of the weaving control system and improve the accuracy of evaluating the weaving effect of the weaving equipment.

[0107] In one embodiment, such as Figure 3 As shown, the webbing weaving equipment includes a spring pressing mechanism for pressing and straightening the woven webbing. After step S21, the webbing weaving control method based on intelligent learning and multi-sensor fusion further includes:

[0108] S211: Based on the target weaving process parameters and equipment weaving data, obtain the corresponding pressing force threshold range of the spring pressing mechanism.

[0109] In this embodiment, the pressure threshold range is the normal pressure range value that the current webbing weaving equipment calculates based on the target weaving process parameters and meets the relevant process requirements.

[0110] Specifically, after the initial weaving is completed, the composite woven webbing needs to be pressed and shaped to improve its flatness; the spring pressing and shaped mechanism is used to press and shape the composite woven webbing that has been initially woven.

[0111] S212: Real-time acquisition of the pressing force data of the spring pressing mechanism, determination of whether the pressing force data is within the corresponding pressing force threshold range, and obtaining the pressing force judgment result.

[0112] In this embodiment, in order to monitor the pressing effect of the spring pressing mechanism on the composite braided webbing, this application compares the pressing force data of the spring pressing mechanism acquired in real time with the pressing force threshold range that meets the pressing process requirements, and determines whether the pressing force data is within the corresponding pressing force threshold range, so as to obtain the pressing force judgment result.

[0113] S213: When a certain pressure data is not within the corresponding pressure threshold range, an abnormal prompt message is generated and sent to the preset user control terminal.

[0114] Specifically, when a certain pressing force data is not within the corresponding pressing force threshold range, it indicates that the pressing force of the spring pressing mechanism is too large or too small, resulting in poor flatness of the pressed composite webbing. It is necessary to adjust the pressing force of the spring pressing mechanism in a timely manner so that its pressing force data returns to the corresponding pressing force threshold range. At this time, an abnormal prompt message is generated to remind the user that the spring pressing mechanism has an abnormality and that handling measures are required. This facilitates weaving control of each production step of the webbing weaving equipment, which is beneficial for quality control and improves the weaving effect of composite webbing.

[0115] In one embodiment, such as Figure 4 As shown, the spring-pressing mechanism includes a pressing elastic element and an adjusting mechanism for adjusting the pressing force of the pressing elastic element; the pressing elastic element includes a pressing spring; after step S213, the webbing weaving control method based on intelligent learning and multi-sensor fusion further includes:

[0116] S2131: Obtain the current clamping force data of the spring clamping mechanism and calculate the minimum clamping force difference required for the current clamping force data to reach the corresponding clamping force threshold range.

[0117] In this embodiment, the minimum pressing force difference is the difference between the current pressing force data and the target pressing force data that meets the process requirements.

[0118] S2132: Compare the minimum pressure difference with the preset critical difference and the preset limit difference. The preset critical difference is less than the preset limit difference.

[0119] In this embodiment, by comparing the minimum pressure difference with the critical difference and the limit difference, the method of adjusting the spring pressure mechanism is determined so as to improve the applicability of the spring pressure mechanism and effectively ensure the flatness of the composite webbing.

[0120] S2133: When the minimum pressure difference is less than the critical difference, an automatic compression command is sent to the regulating mechanism.

[0121] In this embodiment, when the minimum pressure difference is less than the critical difference, it indicates that the pressure data of the spring pressure mechanism decreases slightly and the elastic modulus of the clamping elastic element decreases less. At this time, the clamping elastic element is automatically adjusted by the adjustment mechanism to facilitate the adjustment of the clamping force of the spring pressure mechanism.

[0122] S2134: When the minimum pressure difference is greater than or equal to the critical difference, send a parameter input text box to the user control terminal, obtain the pressure adjustment parameter entered in the text box, and send a compression command to the adjustment mechanism based on the pressure adjustment parameter.

[0123] In this embodiment, when the minimum pressure difference is greater than or equal to the critical difference but less than the limit difference, it indicates that the pressure data of the spring pressure mechanism has dropped significantly, and the elastic modulus of the clamping elastic element has dropped significantly. By sending a message to the user control terminal, the staff needs to determine whether the clamping elastic element needs to be replaced according to the actual usage requirements. If it does not need to be replaced, the method of inputting adjustment parameters is used to make the pressure data of the clamping elastic element fall back into the corresponding pressure threshold range.

[0124] S2135: When the minimum pressure difference is greater than or equal to the limit difference, a replacement prompt message is sent to the user control terminal.

[0125] Specifically, when the minimum compressive force difference is greater than or equal to the limit difference, it indicates that the elastic modulus of the clamping elastic element has dropped significantly, the service life of the clamping elastic element is short, and it needs to be replaced in time. The user is prompted to replace the clamping elastic element in time by outputting a replacement prompt message.

[0126] In one embodiment, the multiple sensor units further include a yarn measuring unit for measuring the length of the yarns involved in weaving; after step S22, the webbing weaving control method based on intelligent learning and multi-sensor fusion further includes:

[0127] S221: Obtain the total yarn length of multiple spools of the ribbon weaving equipment, the historical yarn consumption length of the corresponding yarn metering unit, and the associated spool identifier for each spool.

[0128] In this embodiment, the remaining available yarn length for each bobbin is calculated based on the total yarn length and the historical yarn consumption length; the bobbin is identified by an information code that distinguishes multiple bobbins.

[0129] S222: In the parameter identification and control model, the length of the yarn to be woven corresponding to the webbing weaving request instruction is calculated based on the equipment weaving numerical information and the target weaving process parameters, and the corresponding yarn length data group to be consumed is obtained.

[0130] In this embodiment, the length of the yarn involved in weaving of each yarn bobbin is calculated based on the bobbin identifier to obtain a data combination of the actual length to be consumed, which includes multiple yarn bobbins, i.e., the data group of yarn lengths to be consumed.

[0131] S223: Determine the yarn consumption judgment result based on the total yarn length of multiple bobbins, the historical yarn consumption length, and the corresponding yarn length to be consumed data set.

[0132] S224: During the composite weaving process, based on the yarn consumption judgment result, the bobbin identification and the corresponding weaving speed, determine the time node for outputting the yarn replacement prompt information, and trigger the yarn replacement prompt command at the time node.

[0133] In this embodiment, the time point for outputting the yarn replacement prompt is earlier than the time point when the yarn in the yarn spool is used up, so as to allow the staff time to replace the yarn.

[0134] Specifically, the yarn usage of multiple bobbins in the ribbon weaving equipment is monitored in real time. Based on the required yarn length and actual weaving speed of the composite ribbon being woven, the timing of yarn replacement prompts is calculated in advance. This timing is earlier than the yarn depletion time of the bobbins, thus allowing workers time to replace the yarn. This achieves automated and intelligent production of the ribbon weaving equipment and improves the control effect of the equipment.

[0135] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0136] In one embodiment, a webbing weaving control system based on intelligent learning and multi-sensor fusion is provided, which corresponds to the webbing weaving control method based on intelligent learning and multi-sensor fusion in the above embodiment.

[0137] The webbing knitting control system, based on intelligent learning and multi-sensor fusion, includes a knitting data sensing mechanism and an intelligent control mechanism. Detailed descriptions of each functional module are as follows:

[0138] The weaving data sensing mechanism is constructed using multiple sensor units pre-installed on the weaving equipment to collect weaving data of the composite weaving during the weaving process.

[0139] The intelligent control mechanism is used to extract model training data from the historical weaving dataset of the weaving data sensing mechanism; and to construct the parameter recognition and control model of the corresponding webbing weaving equipment through the deep learning model and the model training data.

[0140] The intelligent control mechanism also includes:

[0141] The request instruction recognition module is used to recognize webbing weaving request instructions and obtain target weaving process parameters based on the webbing weaving request instructions; in the parameter recognition control model, the target weaving trace line of the composite weaving fabric is determined based on the target weaving process parameters.

[0142] The webbing recognition area division module is used to generate several webbing recognition units during the composite webbing weaving process based on the target woven fabric trace lines and target weaving process parameters.

[0143] The trace line deviation calculation module is used to acquire the current weaving sensing data group with each webbing identification unit as the basic unit, and determine the current weaving trace line; calculate the distance between the target weaving trace line of the webbing identification unit and the corresponding current weaving trace line to obtain several trace line deviation information.

[0144] The deviation information collection module is used to trigger a sensor information collection command when it is detected that a certain trace line deviation information meets the preset trace line deviation conditions.

[0145] The equipment parameter adjustment module is used to obtain the adjustment parameter information of the webbing weaving equipment in the parameter identification and control model according to the sensor information collection instructions; and to adjust the parameters of the webbing weaving equipment based on the adjustment parameter information.

[0146] Optionally, the request instruction recognition module includes:

[0147] The equipment weaving value acquisition submodule is used to identify the weaving request command through the weaving equipment and obtain the equipment weaving value information based on the weaving request command; the equipment weaving value information includes weaving speed, tension force, fabric heating temperature, and weaving leveling time.

[0148] The target weaving parameter acquisition submodule is used to acquire yarn attribute selection parameters. In the parameter recognition and control model, the target weaving process parameters are acquired based on the yarn attribute selection parameters and the equipment weaving numerical information.

[0149] The weaving instruction triggering submodule is used to trigger webbing weaving instructions based on the target weaving process parameters and equipment weaving data information.

[0150] For specific limitations regarding the ribbon weaving control system based on intelligent learning and multi-sensor fusion, please refer to the limitations of the ribbon weaving control method based on intelligent learning and multi-sensor fusion mentioned above, which will not be repeated here. Each module in the above-mentioned ribbon weaving control system based on intelligent learning and multi-sensor fusion can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in the processor of the computer device in hardware form or independent of the processor, or it can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0151] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0152] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for controlling the weaving of a braid based on intelligent learning and multi-sensor fusion, characterized in that, The application relates to a parameter identification control method and device for a weaving device. The application comprises: a weaving data sensing mechanism is constructed by using a plurality of sensor units previously arranged on the weaving device; the weaving data sensing mechanism is used for collecting weaving data of a composite fabric during weaving; model training data of a historical weaving data set of the weaving data sensing mechanism is extracted; a parameter identification control model corresponding to the weaving device is constructed by using a deep learning model and the model training data; a weaving request instruction is identified, and target weaving process parameters are obtained according to the weaving request instruction; in the parameter identification control model, target fabric trace lines of the composite fabric are determined based on the target weaving process parameters; a plurality of weaving identification units are generated in the composite fabric weaving process according to the target fabric trace lines and the target weaving process parameters; a current weaving sensing data set is obtained by taking each weaving identification unit as a basic unit, and a current fabric trace line is determined; a distance between the target fabric trace line of the weaving identification unit and the corresponding current fabric trace line is calculated, and a plurality of trace line deviation information are obtained; when it is identified that a certain trace line deviation information meets a preset trace line deviation condition, a sensing information collection instruction is triggered; according to the sensing information collection instruction, adjustment parameter information of the weaving device is obtained in the parameter identification control model; 2. The method of claim 1, wherein the method is based on intelligent learning and multi-sensor fusion for the control of the braiding of the braid. the weaving device is adjusted based on the adjustment parameter information. The application further relates to a parameter identification control method and device for a weaving device. The application comprises: device weaving numerical information is obtained based on the weaving request instruction by identifying the weaving request instruction through the weaving device; the device weaving numerical information comprises weaving speed, tension degree, fabric heating temperature and fabric leveling time; 3. The method of claim 1, wherein the method is based on intelligent learning and multi-sensor fusion for the control of the braiding of the braid. target weaving process parameters are obtained according to the yarn attribute selection parameters and the device weaving numerical information in the parameter identification control model; a weaving instruction is triggered according to the target weaving process parameters and the device weaving data information. The target weaving process parameters comprise weaving speed and fabric attribute parameters; the sensor unit comprises a visual sensing unit; the target fabric trace lines and the target weaving process parameters are used to generate a plurality of weaving identification units in the composite fabric weaving process, which comprises: target trace line change information is obtained according to distance change information of adjacent two trace lines of the target fabric trace lines in a preset weaving process; target weaving parameter change information is outputted when the change value of the target weaving parameter is greater than a preset change threshold value by associating and comparing the change value of the target weaving parameter with the preset change threshold value according to the target weaving parameter change information in the preset weaving process; the target weaving parameter change information comprises weaving speed change information and fabric attribute change information; a first trace line image is obtained by triggering the visual sensing unit to collect a target trace line image based on a fixed time interval within the weaving length of the composite fabric weaving process according to the target fabric trace lines; or When the weaving speed change information or the belt attribute change information is acquired, a visual sensing unit is triggered to collect a target trace line image to obtain a second trace line image; The first trace line image and the second trace line image are corrected according to the target weaving trace line to form a belt recognition unit.

4. The method of claim 2, wherein the method is based on intelligent learning and multi-sensor fusion for the control of the braiding of the braid. The belt weaving device comprises a spring pressing mechanism for pressing the woven belt; After the belt weaving request instruction is recognized and acquired by the belt weaving device and the device weaving data information is acquired based on the belt weaving request instruction, the method further comprises: According to the target weaving process parameter and the device weaving data information, a pressing force threshold interval corresponding to the spring pressing mechanism is acquired; Real-time pressing force data of the spring pressing mechanism is acquired, and it is judged whether the pressing force data is located in the corresponding pressing force threshold interval to obtain a pressing force judgment result; When a certain pressing force data is not in the corresponding pressing force threshold interval, an abnormal prompt information is generated and sent to a preset user control terminal.

5. The method of claim 4, wherein the method is based on intelligent learning and multi-sensor fusion for the control of the braiding of the braid. The spring pressing mechanism comprises a pressing elastic member and an adjusting mechanism for adjusting the pressing force of the pressing elastic member; after the abnormal prompt information is generated and sent to the preset user control terminal when a certain pressing force data is not in the corresponding pressing force threshold interval, the method further comprises: The current pressing force data of the spring pressing mechanism is acquired, and the minimum pressing force difference required for the current pressing force data to reach the corresponding pressing force threshold interval is calculated; The minimum pressing force difference is compared with a preset critical difference value and a preset limit difference value, and the preset critical difference value is smaller than the preset limit difference value; When the minimum pressing force difference is smaller than the critical difference value, an automatic compression instruction is sent to the adjusting mechanism; When the minimum pressing force difference is greater than or equal to the critical difference value, a text box for parameter input is sent to the user control terminal, and a pressing force adjustment parameter input in the text box is acquired, and a compression instruction is sent to the adjusting mechanism based on the pressing force adjustment parameter; When the minimum pressing force difference is greater than or equal to the limit difference value, a replacement prompt information is sent to the user control terminal.

6. The method of claim 2, wherein the method is based on intelligent learning and multi-sensor fusion for the control of the braiding of the braid. The plurality of sensor units further comprise a yarn metering unit for metering the length of yarn participating in weaving; after the target weaving process parameter is acquired in the parameter recognition control model based on the yarn attribute selection parameter and the device weaving numerical value information, the method further comprises: The total length of yarn in each bobbin, the historical yarn consumption length of the corresponding yarn metering unit, and the associated bobbin identifier of the plurality of bobbins of the belt weaving device are acquired; In the parameter recognition control model, the corresponding participating weaving yarn length of the belt weaving request instruction is calculated based on the device weaving numerical value information and the target weaving process parameter, and a corresponding to-be-consumed yarn length data group is obtained; Based on the total length of yarn in each bobbin, the historical yarn consumption length, and the corresponding to-be-consumed yarn length data group of the plurality of bobbins, a yarn consumption judgment result is determined; In the composite braid weaving process, based on the yarn consumption judgment result, the bobbin identifier and the corresponding weaving speed, the time node of outputting the yarn replacement prompt information is determined, and the yarn replacement prompt instruction is triggered at the time node.

7. A system for controlling the weaving of a fabric based on intelligent learning and multi-sensor fusion, characterized in that the system Comprise: Weaving data sensing mechanism, the weaving data sensing mechanism is used for collecting the weaving data of the composite braid in the weaving process by using a plurality of sensor units pre-provided in the braid weaving equipment; The intelligent control mechanism is used for extracting the model training data of the historical weaving data set of the weaving data sensing mechanism; the parameter identification control model corresponding to the braid weaving equipment is constructed by a deep learning model and the model training data; The intelligent control mechanism further comprises: Request instruction identification module, for identifying a braid weaving request instruction, obtaining target weaving process parameters according to the braid weaving request instruction; in the parameter identification control model, the target weaving trace of the composite braid is determined based on the target weaving process parameters; The braid identification region division module is used for generating a plurality of braid identification units in the composite braid weaving process according to the target weaving trace and the target weaving process parameters; Trace deviation calculation module, for taking each braid identification unit as a basic unit, obtaining the current weaving sensing data set, and determining the current weaving trace; the distance between the target weaving trace of the braid identification unit and the corresponding current weaving trace is calculated to obtain a plurality of trace deviation information; The deviation information collection module is used for triggering a sensing information collection instruction when a certain trace deviation information meets a pre-set trace deviation condition; The device parameter adjustment module is used for obtaining the adjustment parameter information of the braid weaving equipment in the parameter identification control model according to the sensing information collection instruction; the parameter adjustment of the braid weaving equipment is carried out based on the adjustment parameter information.

8. The smart learning and multi-sensor fusion based braid control system according to claim 7, wherein, The request instruction identification module comprises: Device weaving value acquisition submodule, for identifying a braid weaving request instruction through the braid weaving equipment, obtaining device weaving value information based on the braid weaving request instruction; the device weaving value information includes weaving speed, tension, weaving heating temperature and braid flattening time; Target weaving parameter acquisition submodule, for obtaining yarn attribute selection parameters, obtaining target weaving process parameters in the parameter identification control model according to the yarn attribute selection parameters and device weaving value information; Weaving instruction triggering submodule, for triggering a braid weaving instruction according to the target weaving process parameters and the device weaving data information.

Images