Knotting machine intelligent operation management method and system based on automatic monitoring

Abstract

The invention discloses an intelligent operation management method and system for a knotting machine based on automatic monitoring, and relates to the technical field of monitoring management.The method includes the steps that automatic monitoring equipment is installed in a set monitoring area, operation data of the knotting machine are collected in real time through the automatic monitoring equipment, and after collection is completed, the operation data of the knotting machine are obtained; knotting machine operation data collected in real time are processed in a data processing mode, after processing is completed, the processed knotting machine operation data are analyzed through a data analysis method, knotting machine performance indexes are determined, and meanwhile a knotting machine state prediction model is constructed based on the knotting machine performance indexes and the processed knotting machine operation data. And finally, based on the knotting machine state prediction model and the knotting machine operation data collected in real time, optimization management is carried out through a digital twinning mode, and the accuracy of intelligent operation management of the knotting machine is improved.

Description

Technical Field

[0001] This invention relates to the field of monitoring and management technology, specifically to an intelligent operation management method and system for knotting machines based on automated monitoring. Background Technology

[0002] In actual operation of the knotting machine, the management of the knotting machine mainly relies on the limited work experience of mechanical engineers to make judgments until the production requirements are met. However, because the judgments are mainly based on the limited work experience of mechanical engineers, the operation monitoring of the knotting machine lacks standardized indicators, resulting in high economic costs and low efficiency in the operation monitoring and management of the knotting machine.

[0003] Existing technology, such as the invention patent application with publication number CN113837479B, discloses an early warning method and system for monitoring the operating status of target equipment. The method includes: using various monitoring indicators of the target equipment at the current moment as input to calculate influencing factors; dynamically optimizing the weights of each neuron in the hidden layer of a BP neural network using these influencing factors to obtain a dynamically optimized BP neural network model for monitoring the operating status of the equipment; and based on this model, providing an early warning of the operating status of the target equipment at the next moment according to the input dataset. The technical solution of this application enables proactive prediction of equipment failures, timely prevention of equipment failures, and guidance for on-site equipment operation and maintenance, providing a scientific basis for the inspection of the operating status of target equipment. Furthermore, the calculation process dynamically reduces network parameters, improving network generalization and achieving accurate early prediction of the operating trends of large target equipment in practical applications.

[0004] As can be seen from the above solutions, most current equipment monitoring methods calculate various monitoring indicators, determine influencing factors, and predict and warn about the operating status based on the determined influencing factors. However, since they only calculate the indicator parameters and lack judgment on the equipment itself, they have certain limitations. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent operation management method and system for knotting machines based on automated monitoring, which solves the problems existing in the background art.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent operation management method for knotting machines based on automated monitoring, specifically including the following steps: S1. Set up a monitoring area for the knotting machine and install automated monitoring equipment within the set monitoring area to collect real-time operating data of the knotting machine through the automated monitoring equipment; S2. The real-time collected knotting machine operation data is processed through data processing methods to obtain the processed knotting machine operation data; S3. Analyze the processed knotting machine operating data using data analysis methods to determine the knotting machine's performance indicators; S4. Construct a knotting machine status prediction model based on the knotting machine performance indicators and the processed knotting machine operation data; S5. Based on the knotting machine status prediction model and real-time collected knotting machine operation data, optimized management is carried out through digital twin method.

[0007] Preferably, the step of setting a monitoring area for the knotting machine and installing automated monitoring equipment within that area to collect real-time operating data of the knotting machine through the automated monitoring equipment includes the following steps: Set collection period set , For the first One collection cycle; The operating data for setting the knotting machine includes: knotting order data, knotting duration data, knotting speed data, and knotting image data.

[0008] Preferably, the step of processing the real-time collected knotting machine operating data to obtain processed knotting machine operating data includes the following steps: S21. Process the knotting order data in the knotting machine operation data through data processing methods to obtain the processed knotting order data; Filtering knotted order data using the Bloom filter algorithm: Create an array of length m, and select... Each hash function iterates through each piece of data in the knotted order data and stores the results in an array; During the traversal using the hash function, if two data entries have the same traversal result, each bit in the two data entries is compared. When the comparison results are consistent, the two data points are set to be the same, and the knotted order data that arrives after the knotted order data is deleted based on the arrival time of the knotted order data. After the traversal is complete, the knotted order data stored in the array during the traversal is summarized to obtain the processed knotted order data; S22. The knotting image data in the knotting machine operation data is processed by image processing to obtain the processed knotting image data. S23. Process the knotting time data and knotting rate data in the knotting machine operation data by unifying the dimensions to obtain the processed knotting time data and knotting rate data. S24. Summarize the processed knotting image data, knotting order data, knotting duration data, and knotting rate data to obtain the processed knotting machine operation data.

[0009] Preferably, the step of processing the knotting image data in the knotting machine operation data using image processing to obtain the processed knotting image data includes the following steps: Binarize the knotted image data to obtain the knotted image outline; Select initial grayscale threshold All pixels in the image data are divided into two categories. and ; set up Less than or equal to the grayscale threshold Pixel category, For values ​​greater than the grayscale threshold Pixel category; Set pixel category The grayscale mean is Pixel category The grayscale mean is The global grayscale mean is ; Define the pixel category in the image data. The probability is It belongs to the pixel category. The probability is ; The binarization formula is shown below: ; ; in, Indicates the binarization threshold; Set gray values ​​greater than the binarization threshold to 255, and gray values ​​less than or equal to the binarization threshold to 0; Summarize the binarized image data to obtain the binarized image data; Image segmentation is performed on the binarized image data to obtain the outline of the knotted image; Standard knot images are acquired in real time and binarized to obtain the outline of the standard knot image. Compare the standard knotted outline with the knotted image outline and calculate the degree of overlap between the two sets of outlines. Set an overlap threshold. When the overlap between two sets of contours is greater than the set threshold, the current knotted image data is considered acceptable; otherwise, it is considered unacceptable. By summarizing the qualified knotted image data, we obtain the processed knotted image data.

[0010] Preferably, the step of analyzing the processed knotting machine operating data using data analysis methods to determine the knotting machine performance indicators includes the following steps: S31. Analyze the production performance indicators of the knotting machine based on the processed knotting machine operation data. The knot-tying tasks can be categorized into: standalone tasks and partially reused tasks; Among them, independent tasks indicate that the current knot type is unrelated to other knot types, while partially reused tasks indicate that the current knot type is partially related to other knot types. Based on the knotting duration and knotting rate data from the processed knotting machine operation data, the time cost required to complete all independent tasks and some reused tasks is calculated as follows: The formulas for calculating the processing time of a single independent task and a single partially reused task are as follows: ; in, This indicates the processing time of a single, independent task. This indicates the processing time for a single reused task, where t represents time. This indicates a function related to processing time. This indicates the proportion of reusable computational resources to the total computational resources. Indicates lookup latency. Knotting time data; Based on the processing time of a single independent task and a single partially reused task, predict and calculate the time cost required to complete the processing of all independent tasks and partially reused tasks. ; in, The time cost required to complete all independent tasks and partially reused tasks. Indicates the number of independent tasks. Total number of tasks; The time cost required to complete all independent tasks and some reused tasks obtained from the prediction calculation is set as the production performance index for the knotting machine. S32. Analyze the production safety indicators of the knotting machine based on the processed knotting machine operation data.

[0011] Preferably, the analysis of production safety indicators of the knotting machine based on the processed knotting machine operating data includes the following steps: The production safety indicators of the knotting machine are analyzed based on the knotting image data in the processed knotting machine operation data. Record the knotting image data of the knotting machine in real time during the production process, and filter out the unqualified knotting image data; The yield rate of the current knotting machine is calculated based on the screening results; The formula for calculating the yield rate is as follows: ; in, Indicates the yield rate. Indicates the number of knots. Indicates the number of defective knots; Set a yield threshold. When the yield within a corresponding time period is less than the set threshold, it indicates that there is a safety hazard in the current knotting machine.

[0012] Preferably, the step of constructing a knotting machine status prediction model based on the knotting machine performance indicators and processed knotting machine operating data includes the following steps: S41. Initialize the processed knotting machine operating data; Set the knotting machine operating status set as ,in, This represents the first set of processed knotting machine operating data input. This represents the nth set of processed knotting machine operating data, and the knotting machine state prediction set is... ,in This represents the first set of knotting machine state prediction data output. This represents the output of the s-th group of knotter state prediction data; S42. Collect and score the knotting machine's operating data, and determine the weight of the knotting machine's operating data; S43. Determine the performance weights of the knotting machine based on its performance indicators; The formula for determining the performance weights of the knotting machine is as follows: ; in, Indicates the performance weight of the knotting machine; S44. Construct a knotting machine state prediction model based on the knotting machine performance weight, knotting machine operation data weight, and initialized knotting machine operation data. The state prediction model for the knotting machine is shown below: ; in, This represents the state prediction model for the knotting machine. This indicates the performance weight of the knotter corresponding to the first set of processed knotter operating data. This indicates the weight of the knotter operation data corresponding to the first set of processed knotter operation data. This represents a constant parameter.

[0013] Preferably, the process of collecting and scoring the knotting machine's operating data and determining the weights of the knotting machine's operating data includes the following steps: Different experts scored the knotted image data in the processed knotting machine operation data based on their own experience, and the scoring results of different experts were converted into fuzzy numbers. At the same time, the weighted average method was used to aggregate the fuzzy numbers. The weighted average method is shown below:

[0014] in, This represents the fuzzy score given by experts to the z-th group of knotted image data. This indicates that expert u has a view on the knotted image data of the z-th group. Let u represent the weight of expert u, and k represent the total number of experts. The fuzzy numbers of each expert's score are used as a set of scoring system data, and all expert scoring system data are aggregated to determine the weight of the knotting machine's operating data.

[0015] Preferably, the optimization management based on the knotting machine status prediction model and real-time collected knotting machine operation data, using a digital twin approach, includes the following steps: Based on the real-time collected operating data of the knotting machine and the knotting machine state prediction model, a knotting machine operating state transition matrix is ​​constructed using a hidden Markov model. Construct the Hidden Markov Model ,in This is a collection space for the operating status of the knotting machine. , for The first knotting machine operating status in the system. for The total number of operating states of the knotting machine; For the state prediction model of the knotting machine, , for The state prediction results of the first group of knotting machines in China for The total number of observed vectors; The state transition matrix for the knotting machine is shown below; The state transition matrix for the knotting machine is as follows:

[0016] in: From State State change, for The Middle The operating status of the knotting machine ; The knotting machine's operating state transition matrix is ​​used to predict each individual state during the knotting machine's operation, and the predicted set of knotting machine operating states and the probability corresponding to each state are output. Based on the predicted set of operating states of the knotting machine, and the probability of each state, corresponding measures are arranged.

[0017] The present invention also provides an intelligent operation management system for a knotting machine based on automated monitoring, which is used to implement an intelligent operation management method for a knotting machine based on automated monitoring. The system includes: a data acquisition module, a data processing module, a data analysis module, a status prediction module, and an optimization management module. The data acquisition module is used to collect real-time operating data of the knotting machine; The data processing module is used to process the real-time collected operating data of the knotting machine; The data analysis module is used to analyze the processed knotting machine operating data and determine the knotting machine performance indicators; The state prediction module is used to construct a state prediction model for the knotting machine based on the performance indicators of the knotting machine and the processed operating data of the knotting machine. The optimization management module is used to perform optimization management based on the knotting machine status prediction model and real-time collected knotting machine operation data through a digital twin approach.

[0018] The beneficial effects of this invention are as follows: (1) This invention installs automated monitoring equipment in a set monitoring area, collects knotting machine operation data in real time through the automated monitoring equipment, processes the real-time knotting machine operation data through data processing, analyzes the processed knotting machine operation data through data analysis methods, determines the knotting machine performance indicators, constructs a knotting machine status prediction model based on the knotting machine performance indicators and the processed knotting machine operation data, and finally optimizes management through digital twin based on the knotting machine status prediction model and the real-time knotting machine operation data, thereby improving the accuracy of intelligent operation management of the knotting machine.

[0019] (2) The present invention processes the knotting order data in the knotting machine operation data through data processing and processes the knotting image data in the knotting machine operation data through image processing, thereby ensuring the reliability of various data processing of the knotting machine operation data.

[0020] (3) This invention analyzes the production performance indicators of the knotting machine based on the processed knotting machine operation data, and analyzes the production safety indicators of the knotting machine based on the processed knotting machine operation data. By analyzing the indicators from two aspects, the rationality of the knotting machine operation data analysis is improved.

[0021] (4) This invention initializes the processed knotting machine operation data, collects knotting machine operation data scores, determines the weight of knotting machine operation data and the weight of knotting machine performance, constructs a knotting machine status prediction model based on the determined weights, and finally optimizes management through digital twin based on the constructed knotting machine status prediction model, thereby improving the intelligence of monitoring and management. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the intelligent operation management method for knotting machines based on automated monitoring according to the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] In a specific embodiment of the present invention, Reference Figure 1 As shown, this invention provides an intelligent operation management method for a knotting machine based on automated monitoring, comprising: S1. Set up a monitoring area for the knotting machine and install automated monitoring equipment within the set monitoring area to collect real-time operating data of the knotting machine through the automated monitoring equipment; S2. The real-time collected knotting machine operation data is processed through data processing methods to obtain the processed knotting machine operation data; S3. Analyze the processed knotting machine operating data using data analysis methods to determine the knotting machine's performance indicators; S4. Construct a knotting machine status prediction model based on the knotting machine performance indicators and the processed knotting machine operation data; S5. Based on the knotting machine status prediction model and real-time collected knotting machine operation data, optimize management through digital twin method; Furthermore, referring to Figure 1As shown, a monitoring area for the knotting machine is set up, and automated monitoring equipment is installed within the set monitoring area. The real-time collection of knotting machine operation data through the automated monitoring equipment includes the following steps: Set the data collection cycle for the installed automated monitoring equipment, and collect real-time data on the operation of the knotting machine within the monitoring area based on the set data collection cycle; Set collection period set , For the first One collection cycle; The operating data for setting the knotting machine includes: knotting order data, knotting duration data, knotting speed data, and knotting image data; Furthermore, referring to Figure 1 As shown, the real-time collected knotting machine operating data is processed using data processing methods to obtain the processed knotting machine operating data, including the following steps: S21. Process the knotting order data in the knotting machine operation data through data processing methods to obtain the processed knotting order data; Filtering knotted order data using the Bloom filter algorithm: Create an array of length m, and select... Each hash function iterates through each piece of data in the knotted order data and stores the results in an array; During the traversal using the hash function, if two data entries have the same traversal result, each bit in the two data entries is compared. When the comparison results are consistent, the two data points are set to be the same, and the knotted order data that arrives after the knotted order data is deleted based on the arrival time of the knotted order data. After the traversal is complete, the knotted order data stored in the array during the traversal is summarized to obtain the processed knotted order data; S22. The knotting image data in the knotting machine operation data is processed by image processing to obtain the processed knotting image data. Binarize the knotted image data to obtain the knotted image outline; Select initial grayscale threshold All pixels in the image data are divided into two categories. and ; set up Less than or equal to the grayscale threshold Pixel category, For values ​​greater than the grayscale threshold Pixel category; Set pixel category The grayscale mean is Pixel category The grayscale mean is The global grayscale mean is ; Define the pixel category in the image data. The probability is It belongs to the pixel category. The probability is ; The binarization formula is shown below: ; ; in, Indicates the binarization threshold; Set gray values ​​greater than the binarization threshold to 255, and gray values ​​less than or equal to the binarization threshold to 0; Summarize the binarized image data to obtain the binarized image data; Image segmentation is performed on the binarized image data to obtain the outline of the knotted image; Furthermore, standard knotted images are acquired in real time and binarized to obtain the outline of the standard knotted image; Furthermore, the standard knotting contour is compared with the knotting image contour, and the degree of overlap between the two sets of contours is calculated. Set an overlap threshold. When the overlap between two sets of contours is greater than the set threshold, the current knotted image data is considered acceptable; otherwise, it is considered unacceptable. Furthermore, the qualified knotted image data are aggregated to obtain the processed knotted image data; S23. Process the knotting time data and knotting rate data in the knotting machine operation data by unifying the dimensions to obtain the processed knotting time data and knotting rate data. S24. Summarize the processed knotting image data, knotting order data, knotting duration data, and knotting speed data to obtain the processed knotting machine operation data. Furthermore, referring to Figure 1 As shown, the following steps are taken to analyze the processed operating data of the knotting machine and determine its performance indicators: S31. Analyze the production performance indicators of the knotting machine based on the processed knotting machine operation data. The knot-tying tasks can be categorized into: standalone tasks and partially reused tasks; Among them, independent tasks indicate that the current knot type is unrelated to other knot types, while partially reused tasks indicate that the current knot type is partially related to other knot types. Based on the knotting duration and knotting rate data from the processed knotting machine operation data, the time cost required to complete all independent tasks and some reused tasks is calculated as follows: The formulas for calculating the processing time of a single independent task and a single partially reused task are as follows: ; in, This indicates the processing time of a single, independent task. This indicates the processing time for a single reused task, where t represents time. This indicates a function related to processing time. This indicates the proportion of reusable computational resources to the total computational resources. Indicates lookup latency. Knotting time data; Furthermore, based on the processing time of a single independent task and a single partially reused task, the time cost required to complete the processing of all independent tasks and partially reused tasks is predicted and calculated. ; in, The time cost required to complete all independent tasks and partially reused tasks. Indicates the number of independent tasks. Total number of tasks; The time cost required to complete all independent tasks and some reused tasks obtained from the prediction calculation is set as the production performance index for the knotting machine. S32. Analyze the production safety indicators of the knotting machine based on the processed knotting machine operation data; The production safety indicators of the knotting machine are analyzed based on the knotting image data in the processed knotting machine operation data. Record the knotting image data of the knotting machine in real time during the production process, and filter out the unqualified knotting image data; Furthermore, the yield rate of the current knotting machine is calculated based on the screening results; The formula for calculating the yield rate is as follows: ; in, Indicates the yield rate. Indicates the number of knots. Indicates the number of defective knots; Set a yield threshold. When the yield within a corresponding time period is less than the set threshold, it indicates that there is a safety hazard in the current knotting machine. Furthermore, referring to Figure 1 As shown, the steps for constructing a knotting machine status prediction model based on the knotting machine's performance indicators and processed knotting machine operating data are as follows: S41. Initialize the processed knotting machine operating data; Set the knotting machine operating status set as ,in, This represents the first set of processed knotting machine operating data input. This represents the nth set of processed knotting machine operating data, and the knotting machine state prediction set is... ,in This represents the first set of knotting machine state prediction data output. This represents the output of the s-th group of knotter state prediction data; S42. Collect and score the knotting machine's operating data, and determine the weight of the knotting machine's operating data; Different experts scored the knotted image data in the processed knotting machine operation data based on their own experience, and the scoring results of different experts were converted into fuzzy numbers. At the same time, the weighted average method was used to aggregate the fuzzy numbers. The weighted average method is shown below:

[0026] in, This represents the fuzzy score given by experts to the z-th group of knotted image data. This indicates that expert u has a view on the knotted image data of the z-th group. Let u represent the weight of expert u, and k represent the total number of experts. Furthermore, the fuzzy numbers of each expert's score are used as a set of scoring system data, and all expert scoring system data are aggregated to determine the weight of the knotting machine's operating data; S43. Determine the performance weights of the knotting machine based on its performance indicators; The formula for determining the performance weights of the knotting machine is as follows: ; in, Indicates the performance weight of the knotting machine; S44. Construct a knotting machine state prediction model based on the knotting machine performance weight, knotting machine operation data weight, and initialized knotting machine operation data. The state prediction model for the knotting machine is shown below: ; in, This represents the state prediction model for the knotting machine. This indicates the performance weight of the knotter corresponding to the first set of processed knotter operating data. This indicates the weight of the knotter operation data corresponding to the first set of processed knotter operation data. Indicates a constant parameter; Furthermore, referring to Figure 1 As shown, the optimization management based on the knotting machine status prediction model and real-time collected knotting machine operation data, using a digital twin approach, includes the following steps: Based on the real-time collected operating data of the knotting machine and the knotting machine state prediction model, a knotting machine operating state transition matrix is ​​constructed using a hidden Markov model. Construct the Hidden Markov Model ,in This is a collection space for the operating status of the knotting machine. , for The first knotting machine operating status in the system. for The total number of operating states of the knotting machine; For the state prediction model of the knotting machine, , for The state prediction results of the first group of knotting machines in China for The total number of observed vectors; The state transition matrix for the knotting machine is shown below; The state transition matrix for the knotting machine is as follows:

[0027] in: From State State change, for The Middle The operating status of the knotting machine ; Furthermore, the knotting machine's operating state transition matrix is ​​used to predict each individual state during the knotting machine's operation, and the predicted set of knotting machine operating states and the probability corresponding to each state are output. Furthermore, based on the predicted set of operating states of the knotting machine and the probability of each state, corresponding measures are arranged. In one specific embodiment, the intelligent operation management system for a knotting machine based on automated monitoring is used to implement an intelligent operation management method for a knotting machine based on automated monitoring. The system includes: a data acquisition module, a data processing module, a data analysis module, a status prediction module, and an optimization management module. The data acquisition module is used to collect real-time operating data of the knotting machine; The data processing module is used to process the real-time collected operating data of the knotting machine; The data analysis module is used to analyze the processed knotting machine operating data and determine the knotting machine performance indicators; The state prediction module is used to construct a state prediction model for the knotting machine based on the performance indicators of the knotting machine and the processed operating data of the knotting machine. The optimization management module is used to perform optimization management based on the knotting machine status prediction model and real-time collected knotting machine operation data through a digital twin approach.

[0028] It should be noted that, The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. An intelligent operation management method for a knotter based on automated monitoring, characterized in that, The method comprises the following steps: S1, setting a monitoring area of the knotting machine, installing an automatic monitoring device in the set monitoring area, and collecting knotting machine operation data in real time through the automatic monitoring device; S2, processing the real-time collected knotting machine operation data through a data processing method to obtain processed knotting machine operation data; S3, analyzing the processed knotting machine operation data through a data analysis method to determine the performance index of the knotting machine; S4, constructing a knotting machine state prediction model based on the performance index of the knotting machine and the processed knotting machine operation data; S5, based on the knotting machine state prediction model and the real-time collected knotting machine operation data, optimizing management through digital twinning.

2. The intelligent operation management method of the knotter based on automatic monitoring according to claim 1, characterized in that, The step of setting a monitoring area of the knotting machine and installing an automatic monitoring device in the set monitoring area to collect knotting machine operation data in real time comprises the following steps: Setting a set of acquisition periods , is a first acquisition period ; The knotting machine operation data includes knotting order data, knotting duration data, knotting rate data, and knotting image data.

3. The intelligent operation management method of the knotter based on automatic monitoring according to claim 1, characterized in that, The step of processing the real-time collected knotting machine operation data through a data processing method to obtain processed knotting machine operation data comprises the following steps: S21, processing the knotting order data in the knotting machine operation data through a data processing method to obtain processed knotting order data; Filtering the knotting order data based on a Bloom filter algorithm: An array of length m is established, and a hash function is selected Each data in the tied order data is traversed, and the traversal result is saved in the array. During the traversal process through the hash function, when the traversal results of two data are the same, compare each bit in the two data; When the comparison result is consistent, set the current two data as the same data, and delete the knotting order data arriving later based on the arrival time of the knotting order data; After the traversal is completed, the knotting order data saved in the array during the traversal process is summarized to obtain the processed knotting order data; S22, processing the knotting image data in the knotting machine operation data through an image processing method to obtain processed knotting image data; S23, processing the knotting duration data and knotting rate data in the knotting machine operation data through a dimension unification method to obtain processed knotting duration data and knotting rate data; S24, summarizing the processed knotting image data, knotting order data, knotting duration data, and knotting rate data to obtain processed knotting machine operation data.

4. The intelligent operation management method of the knotter based on automatic monitoring according to claim 3, characterized in that, The step of processing the knotting image data in the knotting machine operation data through an image processing method to obtain processed knotting image data comprises the following steps: Binaryzation processing the knotting image data to obtain a knotting image contour; Selecting an initial grayscale threshold Classifying all pixels in image data into two classes and ; set up Less than or equal to the grayscale threshold Pixel category, For values ​​greater than the grayscale threshold Pixel category; Set the pixel class The mean gray value of the pixel class is The mean gray value of the pixel class is The global mean gray value is Setting a probability that a pixel in image data belongs to a pixel class is , a probability that a pixel belongs to a pixel class is ; The binaryzation processing formula is as follows: ； ； wherein denotes a binarization threshold value; Setting the gray value greater than the binaryzation threshold value as 255, and setting the gray value less than or equal to the binaryzation threshold value as 0; Summarizing the image data after binaryzation processing to obtain binaryzation processed image data; Image segmentation is performed on the binaryzation processed image data to obtain a knotting image contour; Real-time collection of a standard knotting image, and binaryzation processing of the standard knotting image to obtain a standard knotting image contour; Comparing the standard knotting contour with the knotting image contour to calculate the coincidence degree of the two contours; A coincidence degree threshold is set, and when the coincidence degree of the two groups of contours is greater than the set threshold, it is indicated that the current knotting image data is qualified, and otherwise, it is not qualified; The qualified knotting image data is summarized to obtain processed knotting image data.

5. The intelligent operation management method of the knotter based on automatic monitoring according to claim 1, characterized in that, The processed knotting machine running data is analyzed by the data analysis method to determine the knotting machine performance index, including the following steps: S31, production performance index analysis of the knotting machine based on the processed knotting machine running data; The knotting task setting includes an independent task and a partially reusable task; The independent task indicates that the current knotting type is not related to other knotting types, and the partially reusable task indicates that the current knotting type is partially related to other knotting types; The time cost required for processing all independent tasks and partially reusable tasks is calculated based on the knotting duration data and the knotting rate data in the processed knotting machine running data as follows: The processing duration of a single independent task and a single partially reusable task is calculated as follows: ； wherein, represents a single independent task processing duration, represents a single partial reuse task processing duration, t represents time, represents a processing duration related function, represents a proportion of reusable computation to total computation, represents a lookup latency, knotting duration data; Based on the processing duration of a single independent task and a single partially reusable task, the time cost required for processing all independent tasks and partially reusable tasks is predicted and calculated; ； wherein, the time cost required for all independent tasks and partially reused task processing to be completed, denotes the number of independent tasks, is the total number of tasks; The time cost required for processing all independent tasks and partially reusable tasks predicted and calculated is set as the production performance index of the knotting machine; S32, production safety index analysis of the knotting machine based on the processed knotting machine running data.

6. The intelligent operation management method of the knotter based on automatic monitoring according to claim 5, characterized in that, The production safety index analysis of the knotting machine based on the processed knotting machine running data includes the following steps: Production safety index analysis of the knotting machine based on the knotting image data in the processed knotting machine running data; Real-time recording of the knotting image data of the knotting machine in the production process, and screening out unqualified knotting image data therefrom; Based on the screening result, the yield of the current knotting machine is calculated; The yield calculation formula is as follows: ； wherein, represents the yield, represents the number of knots, represents the number of non-conforming knots; A yield threshold is set, and when the yield in the corresponding time period is less than the set threshold, it is indicated that the current knotting machine has safety hazards.

7. The intelligent operation management method of the knotter based on automatic monitoring according to claim 1, characterized in that, The knotting machine state prediction model is constructed based on the knotting machine performance index and the processed knotting machine running data, including the following steps: S41, initialization of the processed knotting machine running data; The set of running states of the knotter is set as wherein, represents the first set of processed knotter running data input, represents the nth set of processed knotter running data input, and the set of knotter state predictions is wherein represents the first set of knotter state prediction data output, represents the s th set of knotter state prediction data output. S42, collection of knotting machine running data scores and determination of knotting machine running data weights; S43, determination of knotting machine performance weights based on knotting machine performance indexes; The knotting machine performance weight determination formula is as follows: ； wherein, represents the knotter performance weight; S44, construction of a knotting machine state prediction model based on knotting machine performance weights, knotting machine running data weights, and initialized knotting machine running data; The knotting machine state prediction model is as follows: ； wherein, represents a knotter state prediction model, represents a first set of processed knotter performance data corresponding to a knotter performance weight, represents a first set of processed knotter performance data corresponding to a knotter performance weight, represents a constant parameter.

8. The intelligent operation management method of the knotter based on automatic monitoring according to claim 7, characterized in that, The collection of knotting machine running data scores and the determination of knotting machine running data weights include the following steps: Different experts judge the knotting image data scores in the processed knotting machine running data according to their own experience, and convert the scoring results of different experts into fuzzy numbers, and at the same time, the fuzzy numbers are aggregated by using the weighted average method; The weighted average method is as follows: ； wherein, represents the fuzzy score of the expert for the z-th set of tied image data, represents the expert u's score for the z-th set of tied image data, represents the weight of the expert u, and k represents the total number of experts; The fuzzy numbers of each expert's scoring results are taken as a group of scoring system data, and all expert scoring system data are summarized to determine the knotting machine running data weights.

9. The intelligent operation management method of the knotter based on automatic monitoring according to claim 1, characterized in that, The optimization management based on the knotting machine state prediction model and the real-time collected knotting machine operation data is performed through a digital twin method, and includes the following steps: Based on the real-time collected knotting machine operation data and the knotting machine state prediction model, a knotting machine operation state transition matrix is constructed through a hidden Markov model; constructing the hidden markov model wherein is a set of knotter operating state spaces, , is the first knotter operating state in , is the total number of knotter operating states in ; is a knotter state prediction model, , is the first group of knotter state prediction results in , is the total number of observation vectors in ; is a knotter operating state transition matrix, as follows: The knotting machine operating state transition matrix is as follows: ； wherein: is from state to state change, is the knotter machine operating state, ; Each individual state in the knotting machine operation process is predicted through the knotting machine operation state transition matrix, and a predicted knotting machine operation state set and a probability corresponding to each state are outputted; Based on the predicted knotting machine operation state set and the probability corresponding to each state, corresponding measures are arranged.

10. A system for implementing the intelligent operation management method of the knotter based on automatic monitoring according to claims 1-9, characterized in that, It includes: a data acquisition module, a data processing module, a data analysis module, a state prediction module, and an optimization management module; The data acquisition module is used for real-time acquisition of knotting machine operation data; The data processing module is used for processing the real-time collected knotting machine operation data; The data analysis module is used for analyzing the processed knotting machine operation data to determine the knotting machine performance index; The state prediction module is used for constructing a knotting machine state prediction model according to the knotting machine performance index and the processed knotting machine operation data; The optimization management module is used for optimization management through a digital twin method according to the knotting machine state prediction model and the real-time collected knotting machine operation data.

Images